1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988--2023 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.
48 * gdbserver: (gdb) Server. The GNU debugging server.
52 @c man begin COPYRIGHT
53 Copyright @copyright{} 1988-2023 Free Software Foundation, Inc.
55 Permission is granted to copy, distribute and/or modify this document
56 under the terms of the GNU Free Documentation License, Version 1.3 or
57 any later version published by the Free Software Foundation; with the
58 Invariant Sections being ``Free Software'' and ``Free Software Needs
59 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
60 and with the Back-Cover Texts as in (a) below.
62 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
63 this GNU Manual. Buying copies from GNU Press supports the FSF in
64 developing GNU and promoting software freedom.''
69 This file documents the @sc{gnu} debugger @value{GDBN}.
71 This is the @value{EDITION} Edition, of @cite{Debugging with
72 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
73 @ifset VERSION_PACKAGE
74 @value{VERSION_PACKAGE}
76 Version @value{GDBVN}.
82 @title Debugging with @value{GDBN}
83 @subtitle The @sc{gnu} Source-Level Debugger
85 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
86 @ifset VERSION_PACKAGE
88 @subtitle @value{VERSION_PACKAGE}
90 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
94 \hfill (Send bugs and comments on @value{GDBN} to @value{BUGURL}.)\par
95 \hfill {\it Debugging with @value{GDBN}}\par
96 \hfill \TeX{}info \texinfoversion\par
100 @vskip 0pt plus 1filll
101 Published by the Free Software Foundation @*
102 51 Franklin Street, Fifth Floor,
103 Boston, MA 02110-1301, USA@*
104 ISBN 978-0-9831592-3-0 @*
113 @top Debugging with @value{GDBN}
115 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
117 This is the @value{EDITION} Edition, for @value{GDBN}
118 @ifset VERSION_PACKAGE
119 @value{VERSION_PACKAGE}
121 Version @value{GDBVN}.
123 Copyright (C) 1988-2023 Free Software Foundation, Inc.
125 This edition of the GDB manual is dedicated to the memory of Fred
126 Fish. Fred was a long-standing contributor to GDB and to Free
127 software in general. We will miss him.
130 * Summary:: Summary of @value{GDBN}
131 * Sample Session:: A sample @value{GDBN} session
133 * Invocation:: Getting in and out of @value{GDBN}
134 * Commands:: @value{GDBN} commands
135 * Running:: Running programs under @value{GDBN}
136 * Stopping:: Stopping and continuing
137 * Reverse Execution:: Running programs backward
138 * Process Record and Replay:: Recording inferior's execution and replaying it
139 * Stack:: Examining the stack
140 * Source:: Examining source files
141 * Data:: Examining data
142 * Optimized Code:: Debugging optimized code
143 * Macros:: Preprocessor Macros
144 * Tracepoints:: Debugging remote targets non-intrusively
145 * Overlays:: Debugging programs that use overlays
147 * Languages:: Using @value{GDBN} with different languages
149 * Symbols:: Examining the symbol table
150 * Altering:: Altering execution
151 * GDB Files:: @value{GDBN} files
152 * Targets:: Specifying a debugging target
153 * Remote Debugging:: Debugging remote programs
154 * Configurations:: Configuration-specific information
155 * Controlling GDB:: Controlling @value{GDBN}
156 * Extending GDB:: Extending @value{GDBN}
157 * Interpreters:: Command Interpreters
158 * TUI:: @value{GDBN} Text User Interface
159 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
160 * GDB/MI:: @value{GDBN}'s Machine Interface.
161 * Annotations:: @value{GDBN}'s annotation interface.
162 * Debugger Adapter Protocol:: The Debugger Adapter Protocol.
163 * JIT Interface:: Using the JIT debugging interface.
164 * In-Process Agent:: In-Process Agent
166 * GDB Bugs:: Reporting bugs in @value{GDBN}
168 @ifset SYSTEM_READLINE
169 * Command Line Editing: (rluserman). Command Line Editing
170 * Using History Interactively: (history). Using History Interactively
172 @ifclear SYSTEM_READLINE
173 * Command Line Editing:: Command Line Editing
174 * Using History Interactively:: Using History Interactively
176 * In Memoriam:: In Memoriam
177 * Formatting Documentation:: How to format and print @value{GDBN} documentation
178 * Installing GDB:: Installing @value{GDBN}
179 * Maintenance Commands:: Maintenance Commands
180 * Remote Protocol:: GDB Remote Serial Protocol
181 * Agent Expressions:: The @value{GDBN} Agent Expression Mechanism
182 * Target Descriptions:: How targets can describe themselves to
184 * Operating System Information:: Getting additional information from
186 * Trace File Format:: @value{GDBN} trace file format
187 * Index Section Format:: .gdb_index section format
188 * Debuginfod:: Download debugging resources with @code{debuginfod}
189 * Man Pages:: Manual pages
190 * Copying:: GNU General Public License says
191 how you can copy and share @value{GDBN}
192 * GNU Free Documentation License:: The license for this documentation
193 * Concept Index:: Index of @value{GDBN} concepts
194 * Command and Variable Index:: Index of @value{GDBN} commands, variables,
195 functions, and Python data types
203 @unnumbered Summary of @value{GDBN}
205 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
206 going on ``inside'' another program while it executes---or what another
207 program was doing at the moment it crashed.
209 @value{GDBN} can do four main kinds of things (plus other things in support of
210 these) to help you catch bugs in the act:
214 Start your program, specifying anything that might affect its behavior.
217 Make your program stop on specified conditions.
220 Examine what has happened, when your program has stopped.
223 Change things in your program, so you can experiment with correcting the
224 effects of one bug and go on to learn about another.
227 You can use @value{GDBN} to debug programs written in C and C@t{++}.
228 For more information, see @ref{Supported Languages,,Supported Languages}.
229 For more information, see @ref{C,,C and C++}.
231 Support for D is partial. For information on D, see
235 Support for Modula-2 is partial. For information on Modula-2, see
236 @ref{Modula-2,,Modula-2}.
238 Support for OpenCL C is partial. For information on OpenCL C, see
239 @ref{OpenCL C,,OpenCL C}.
242 Debugging Pascal programs which use sets, subranges, file variables, or
243 nested functions does not currently work. @value{GDBN} does not support
244 entering expressions, printing values, or similar features using Pascal
248 @value{GDBN} can be used to debug programs written in Fortran, although
249 it may be necessary to refer to some variables with a trailing
252 @value{GDBN} can be used to debug programs written in Objective-C,
253 using either the Apple/NeXT or the GNU Objective-C runtime.
256 * Free Software:: Freely redistributable software
257 * Free Documentation:: Free Software Needs Free Documentation
258 * Contributors:: Contributors to GDB
262 @unnumberedsec Free Software
264 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
265 General Public License
266 (GPL). The GPL gives you the freedom to copy or adapt a licensed
267 program---but every person getting a copy also gets with it the
268 freedom to modify that copy (which means that they must get access to
269 the source code), and the freedom to distribute further copies.
270 Typical software companies use copyrights to limit your freedoms; the
271 Free Software Foundation uses the GPL to preserve these freedoms.
273 Fundamentally, the General Public License is a license which says that
274 you have these freedoms and that you cannot take these freedoms away
277 @node Free Documentation
278 @unnumberedsec Free Software Needs Free Documentation
280 The biggest deficiency in the free software community today is not in
281 the software---it is the lack of good free documentation that we can
282 include with the free software. Many of our most important
283 programs do not come with free reference manuals and free introductory
284 texts. Documentation is an essential part of any software package;
285 when an important free software package does not come with a free
286 manual and a free tutorial, that is a major gap. We have many such
289 Consider Perl, for instance. The tutorial manuals that people
290 normally use are non-free. How did this come about? Because the
291 authors of those manuals published them with restrictive terms---no
292 copying, no modification, source files not available---which exclude
293 them from the free software world.
295 That wasn't the first time this sort of thing happened, and it was far
296 from the last. Many times we have heard a GNU user eagerly describe a
297 manual that he is writing, his intended contribution to the community,
298 only to learn that he had ruined everything by signing a publication
299 contract to make it non-free.
301 Free documentation, like free software, is a matter of freedom, not
302 price. The problem with the non-free manual is not that publishers
303 charge a price for printed copies---that in itself is fine. (The Free
304 Software Foundation sells printed copies of manuals, too.) The
305 problem is the restrictions on the use of the manual. Free manuals
306 are available in source code form, and give you permission to copy and
307 modify. Non-free manuals do not allow this.
309 The criteria of freedom for a free manual are roughly the same as for
310 free software. Redistribution (including the normal kinds of
311 commercial redistribution) must be permitted, so that the manual can
312 accompany every copy of the program, both on-line and on paper.
314 Permission for modification of the technical content is crucial too.
315 When people modify the software, adding or changing features, if they
316 are conscientious they will change the manual too---so they can
317 provide accurate and clear documentation for the modified program. A
318 manual that leaves you no choice but to write a new manual to document
319 a changed version of the program is not really available to our
322 Some kinds of limits on the way modification is handled are
323 acceptable. For example, requirements to preserve the original
324 author's copyright notice, the distribution terms, or the list of
325 authors, are ok. It is also no problem to require modified versions
326 to include notice that they were modified. Even entire sections that
327 may not be deleted or changed are acceptable, as long as they deal
328 with nontechnical topics (like this one). These kinds of restrictions
329 are acceptable because they don't obstruct the community's normal use
332 However, it must be possible to modify all the @emph{technical}
333 content of the manual, and then distribute the result in all the usual
334 media, through all the usual channels. Otherwise, the restrictions
335 obstruct the use of the manual, it is not free, and we need another
336 manual to replace it.
338 Please spread the word about this issue. Our community continues to
339 lose manuals to proprietary publishing. If we spread the word that
340 free software needs free reference manuals and free tutorials, perhaps
341 the next person who wants to contribute by writing documentation will
342 realize, before it is too late, that only free manuals contribute to
343 the free software community.
345 If you are writing documentation, please insist on publishing it under
346 the GNU Free Documentation License or another free documentation
347 license. Remember that this decision requires your approval---you
348 don't have to let the publisher decide. Some commercial publishers
349 will use a free license if you insist, but they will not propose the
350 option; it is up to you to raise the issue and say firmly that this is
351 what you want. If the publisher you are dealing with refuses, please
352 try other publishers. If you're not sure whether a proposed license
353 is free, write to @email{licensing@@gnu.org}.
355 You can encourage commercial publishers to sell more free, copylefted
356 manuals and tutorials by buying them, and particularly by buying
357 copies from the publishers that paid for their writing or for major
358 improvements. Meanwhile, try to avoid buying non-free documentation
359 at all. Check the distribution terms of a manual before you buy it,
360 and insist that whoever seeks your business must respect your freedom.
361 Check the history of the book, and try to reward the publishers that
362 have paid or pay the authors to work on it.
364 The Free Software Foundation maintains a list of free documentation
365 published by other publishers, at
366 @url{http://www.fsf.org/doc/other-free-books.html}.
369 @unnumberedsec Contributors to @value{GDBN}
371 Richard Stallman was the original author of @value{GDBN}, and of many
372 other @sc{gnu} programs. Many others have contributed to its
373 development. This section attempts to credit major contributors. One
374 of the virtues of free software is that everyone is free to contribute
375 to it; with regret, we cannot actually acknowledge everyone here. The
376 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
377 blow-by-blow account.
379 Changes much prior to version 2.0 are lost in the mists of time.
382 @emph{Plea:} Additions to this section are particularly welcome. If you
383 or your friends (or enemies, to be evenhanded) have been unfairly
384 omitted from this list, we would like to add your names!
387 So that they may not regard their many labors as thankless, we
388 particularly thank those who shepherded @value{GDBN} through major
390 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
391 Jim Blandy (release 4.18);
392 Jason Molenda (release 4.17);
393 Stan Shebs (release 4.14);
394 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
395 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
396 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
397 Jim Kingdon (releases 3.5, 3.4, and 3.3);
398 and Randy Smith (releases 3.2, 3.1, and 3.0).
400 Richard Stallman, assisted at various times by Peter TerMaat, Chris
401 Hanson, and Richard Mlynarik, handled releases through 2.8.
403 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
404 in @value{GDBN}, with significant additional contributions from Per
405 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
406 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
407 much general update work leading to release 3.0).
409 @value{GDBN} uses the BFD subroutine library to examine multiple
410 object-file formats; BFD was a joint project of David V.
411 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
413 David Johnson wrote the original COFF support; Pace Willison did
414 the original support for encapsulated COFF.
416 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
418 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
419 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
421 Jean-Daniel Fekete contributed Sun 386i support.
422 Chris Hanson improved the HP9000 support.
423 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
424 David Johnson contributed Encore Umax support.
425 Jyrki Kuoppala contributed Altos 3068 support.
426 Jeff Law contributed HP PA and SOM support.
427 Keith Packard contributed NS32K support.
428 Doug Rabson contributed Acorn Risc Machine support.
429 Bob Rusk contributed Harris Nighthawk CX-UX support.
430 Chris Smith contributed Convex support (and Fortran debugging).
431 Jonathan Stone contributed Pyramid support.
432 Michael Tiemann contributed SPARC support.
433 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
434 Pace Willison contributed Intel 386 support.
435 Jay Vosburgh contributed Symmetry support.
436 Marko Mlinar contributed OpenRISC 1000 support.
438 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
440 Rich Schaefer and Peter Schauer helped with support of SunOS shared
443 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
444 about several machine instruction sets.
446 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
447 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
448 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
449 and RDI targets, respectively.
451 Brian Fox is the author of the readline libraries providing
452 command-line editing and command history.
454 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
455 Modula-2 support, and contributed the Languages chapter of this manual.
457 Fred Fish wrote most of the support for Unix System Vr4.
458 He also enhanced the command-completion support to cover C@t{++} overloaded
461 Hitachi America (now Renesas America), Ltd. sponsored the support for
462 H8/300, H8/500, and Super-H processors.
464 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
466 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
469 Toshiba sponsored the support for the TX39 Mips processor.
471 Matsushita sponsored the support for the MN10200 and MN10300 processors.
473 Fujitsu sponsored the support for SPARClite and FR30 processors.
475 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
478 Michael Snyder added support for tracepoints.
480 Stu Grossman wrote gdbserver.
482 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
483 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
485 The following people at the Hewlett-Packard Company contributed
486 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
487 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
488 compiler, and the Text User Interface (nee Terminal User Interface):
489 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
490 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
491 provided HP-specific information in this manual.
493 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
494 Robert Hoehne made significant contributions to the DJGPP port.
496 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
497 development since 1991. Cygnus engineers who have worked on @value{GDBN}
498 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
499 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
500 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
501 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
502 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
503 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
504 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
505 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
506 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
507 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
508 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
509 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
510 Zuhn have made contributions both large and small.
512 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
513 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
515 Jim Blandy added support for preprocessor macros, while working for Red
518 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
519 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
520 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
521 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
522 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
523 with the migration of old architectures to this new framework.
525 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
526 unwinder framework, this consisting of a fresh new design featuring
527 frame IDs, independent frame sniffers, and the sentinel frame. Mark
528 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
529 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
530 trad unwinders. The architecture-specific changes, each involving a
531 complete rewrite of the architecture's frame code, were carried out by
532 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
533 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
534 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
535 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
538 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
539 Tensilica, Inc.@: contributed support for Xtensa processors. Others
540 who have worked on the Xtensa port of @value{GDBN} in the past include
541 Steve Tjiang, John Newlin, and Scott Foehner.
543 Michael Eager and staff of Xilinx, Inc., contributed support for the
544 Xilinx MicroBlaze architecture.
546 Initial support for the FreeBSD/mips target and native configuration
547 was developed by SRI International and the University of Cambridge
548 Computer Laboratory under DARPA/AFRL contract FA8750-10-C-0237
549 ("CTSRD"), as part of the DARPA CRASH research programme.
551 Initial support for the FreeBSD/riscv target and native configuration
552 was developed by SRI International and the University of Cambridge
553 Computer Laboratory (Department of Computer Science and Technology)
554 under DARPA contract HR0011-18-C-0016 ("ECATS"), as part of the DARPA
555 SSITH research programme.
557 The original port to the OpenRISC 1000 is believed to be due to
558 Alessandro Forin and Per Bothner. More recent ports have been the work
559 of Jeremy Bennett, Franck Jullien, Stefan Wallentowitz and
562 Weimin Pan, David Faust and Jose E. Marchesi contributed support for
563 the Linux kernel BPF virtual architecture. This work was sponsored by
567 @chapter A Sample @value{GDBN} Session
569 You can use this manual at your leisure to read all about @value{GDBN}.
570 However, a handful of commands are enough to get started using the
571 debugger. This chapter illustrates those commands.
574 In this sample session, we emphasize user input like this: @b{input},
575 to make it easier to pick out from the surrounding output.
578 @c FIXME: this example may not be appropriate for some configs, where
579 @c FIXME...primary interest is in remote use.
581 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
582 processor) exhibits the following bug: sometimes, when we change its
583 quote strings from the default, the commands used to capture one macro
584 definition within another stop working. In the following short @code{m4}
585 session, we define a macro @code{foo} which expands to @code{0000}; we
586 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
587 same thing. However, when we change the open quote string to
588 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
589 procedure fails to define a new synonym @code{baz}:
598 @b{define(bar,defn(`foo'))}
602 @b{changequote(<QUOTE>,<UNQUOTE>)}
604 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
607 m4: End of input: 0: fatal error: EOF in string
611 Let us use @value{GDBN} to try to see what is going on.
614 $ @b{@value{GDBP} m4}
615 @c FIXME: this falsifies the exact text played out, to permit smallbook
616 @c FIXME... format to come out better.
617 @value{GDBN} is free software and you are welcome to distribute copies
618 of it under certain conditions; type "show copying" to see
620 There is absolutely no warranty for @value{GDBN}; type "show warranty"
623 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
628 @value{GDBN} reads only enough symbol data to know where to find the
629 rest when needed; as a result, the first prompt comes up very quickly.
630 We now tell @value{GDBN} to use a narrower display width than usual, so
631 that examples fit in this manual.
634 (@value{GDBP}) @b{set width 70}
638 We need to see how the @code{m4} built-in @code{changequote} works.
639 Having looked at the source, we know the relevant subroutine is
640 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
641 @code{break} command.
644 (@value{GDBP}) @b{break m4_changequote}
645 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
649 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
650 control; as long as control does not reach the @code{m4_changequote}
651 subroutine, the program runs as usual:
654 (@value{GDBP}) @b{run}
655 Starting program: /work/Editorial/gdb/gnu/m4/m4
663 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
664 suspends execution of @code{m4}, displaying information about the
665 context where it stops.
668 @b{changequote(<QUOTE>,<UNQUOTE>)}
670 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
672 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
676 Now we use the command @code{n} (@code{next}) to advance execution to
677 the next line of the current function.
681 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
686 @code{set_quotes} looks like a promising subroutine. We can go into it
687 by using the command @code{s} (@code{step}) instead of @code{next}.
688 @code{step} goes to the next line to be executed in @emph{any}
689 subroutine, so it steps into @code{set_quotes}.
693 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
695 530 if (lquote != def_lquote)
699 The display that shows the subroutine where @code{m4} is now
700 suspended (and its arguments) is called a stack frame display. It
701 shows a summary of the stack. We can use the @code{backtrace}
702 command (which can also be spelled @code{bt}), to see where we are
703 in the stack as a whole: the @code{backtrace} command displays a
704 stack frame for each active subroutine.
707 (@value{GDBP}) @b{bt}
708 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
710 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
712 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
713 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
715 #4 0x79dc in expand_input () at macro.c:40
716 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
720 We step through a few more lines to see what happens. The first two
721 times, we can use @samp{s}; the next two times we use @code{n} to avoid
722 falling into the @code{xstrdup} subroutine.
726 0x3b5c 532 if (rquote != def_rquote)
728 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
729 def_lquote : xstrdup(lq);
731 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
734 538 len_lquote = strlen(rquote);
738 The last line displayed looks a little odd; we can examine the variables
739 @code{lquote} and @code{rquote} to see if they are in fact the new left
740 and right quotes we specified. We use the command @code{p}
741 (@code{print}) to see their values.
744 (@value{GDBP}) @b{p lquote}
745 $1 = 0x35d40 "<QUOTE>"
746 (@value{GDBP}) @b{p rquote}
747 $2 = 0x35d50 "<UNQUOTE>"
751 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
752 To look at some context, we can display ten lines of source
753 surrounding the current line with the @code{l} (@code{list}) command.
759 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
761 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
764 538 len_lquote = strlen(rquote);
765 539 len_rquote = strlen(lquote);
772 Let us step past the two lines that set @code{len_lquote} and
773 @code{len_rquote}, and then examine the values of those variables.
777 539 len_rquote = strlen(lquote);
780 (@value{GDBP}) @b{p len_lquote}
782 (@value{GDBP}) @b{p len_rquote}
787 That certainly looks wrong, assuming @code{len_lquote} and
788 @code{len_rquote} are meant to be the lengths of @code{lquote} and
789 @code{rquote} respectively. We can set them to better values using
790 the @code{p} command, since it can print the value of
791 any expression---and that expression can include subroutine calls and
795 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
797 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
802 Is that enough to fix the problem of using the new quotes with the
803 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
804 executing with the @code{c} (@code{continue}) command, and then try the
805 example that caused trouble initially:
811 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
818 Success! The new quotes now work just as well as the default ones. The
819 problem seems to have been just the two typos defining the wrong
820 lengths. We allow @code{m4} exit by giving it an EOF as input:
824 Program exited normally.
828 The message @samp{Program exited normally.} is from @value{GDBN}; it
829 indicates @code{m4} has finished executing. We can end our @value{GDBN}
830 session with the @value{GDBN} @code{quit} command.
833 (@value{GDBP}) @b{quit}
837 @chapter Getting In and Out of @value{GDBN}
839 This chapter discusses how to start @value{GDBN}, and how to get out of it.
843 type @samp{@value{GDBP}} to start @value{GDBN}.
845 type @kbd{quit}, @kbd{exit} or @kbd{Ctrl-d} to exit.
849 * Invoking GDB:: How to start @value{GDBN}
850 * Quitting GDB:: How to quit @value{GDBN}
851 * Shell Commands:: How to use shell commands inside @value{GDBN}
852 * Logging Output:: How to log @value{GDBN}'s output to a file
856 @section Invoking @value{GDBN}
858 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
859 @value{GDBN} reads commands from the terminal until you tell it to exit.
861 You can also run @code{@value{GDBP}} with a variety of arguments and options,
862 to specify more of your debugging environment at the outset.
864 The command-line options described here are designed
865 to cover a variety of situations; in some environments, some of these
866 options may effectively be unavailable.
868 The most usual way to start @value{GDBN} is with one argument,
869 specifying an executable program:
872 @value{GDBP} @var{program}
876 You can also start with both an executable program and a core file
880 @value{GDBP} @var{program} @var{core}
883 You can, instead, specify a process ID as a second argument or use option
884 @code{-p}, if you want to debug a running process:
887 @value{GDBP} @var{program} 1234
892 would attach @value{GDBN} to process @code{1234}. With option @option{-p} you
893 can omit the @var{program} filename.
895 Taking advantage of the second command-line argument requires a fairly
896 complete operating system; when you use @value{GDBN} as a remote
897 debugger attached to a bare board, there may not be any notion of
898 ``process'', and there is often no way to get a core dump. @value{GDBN}
899 will warn you if it is unable to attach or to read core dumps.
901 You can optionally have @code{@value{GDBP}} pass any arguments after the
902 executable file to the inferior using @code{--args}. This option stops
905 @value{GDBP} --args gcc -O2 -c foo.c
907 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
908 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
910 You can run @code{@value{GDBP}} without printing the front material, which describes
911 @value{GDBN}'s non-warranty, by specifying @code{--silent}
912 (or @code{-q}/@code{--quiet}):
915 @value{GDBP} --silent
919 You can further control how @value{GDBN} starts up by using command-line
920 options. @value{GDBN} itself can remind you of the options available.
930 to display all available options and briefly describe their use
931 (@samp{@value{GDBP} -h} is a shorter equivalent).
933 All options and command line arguments you give are processed
934 in sequential order. The order makes a difference when the
935 @samp{-x} option is used.
939 * File Options:: Choosing files
940 * Mode Options:: Choosing modes
941 * Startup:: What @value{GDBN} does during startup
942 * Initialization Files:: Initialization Files
946 @subsection Choosing Files
948 When @value{GDBN} starts, it reads any arguments other than options as
949 specifying an executable file and core file (or process ID). This is
950 the same as if the arguments were specified by the @samp{-se} and
951 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
952 first argument that does not have an associated option flag as
953 equivalent to the @samp{-se} option followed by that argument; and the
954 second argument that does not have an associated option flag, if any, as
955 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
956 If the second argument begins with a decimal digit, @value{GDBN} will
957 first attempt to attach to it as a process, and if that fails, attempt
958 to open it as a corefile. If you have a corefile whose name begins with
959 a digit, you can prevent @value{GDBN} from treating it as a pid by
960 prefixing it with @file{./}, e.g.@: @file{./12345}.
962 If @value{GDBN} has not been configured to included core file support,
963 such as for most embedded targets, then it will complain about a second
964 argument and ignore it.
966 For the @samp{-s}, @samp{-e}, and @samp{-se} options, and their long
967 form equivalents, the method used to search the file system for the
968 symbol and/or executable file is the same as that used by the
969 @code{file} command. @xref{Files, ,file}.
971 Many options have both long and short forms; both are shown in the
972 following list. @value{GDBN} also recognizes the long forms if you truncate
973 them, so long as enough of the option is present to be unambiguous.
974 (If you prefer, you can flag option arguments with @samp{--} rather
975 than @samp{-}, though we illustrate the more usual convention.)
977 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
978 @c way, both those who look for -foo and --foo in the index, will find
982 @item -symbols @var{file}
984 @cindex @code{--symbols}
986 Read symbol table from file @var{file}.
988 @item -exec @var{file}
990 @cindex @code{--exec}
992 Use file @var{file} as the executable file to execute when appropriate,
993 and for examining pure data in conjunction with a core dump.
997 Read symbol table from file @var{file} and use it as the executable
1000 @item -core @var{file}
1001 @itemx -c @var{file}
1002 @cindex @code{--core}
1004 Use file @var{file} as a core dump to examine.
1006 @item -pid @var{number}
1007 @itemx -p @var{number}
1008 @cindex @code{--pid}
1010 Connect to process ID @var{number}, as with the @code{attach} command.
1012 @item -command @var{file}
1013 @itemx -x @var{file}
1014 @cindex @code{--command}
1016 Execute commands from file @var{file}. The contents of this file is
1017 evaluated exactly as the @code{source} command would.
1018 @xref{Command Files,, Command files}.
1020 @item -eval-command @var{command}
1021 @itemx -ex @var{command}
1022 @cindex @code{--eval-command}
1024 Execute a single @value{GDBN} command.
1026 This option may be used multiple times to call multiple commands. It may
1027 also be interleaved with @samp{-command} as required.
1030 @value{GDBP} -ex 'target sim' -ex 'load' \
1031 -x setbreakpoints -ex 'run' a.out
1034 @item -init-command @var{file}
1035 @itemx -ix @var{file}
1036 @cindex @code{--init-command}
1038 Execute commands from file @var{file} before loading the inferior (but
1039 after loading gdbinit files).
1042 @item -init-eval-command @var{command}
1043 @itemx -iex @var{command}
1044 @cindex @code{--init-eval-command}
1046 Execute a single @value{GDBN} command before loading the inferior (but
1047 after loading gdbinit files).
1050 @item -early-init-command @var{file}
1051 @itemx -eix @var{file}
1052 @cindex @code{--early-init-command}
1054 Execute commands from @var{file} very early in the initialization
1055 process, before any output is produced. @xref{Startup}.
1057 @item -early-init-eval-command @var{command}
1058 @itemx -eiex @var{command}
1059 @cindex @code{--early-init-eval-command}
1060 @cindex @code{-eiex}
1061 Execute a single @value{GDBN} command very early in the initialization
1062 process, before any output is produced.
1064 @item -directory @var{directory}
1065 @itemx -d @var{directory}
1066 @cindex @code{--directory}
1068 Add @var{directory} to the path to search for source and script files.
1072 @cindex @code{--readnow}
1074 Read each symbol file's entire symbol table immediately, rather than
1075 the default, which is to read it incrementally as it is needed.
1076 This makes startup slower, but makes future operations faster.
1079 @anchor{--readnever}
1080 @cindex @code{--readnever}, command-line option
1081 Do not read each symbol file's symbolic debug information. This makes
1082 startup faster but at the expense of not being able to perform
1083 symbolic debugging. DWARF unwind information is also not read,
1084 meaning backtraces may become incomplete or inaccurate. One use of
1085 this is when a user simply wants to do the following sequence: attach,
1086 dump core, detach. Loading the debugging information in this case is
1087 an unnecessary cause of delay.
1091 @subsection Choosing Modes
1093 You can run @value{GDBN} in various alternative modes---for example, in
1094 batch mode or quiet mode.
1102 Do not execute commands found in any initialization files
1103 (@pxref{Initialization Files}).
1108 Do not execute commands found in any home directory initialization
1109 file (@pxref{Initialization Files,,Home directory initialization
1110 file}). The system wide and current directory initialization files
1116 @cindex @code{--quiet}
1117 @cindex @code{--silent}
1119 ``Quiet''. Do not print the introductory and copyright messages. These
1120 messages are also suppressed in batch mode.
1122 @kindex set startup-quietly
1123 @kindex show startup-quietly
1124 This can also be enabled using @code{set startup-quietly on}. The
1125 default is @code{off}. Use @code{show startup-quietly} to see the
1126 current setting. Place @code{set startup-quietly on} into your early
1127 initialization file (@pxref{Initialization Files,,Initialization
1128 Files}) to have future @value{GDBN} sessions startup quietly.
1131 @cindex @code{--batch}
1132 Run in batch mode. Exit with status @code{0} after processing all the
1133 command files specified with @samp{-x} (and all commands from
1134 initialization files, if not inhibited with @samp{-n}). Exit with
1135 nonzero status if an error occurs in executing the @value{GDBN} commands
1136 in the command files. Batch mode also disables pagination, sets unlimited
1137 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1138 off} were in effect (@pxref{Messages/Warnings}).
1140 Batch mode may be useful for running @value{GDBN} as a filter, for
1141 example to download and run a program on another computer; in order to
1142 make this more useful, the message
1145 Program exited normally.
1149 (which is ordinarily issued whenever a program running under
1150 @value{GDBN} control terminates) is not issued when running in batch
1154 @cindex @code{--batch-silent}
1155 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1156 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1157 unaffected). This is much quieter than @samp{-silent} and would be useless
1158 for an interactive session.
1160 This is particularly useful when using targets that give @samp{Loading section}
1161 messages, for example.
1163 Note that targets that give their output via @value{GDBN}, as opposed to
1164 writing directly to @code{stdout}, will also be made silent.
1166 @item -return-child-result
1167 @cindex @code{--return-child-result}
1168 The return code from @value{GDBN} will be the return code from the child
1169 process (the process being debugged), with the following exceptions:
1173 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1174 internal error. In this case the exit code is the same as it would have been
1175 without @samp{-return-child-result}.
1177 The user quits with an explicit value. E.g., @samp{quit 1}.
1179 The child process never runs, or is not allowed to terminate, in which case
1180 the exit code will be -1.
1183 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1184 when @value{GDBN} is being used as a remote program loader or simulator
1189 @cindex @code{--nowindows}
1191 ``No windows''. If @value{GDBN} comes with a graphical user interface
1192 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1193 interface. If no GUI is available, this option has no effect.
1197 @cindex @code{--windows}
1199 If @value{GDBN} includes a GUI, then this option requires it to be
1202 @item -cd @var{directory}
1204 Run @value{GDBN} using @var{directory} as its working directory,
1205 instead of the current directory.
1207 @item -data-directory @var{directory}
1208 @itemx -D @var{directory}
1209 @cindex @code{--data-directory}
1211 Run @value{GDBN} using @var{directory} as its data directory.
1212 The data directory is where @value{GDBN} searches for its
1213 auxiliary files. @xref{Data Files}.
1217 @cindex @code{--fullname}
1219 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1220 subprocess. It tells @value{GDBN} to output the full file name and line
1221 number in a standard, recognizable fashion each time a stack frame is
1222 displayed (which includes each time your program stops). This
1223 recognizable format looks like two @samp{\032} characters, followed by
1224 the file name, line number and character position separated by colons,
1225 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1226 @samp{\032} characters as a signal to display the source code for the
1229 @item -annotate @var{level}
1230 @cindex @code{--annotate}
1231 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1232 effect is identical to using @samp{set annotate @var{level}}
1233 (@pxref{Annotations}). The annotation @var{level} controls how much
1234 information @value{GDBN} prints together with its prompt, values of
1235 expressions, source lines, and other types of output. Level 0 is the
1236 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1237 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1238 that control @value{GDBN}, and level 2 has been deprecated.
1240 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1244 @cindex @code{--args}
1245 Change interpretation of command line so that arguments following the
1246 executable file are passed as command line arguments to the inferior.
1247 This option stops option processing.
1249 @item -baud @var{bps}
1251 @cindex @code{--baud}
1253 Set the line speed (baud rate or bits per second) of any serial
1254 interface used by @value{GDBN} for remote debugging.
1256 @item -l @var{timeout}
1258 Set the timeout (in seconds) of any communication used by @value{GDBN}
1259 for remote debugging.
1261 @item -tty @var{device}
1262 @itemx -t @var{device}
1263 @cindex @code{--tty}
1265 Run using @var{device} for your program's standard input and output.
1266 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1268 @c resolve the situation of these eventually
1270 @cindex @code{--tui}
1271 Activate the @dfn{Text User Interface} when starting. The Text User
1272 Interface manages several text windows on the terminal, showing
1273 source, assembly, registers and @value{GDBN} command outputs
1274 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Do not use this
1275 option if you run @value{GDBN} from Emacs (@pxref{Emacs, ,
1276 Using @value{GDBN} under @sc{gnu} Emacs}).
1278 @item -interpreter @var{interp}
1279 @cindex @code{--interpreter}
1280 Use the interpreter @var{interp} for interface with the controlling
1281 program or device. This option is meant to be set by programs which
1282 communicate with @value{GDBN} using it as a back end.
1283 @xref{Interpreters, , Command Interpreters}.
1285 @samp{--interpreter=mi} (or @samp{--interpreter=mi3}) causes
1286 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} version 3 (@pxref{GDB/MI, ,
1287 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 9.1. @sc{gdb/mi}
1288 version 2 (@code{mi2}), included in @value{GDBN} 6.0 and version 1 (@code{mi1}),
1289 included in @value{GDBN} 5.3, are also available. Earlier @sc{gdb/mi}
1290 interfaces are no longer supported.
1293 @cindex @code{--write}
1294 Open the executable and core files for both reading and writing. This
1295 is equivalent to the @samp{set write on} command inside @value{GDBN}
1299 @cindex @code{--statistics}
1300 This option causes @value{GDBN} to print statistics about time and
1301 memory usage after it completes each command and returns to the prompt.
1304 @cindex @code{--version}
1305 This option causes @value{GDBN} to print its version number and
1306 no-warranty blurb, and exit.
1308 @item -configuration
1309 @cindex @code{--configuration}
1310 This option causes @value{GDBN} to print details about its build-time
1311 configuration parameters, and then exit. These details can be
1312 important when reporting @value{GDBN} bugs (@pxref{GDB Bugs}).
1317 @subsection What @value{GDBN} Does During Startup
1318 @cindex @value{GDBN} startup
1320 Here's the description of what @value{GDBN} does during session startup:
1325 Performs minimal setup required to initialize basic internal state.
1328 @cindex early initialization file
1329 Reads commands from the early initialization file (if any) in your
1330 home directory. Only a restricted set of commands can be placed into
1331 an early initialization file, see @ref{Initialization Files}, for
1335 Executes commands and command files specified by the @samp{-eiex} and
1336 @samp{-eix} command line options in their specified order. Only a
1337 restricted set of commands can be used with @samp{-eiex} and
1338 @samp{eix}, see @ref{Initialization Files}, for details.
1341 Sets up the command interpreter as specified by the command line
1342 (@pxref{Mode Options, interpreter}).
1346 Reads the system wide initialization file and the files from the
1347 system wide initialization directory, @pxref{System Wide Init Files}.
1350 Reads the initialization file (if any) in your home directory and
1351 executes all the commands in that file, @pxref{Home Directory Init
1354 @anchor{Option -init-eval-command}
1356 Executes commands and command files specified by the @samp{-iex} and
1357 @samp{-ix} options in their specified order. Usually you should use the
1358 @samp{-ex} and @samp{-x} options instead, but this way you can apply
1359 settings before @value{GDBN} init files get executed and before inferior
1363 Processes command line options and operands.
1366 Reads and executes the commands from the initialization file (if any)
1367 in the current working directory as long as @samp{set auto-load
1368 local-gdbinit} is set to @samp{on} (@pxref{Init File in the Current
1369 Directory}). This is only done if the current directory is different
1370 from your home directory. Thus, you can have more than one init file,
1371 one generic in your home directory, and another, specific to the
1372 program you are debugging, in the directory where you invoke
1373 @value{GDBN}. @xref{Init File in the Current Directory during
1377 If the command line specified a program to debug, or a process to
1378 attach to, or a core file, @value{GDBN} loads any auto-loaded
1379 scripts provided for the program or for its loaded shared libraries.
1380 @xref{Auto-loading}.
1382 If you wish to disable the auto-loading during startup,
1383 you must do something like the following:
1386 $ gdb -iex "set auto-load python-scripts off" myprogram
1389 Option @samp{-ex} does not work because the auto-loading is then turned
1393 Executes commands and command files specified by the @samp{-ex} and
1394 @samp{-x} options in their specified order. @xref{Command Files}, for
1395 more details about @value{GDBN} command files.
1398 Reads the command history recorded in the @dfn{history file}.
1399 @xref{Command History}, for more details about the command history and the
1400 files where @value{GDBN} records it.
1403 @node Initialization Files
1404 @subsection Initialization Files
1405 @cindex init file name
1407 During startup (@pxref{Startup}) @value{GDBN} will execute commands
1408 from several initialization files. These initialization files use the
1409 same syntax as @dfn{command files} (@pxref{Command Files}) and are
1410 processed by @value{GDBN} in the same way.
1412 To display the list of initialization files loaded by @value{GDBN} at
1413 startup, in the order they will be loaded, you can use @kbd{gdb
1416 @cindex early initialization
1417 The @dfn{early initialization} file is loaded very early in
1418 @value{GDBN}'s initialization process, before the interpreter
1419 (@pxref{Interpreters}) has been initialized, and before the default
1420 target (@pxref{Targets}) is initialized. Only @code{set} or
1421 @code{source} commands should be placed into an early initialization
1422 file, and the only @code{set} commands that can be used are those that
1423 control how @value{GDBN} starts up.
1425 Commands that can be placed into an early initialization file will be
1426 documented as such throughout this manual. Any command that is not
1427 documented as being suitable for an early initialization file should
1428 instead be placed into a general initialization file. Command files
1429 passed to @code{--early-init-command} or @code{-eix} are also early
1430 initialization files, with the same command restrictions. Only
1431 commands that can appear in an early initialization file should be
1432 passed to @code{--early-init-eval-command} or @code{-eiex}.
1434 @cindex general initialization
1435 In contrast, the @dfn{general initialization} files are processed
1436 later, after @value{GDBN} has finished its own internal initialization
1437 process, any valid command can be used in these files.
1439 @cindex initialization file
1440 Throughout the rest of this document the term @dfn{initialization
1441 file} refers to one of the general initialization files, not the early
1442 initialization file. Any discussion of the early initialization file
1443 will specifically mention that it is the early initialization file
1446 As the system wide and home directory initialization files are
1447 processed before most command line options, changes to settings
1448 (e.g.@: @samp{set complaints}) can affect subsequent processing of
1449 command line options and operands.
1451 The following sections describe where @value{GDBN} looks for the early
1452 initialization and initialization files, and the order that the files
1455 @subsubsection Home directory early initialization files
1457 @value{GDBN} initially looks for an early initialization file in the
1458 users home directory@footnote{On DOS/Windows systems, the home
1459 directory is the one pointed to by the @env{HOME} environment
1460 variable.}. There are a number of locations that @value{GDBN} will
1461 search in the home directory, these locations are searched in order
1462 and @value{GDBN} will load the first file that it finds, and
1463 subsequent locations will not be checked.
1465 On non-macOS hosts the locations searched are:
1468 The file @file{gdb/gdbearlyinit} within the directory pointed to by the
1469 environment variable @env{XDG_CONFIG_HOME}, if it is defined.
1471 The file @file{.config/gdb/gdbearlyinit} within the directory pointed to
1472 by the environment variable @env{HOME}, if it is defined.
1474 The file @file{.gdbearlyinit} within the directory pointed to by the
1475 environment variable @env{HOME}, if it is defined.
1478 By contrast, on macOS hosts the locations searched are:
1481 The file @file{Library/Preferences/gdb/gdbearlyinit} within the
1482 directory pointed to by the environment variable @env{HOME}, if it is
1485 The file @file{.gdbearlyinit} within the directory pointed to by the
1486 environment variable @env{HOME}, if it is defined.
1489 It is possible to prevent the home directory early initialization file
1490 from being loaded using the @samp{-nx} or @samp{-nh} command line
1491 options, @pxref{Mode Options,,Choosing Modes}.
1493 @anchor{System Wide Init Files}
1494 @subsubsection System wide initialization files
1496 There are two locations that are searched for system wide
1497 initialization files. Both of these locations are always checked:
1501 @item @file{system.gdbinit}
1502 This is a single system-wide initialization file. Its location is
1503 specified with the @code{--with-system-gdbinit} configure option
1504 (@pxref{System-wide configuration}). It is loaded first when
1505 @value{GDBN} starts, before command line options have been processed.
1507 @item @file{system.gdbinit.d}
1508 This is the system-wide initialization directory. Its location is
1509 specified with the @code{--with-system-gdbinit-dir} configure option
1510 (@pxref{System-wide configuration}). Files in this directory are
1511 loaded in alphabetical order immediately after @file{system.gdbinit}
1512 (if enabled) when @value{GDBN} starts, before command line options
1513 have been processed. Files need to have a recognized scripting
1514 language extension (@file{.py}/@file{.scm}) or be named with a
1515 @file{.gdb} extension to be interpreted as regular @value{GDBN}
1516 commands. @value{GDBN} will not recurse into any subdirectories of
1521 It is possible to prevent the system wide initialization files from
1522 being loaded using the @samp{-nx} command line option, @pxref{Mode
1523 Options,,Choosing Modes}.
1525 @anchor{Home Directory Init File}
1526 @subsubsection Home directory initialization file
1527 @cindex @file{gdbinit}
1528 @cindex @file{.gdbinit}
1529 @cindex @file{gdb.ini}
1531 After loading the system wide initialization files @value{GDBN} will
1532 look for an initialization file in the users home
1533 directory@footnote{On DOS/Windows systems, the home directory is the
1534 one pointed to by the @env{HOME} environment variable.}. There are a
1535 number of locations that @value{GDBN} will search in the home
1536 directory, these locations are searched in order and @value{GDBN} will
1537 load the first file that it finds, and subsequent locations will not
1540 On non-Apple hosts the locations searched are:
1542 @item $XDG_CONFIG_HOME/gdb/gdbinit
1543 @item $HOME/.config/gdb/gdbinit
1544 @item $HOME/.gdbinit
1547 While on Apple hosts the locations searched are:
1549 @item $HOME/Library/Preferences/gdb/gdbinit
1550 @item $HOME/.gdbinit
1553 It is possible to prevent the home directory initialization file from
1554 being loaded using the @samp{-nx} or @samp{-nh} command line options,
1555 @pxref{Mode Options,,Choosing Modes}.
1557 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini} instead of
1558 @file{.gdbinit} or @file{gdbinit}, due to the limitations of file
1559 names imposed by DOS filesystems. The Windows port of @value{GDBN}
1560 uses the standard name, but if it finds a @file{gdb.ini} file in your
1561 home directory, it warns you about that and suggests to rename the
1562 file to the standard name.
1564 @anchor{Init File in the Current Directory during Startup}
1565 @subsubsection Local directory initialization file
1567 @value{GDBN} will check the current directory for a file called
1568 @file{.gdbinit}. It is loaded last, after command line options
1569 other than @samp{-x} and @samp{-ex} have been processed. The command
1570 line options @samp{-x} and @samp{-ex} are processed last, after
1571 @file{.gdbinit} has been loaded, @pxref{File Options,,Choosing
1574 If the file in the current directory was already loaded as the home
1575 directory initialization file then it will not be loaded a second
1578 It is possible to prevent the local directory initialization file from
1579 being loaded using the @samp{-nx} command line option, @pxref{Mode
1580 Options,,Choosing Modes}.
1583 @section Quitting @value{GDBN}
1584 @cindex exiting @value{GDBN}
1585 @cindex leaving @value{GDBN}
1588 @kindex quit @r{[}@var{expression}@r{]}
1589 @kindex exit @r{[}@var{expression}@r{]}
1590 @kindex q @r{(@code{quit})}
1591 @item quit @r{[}@var{expression}@r{]}
1592 @itemx exit @r{[}@var{expression}@r{]}
1594 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1595 @code{q}), the @code{exit} command, or type an end-of-file
1596 character (usually @kbd{Ctrl-d}). If you do not supply @var{expression},
1597 @value{GDBN} will terminate normally; otherwise it will terminate using
1598 the result of @var{expression} as the error code.
1602 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1603 terminates the action of any @value{GDBN} command that is in progress and
1604 returns to @value{GDBN} command level. It is safe to type the interrupt
1605 character at any time because @value{GDBN} does not allow it to take effect
1606 until a time when it is safe.
1608 If you have been using @value{GDBN} to control an attached process or
1609 device, you can release it with the @code{detach} command
1610 (@pxref{Attach, ,Debugging an Already-running Process}).
1612 @node Shell Commands
1613 @section Shell Commands
1615 If you need to execute occasional shell commands during your
1616 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1617 just use the @code{shell} command.
1622 @cindex shell escape
1623 @item shell @var{command-string}
1624 @itemx !@var{command-string}
1625 Invoke a shell to execute @var{command-string}.
1626 Note that no space is needed between @code{!} and @var{command-string}.
1627 On GNU and Unix systems, the environment variable @env{SHELL}, if it
1628 exists, determines which shell to run. Otherwise @value{GDBN} uses
1629 the default shell (@file{/bin/sh} on GNU and Unix systems,
1630 @file{cmd.exe} on MS-Windows, @file{COMMAND.COM} on MS-DOS, etc.).
1633 You may also invoke shell commands from expressions, using the
1634 @code{$_shell} convenience function. @xref{$_shell convenience
1637 The utility @code{make} is often needed in development environments.
1638 You do not have to use the @code{shell} command for this purpose in
1643 @cindex calling make
1644 @item make @var{make-args}
1645 Execute the @code{make} program with the specified
1646 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1652 @cindex send the output of a gdb command to a shell command
1654 @item pipe [@var{command}] | @var{shell_command}
1655 @itemx | [@var{command}] | @var{shell_command}
1656 @itemx pipe -d @var{delim} @var{command} @var{delim} @var{shell_command}
1657 @itemx | -d @var{delim} @var{command} @var{delim} @var{shell_command}
1658 Executes @var{command} and sends its output to @var{shell_command}.
1659 Note that no space is needed around @code{|}.
1660 If no @var{command} is provided, the last command executed is repeated.
1662 In case the @var{command} contains a @code{|}, the option @code{-d @var{delim}}
1663 can be used to specify an alternate delimiter string @var{delim} that separates
1664 the @var{command} from the @var{shell_command}.
1669 (@value{GDBP}) p var
1679 (@value{GDBP}) pipe p var|wc
1681 (@value{GDBP}) |p var|wc -l
1685 (@value{GDBP}) p /x var
1693 (@value{GDBP}) ||grep red
1697 (@value{GDBP}) | -d ! echo this contains a | char\n ! sed -e 's/|/PIPE/'
1698 this contains a PIPE char
1699 (@value{GDBP}) | -d xxx echo this contains a | char!\n xxx sed -e 's/|/PIPE/'
1700 this contains a PIPE char!
1706 The convenience variables @code{$_shell_exitcode} and @code{$_shell_exitsignal}
1707 can be used to examine the exit status of the last shell command launched
1708 by @code{shell}, @code{make}, @code{pipe} and @code{|}.
1709 @xref{Convenience Vars,, Convenience Variables}.
1711 @node Logging Output
1712 @section Logging Output
1713 @cindex logging @value{GDBN} output
1714 @cindex save @value{GDBN} output to a file
1716 You may want to save the output of @value{GDBN} commands to a file.
1717 There are several commands to control @value{GDBN}'s logging.
1720 @kindex set logging enabled
1721 @item set logging enabled [on|off]
1722 Enable or disable logging.
1723 @cindex logging file name
1724 @item set logging file @var{file}
1725 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1726 @item set logging overwrite [on|off]
1727 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1728 you want @code{set logging enabled on} to overwrite the logfile instead.
1729 @item set logging redirect [on|off]
1730 By default, @value{GDBN} output will go to both the terminal and the logfile.
1731 Set @code{redirect} if you want output to go only to the log file.
1732 @item set logging debugredirect [on|off]
1733 By default, @value{GDBN} debug output will go to both the terminal and the logfile.
1734 Set @code{debugredirect} if you want debug output to go only to the log file.
1735 @kindex show logging
1737 Show the current values of the logging settings.
1740 You can also redirect the output of a @value{GDBN} command to a
1741 shell command. @xref{pipe}.
1743 @chapter @value{GDBN} Commands
1745 You can abbreviate a @value{GDBN} command to the first few letters of the command
1746 name, if that abbreviation is unambiguous; and you can repeat certain
1747 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1748 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1749 show you the alternatives available, if there is more than one possibility).
1752 * Command Syntax:: How to give commands to @value{GDBN}
1753 * Command Settings:: How to change default behavior of commands
1754 * Completion:: Command completion
1755 * Command Options:: Command options
1756 * Help:: How to ask @value{GDBN} for help
1759 @node Command Syntax
1760 @section Command Syntax
1762 A @value{GDBN} command is a single line of input. There is no limit on
1763 how long it can be. It starts with a command name, which is followed by
1764 arguments whose meaning depends on the command name. For example, the
1765 command @code{step} accepts an argument which is the number of times to
1766 step, as in @samp{step 5}. You can also use the @code{step} command
1767 with no arguments. Some commands do not allow any arguments.
1769 @cindex abbreviation
1770 @value{GDBN} command names may always be truncated if that abbreviation is
1771 unambiguous. Other possible command abbreviations are listed in the
1772 documentation for individual commands. In some cases, even ambiguous
1773 abbreviations are allowed; for example, @code{s} is specially defined as
1774 equivalent to @code{step} even though there are other commands whose
1775 names start with @code{s}. You can test abbreviations by using them as
1776 arguments to the @code{help} command.
1778 @cindex repeating commands
1779 @kindex RET @r{(repeat last command)}
1780 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1781 repeat the previous command. Certain commands (for example, @code{run})
1782 will not repeat this way; these are commands whose unintentional
1783 repetition might cause trouble and which you are unlikely to want to
1784 repeat. User-defined commands can disable this feature; see
1785 @ref{Define, dont-repeat}.
1787 The @code{list} and @code{x} commands, when you repeat them with
1788 @key{RET}, construct new arguments rather than repeating
1789 exactly as typed. This permits easy scanning of source or memory.
1791 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1792 output, in a way similar to the common utility @code{more}
1793 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1794 @key{RET} too many in this situation, @value{GDBN} disables command
1795 repetition after any command that generates this sort of display.
1797 @kindex # @r{(a comment)}
1799 Any text from a @kbd{#} to the end of the line is a comment; it does
1800 nothing. This is useful mainly in command files (@pxref{Command
1801 Files,,Command Files}).
1803 @cindex repeating command sequences
1804 @kindex Ctrl-o @r{(operate-and-get-next)}
1805 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1806 commands. This command accepts the current line, like @key{RET}, and
1807 then fetches the next line relative to the current line from the history
1811 @node Command Settings
1812 @section Command Settings
1813 @cindex default behavior of commands, changing
1814 @cindex default settings, changing
1816 Many commands change their behavior according to command-specific
1817 variables or settings. These settings can be changed with the
1818 @code{set} subcommands. For example, the @code{print} command
1819 (@pxref{Data, ,Examining Data}) prints arrays differently depending on
1820 settings changeable with the commands @code{set print elements
1821 NUMBER-OF-ELEMENTS} and @code{set print array-indexes}, among others.
1823 You can change these settings to your preference in the gdbinit files
1824 loaded at @value{GDBN} startup. @xref{Startup}.
1826 The settings can also be changed interactively during the debugging
1827 session. For example, to change the limit of array elements to print,
1828 you can do the following:
1830 (@value{GDBP}) set print elements 10
1831 (@value{GDBP}) print some_array
1832 $1 = @{0, 10, 20, 30, 40, 50, 60, 70, 80, 90...@}
1835 The above @code{set print elements 10} command changes the number of
1836 elements to print from the default of 200 to 10. If you only intend
1837 this limit of 10 to be used for printing @code{some_array}, then you
1838 must restore the limit back to 200, with @code{set print elements
1841 Some commands allow overriding settings with command options. For
1842 example, the @code{print} command supports a number of options that
1843 allow overriding relevant global print settings as set by @code{set
1844 print} subcommands. @xref{print options}. The example above could be
1847 (@value{GDBP}) print -elements 10 -- some_array
1848 $1 = @{0, 10, 20, 30, 40, 50, 60, 70, 80, 90...@}
1851 Alternatively, you can use the @code{with} command to change a setting
1852 temporarily, for the duration of a command invocation.
1855 @kindex with command
1856 @kindex w @r{(@code{with})}
1858 @cindex temporarily change settings
1859 @item with @var{setting} [@var{value}] [-- @var{command}]
1860 @itemx w @var{setting} [@var{value}] [-- @var{command}]
1861 Temporarily set @var{setting} to @var{value} for the duration of
1864 @var{setting} is any setting you can change with the @code{set}
1865 subcommands. @var{value} is the value to assign to @code{setting}
1866 while running @code{command}.
1868 If no @var{command} is provided, the last command executed is
1871 If a @var{command} is provided, it must be preceded by a double dash
1872 (@code{--}) separator. This is required because some settings accept
1873 free-form arguments, such as expressions or filenames.
1875 For example, the command
1877 (@value{GDBP}) with print array on -- print some_array
1880 is equivalent to the following 3 commands:
1882 (@value{GDBP}) set print array on
1883 (@value{GDBP}) print some_array
1884 (@value{GDBP}) set print array off
1887 The @code{with} command is particularly useful when you want to
1888 override a setting while running user-defined commands, or commands
1889 defined in Python or Guile. @xref{Extending GDB,, Extending GDB}.
1892 (@value{GDBP}) with print pretty on -- my_complex_command
1895 To change several settings for the same command, you can nest
1896 @code{with} commands. For example, @code{with language ada -- with
1897 print elements 10} temporarily changes the language to Ada and sets a
1898 limit of 10 elements to print for arrays and strings.
1903 @section Command Completion
1906 @cindex word completion
1907 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1908 only one possibility; it can also show you what the valid possibilities
1909 are for the next word in a command, at any time. This works for @value{GDBN}
1910 commands, @value{GDBN} subcommands, command options, and the names of symbols
1913 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1914 of a word. If there is only one possibility, @value{GDBN} fills in the
1915 word, and waits for you to finish the command (or press @key{RET} to
1916 enter it). For example, if you type
1918 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1919 @c complete accuracy in these examples; space introduced for clarity.
1920 @c If texinfo enhancements make it unnecessary, it would be nice to
1921 @c replace " @key" by "@key" in the following...
1923 (@value{GDBP}) info bre@key{TAB}
1927 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1928 the only @code{info} subcommand beginning with @samp{bre}:
1931 (@value{GDBP}) info breakpoints
1935 You can either press @key{RET} at this point, to run the @code{info
1936 breakpoints} command, or backspace and enter something else, if
1937 @samp{breakpoints} does not look like the command you expected. (If you
1938 were sure you wanted @code{info breakpoints} in the first place, you
1939 might as well just type @key{RET} immediately after @samp{info bre},
1940 to exploit command abbreviations rather than command completion).
1942 If there is more than one possibility for the next word when you press
1943 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1944 characters and try again, or just press @key{TAB} a second time;
1945 @value{GDBN} displays all the possible completions for that word. For
1946 example, you might want to set a breakpoint on a subroutine whose name
1947 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1948 just sounds the bell. Typing @key{TAB} again displays all the
1949 function names in your program that begin with those characters, for
1953 (@value{GDBP}) b make_@key{TAB}
1954 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1955 make_a_section_from_file make_environ
1956 make_abs_section make_function_type
1957 make_blockvector make_pointer_type
1958 make_cleanup make_reference_type
1959 make_command make_symbol_completion_list
1960 (@value{GDBP}) b make_
1964 After displaying the available possibilities, @value{GDBN} copies your
1965 partial input (@samp{b make_} in the example) so you can finish the
1968 If the command you are trying to complete expects either a keyword or a
1969 number to follow, then @samp{NUMBER} will be shown among the available
1970 completions, for example:
1973 (@value{GDBP}) print -elements @key{TAB}@key{TAB}
1975 (@value{GDBP}) print -elements@tie{}
1979 Here, the option expects a number (e.g., @code{100}), not literal
1980 @code{NUMBER}. Such metasyntactical arguments are always presented in
1983 If you just want to see the list of alternatives in the first place, you
1984 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1985 means @kbd{@key{META} ?}. You can type this either by holding down a
1986 key designated as the @key{META} shift on your keyboard (if there is
1987 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1989 If the number of possible completions is large, @value{GDBN} will
1990 print as much of the list as it has collected, as well as a message
1991 indicating that the list may be truncated.
1994 (@value{GDBP}) b m@key{TAB}@key{TAB}
1996 <... the rest of the possible completions ...>
1997 *** List may be truncated, max-completions reached. ***
2002 This behavior can be controlled with the following commands:
2005 @kindex set max-completions
2006 @item set max-completions @var{limit}
2007 @itemx set max-completions unlimited
2008 Set the maximum number of completion candidates. @value{GDBN} will
2009 stop looking for more completions once it collects this many candidates.
2010 This is useful when completing on things like function names as collecting
2011 all the possible candidates can be time consuming.
2012 The default value is 200. A value of zero disables tab-completion.
2013 Note that setting either no limit or a very large limit can make
2015 @kindex show max-completions
2016 @item show max-completions
2017 Show the maximum number of candidates that @value{GDBN} will collect and show
2021 @cindex quotes in commands
2022 @cindex completion of quoted strings
2023 Sometimes the string you need, while logically a ``word'', may contain
2024 parentheses or other characters that @value{GDBN} normally excludes from
2025 its notion of a word. To permit word completion to work in this
2026 situation, you may enclose words in @code{'} (single quote marks) in
2027 @value{GDBN} commands.
2029 A likely situation where you might need this is in typing an
2030 expression that involves a C@t{++} symbol name with template
2031 parameters. This is because when completing expressions, GDB treats
2032 the @samp{<} character as word delimiter, assuming that it's the
2033 less-than comparison operator (@pxref{C Operators, , C and C@t{++}
2036 For example, when you want to call a C@t{++} template function
2037 interactively using the @code{print} or @code{call} commands, you may
2038 need to distinguish whether you mean the version of @code{name} that
2039 was specialized for @code{int}, @code{name<int>()}, or the version
2040 that was specialized for @code{float}, @code{name<float>()}. To use
2041 the word-completion facilities in this situation, type a single quote
2042 @code{'} at the beginning of the function name. This alerts
2043 @value{GDBN} that it may need to consider more information than usual
2044 when you press @key{TAB} or @kbd{M-?} to request word completion:
2047 (@value{GDBP}) p 'func<@kbd{M-?}
2048 func<int>() func<float>()
2049 (@value{GDBP}) p 'func<
2052 When setting breakpoints however (@pxref{Location Specifications}), you don't
2053 usually need to type a quote before the function name, because
2054 @value{GDBN} understands that you want to set a breakpoint on a
2058 (@value{GDBP}) b func<@kbd{M-?}
2059 func<int>() func<float>()
2060 (@value{GDBP}) b func<
2063 This is true even in the case of typing the name of C@t{++} overloaded
2064 functions (multiple definitions of the same function, distinguished by
2065 argument type). For example, when you want to set a breakpoint you
2066 don't need to distinguish whether you mean the version of @code{name}
2067 that takes an @code{int} parameter, @code{name(int)}, or the version
2068 that takes a @code{float} parameter, @code{name(float)}.
2071 (@value{GDBP}) b bubble(@kbd{M-?}
2072 bubble(int) bubble(double)
2073 (@value{GDBP}) b bubble(dou@kbd{M-?}
2077 See @ref{quoting names} for a description of other scenarios that
2080 For more information about overloaded functions, see @ref{C Plus Plus
2081 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
2082 overload-resolution off} to disable overload resolution;
2083 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
2085 @cindex completion of structure field names
2086 @cindex structure field name completion
2087 @cindex completion of union field names
2088 @cindex union field name completion
2089 When completing in an expression which looks up a field in a
2090 structure, @value{GDBN} also tries@footnote{The completer can be
2091 confused by certain kinds of invalid expressions. Also, it only
2092 examines the static type of the expression, not the dynamic type.} to
2093 limit completions to the field names available in the type of the
2097 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
2098 magic to_fputs to_rewind
2099 to_data to_isatty to_write
2100 to_delete to_put to_write_async_safe
2105 This is because the @code{gdb_stdout} is a variable of the type
2106 @code{struct ui_file} that is defined in @value{GDBN} sources as
2113 ui_file_flush_ftype *to_flush;
2114 ui_file_write_ftype *to_write;
2115 ui_file_write_async_safe_ftype *to_write_async_safe;
2116 ui_file_fputs_ftype *to_fputs;
2117 ui_file_read_ftype *to_read;
2118 ui_file_delete_ftype *to_delete;
2119 ui_file_isatty_ftype *to_isatty;
2120 ui_file_rewind_ftype *to_rewind;
2121 ui_file_put_ftype *to_put;
2126 @node Command Options
2127 @section Command options
2129 @cindex command options
2130 Some commands accept options starting with a leading dash. For
2131 example, @code{print -pretty}. Similarly to command names, you can
2132 abbreviate a @value{GDBN} option to the first few letters of the
2133 option name, if that abbreviation is unambiguous, and you can also use
2134 the @key{TAB} key to get @value{GDBN} to fill out the rest of a word
2135 in an option (or to show you the alternatives available, if there is
2136 more than one possibility).
2138 @cindex command options, raw input
2139 Some commands take raw input as argument. For example, the print
2140 command processes arbitrary expressions in any of the languages
2141 supported by @value{GDBN}. With such commands, because raw input may
2142 start with a leading dash that would be confused with an option or any
2143 of its abbreviations, e.g.@: @code{print -p} (short for @code{print
2144 -pretty} or printing negative @code{p}?), if you specify any command
2145 option, then you must use a double-dash (@code{--}) delimiter to
2146 indicate the end of options.
2148 @cindex command options, boolean
2150 Some options are described as accepting an argument which can be
2151 either @code{on} or @code{off}. These are known as @dfn{boolean
2152 options}. Similarly to boolean settings commands---@code{on} and
2153 @code{off} are the typical values, but any of @code{1}, @code{yes} and
2154 @code{enable} can also be used as ``true'' value, and any of @code{0},
2155 @code{no} and @code{disable} can also be used as ``false'' value. You
2156 can also omit a ``true'' value, as it is implied by default.
2158 For example, these are equivalent:
2161 (@value{GDBP}) print -object on -pretty off -element unlimited -- *myptr
2162 (@value{GDBP}) p -o -p 0 -e u -- *myptr
2165 You can discover the set of options some command accepts by completing
2166 on @code{-} after the command name. For example:
2169 (@value{GDBP}) print -@key{TAB}@key{TAB}
2170 -address -max-depth -object -static-members
2171 -array -memory-tag-violations -pretty -symbol
2172 -array-indexes -nibbles -raw-values -union
2173 -elements -null-stop -repeats -vtbl
2176 Completion will in some cases guide you with a suggestion of what kind
2177 of argument an option expects. For example:
2180 (@value{GDBP}) print -elements @key{TAB}@key{TAB}
2185 Here, the option expects a number (e.g., @code{100}), not literal
2186 @code{NUMBER}. Such metasyntactical arguments are always presented in
2189 (For more on using the @code{print} command, see @ref{Data, ,Examining
2193 @section Getting Help
2194 @cindex online documentation
2197 You can always ask @value{GDBN} itself for information on its commands,
2198 using the command @code{help}.
2201 @kindex h @r{(@code{help})}
2204 You can use @code{help} (abbreviated @code{h}) with no arguments to
2205 display a short list of named classes of commands:
2209 List of classes of commands:
2211 aliases -- User-defined aliases of other commands
2212 breakpoints -- Making program stop at certain points
2213 data -- Examining data
2214 files -- Specifying and examining files
2215 internals -- Maintenance commands
2216 obscure -- Obscure features
2217 running -- Running the program
2218 stack -- Examining the stack
2219 status -- Status inquiries
2220 support -- Support facilities
2221 tracepoints -- Tracing of program execution without
2222 stopping the program
2223 user-defined -- User-defined commands
2225 Type "help" followed by a class name for a list of
2226 commands in that class.
2227 Type "help" followed by command name for full
2229 Command name abbreviations are allowed if unambiguous.
2232 @c the above line break eliminates huge line overfull...
2234 @item help @var{class}
2235 Using one of the general help classes as an argument, you can get a
2236 list of the individual commands in that class. If a command has
2237 aliases, the aliases are given after the command name, separated by
2238 commas. If an alias has default arguments, the full definition of
2239 the alias is given after the first line.
2240 For example, here is the help display for the class @code{status}:
2243 (@value{GDBP}) help status
2248 @c Line break in "show" line falsifies real output, but needed
2249 @c to fit in smallbook page size.
2250 info, inf, i -- Generic command for showing things
2251 about the program being debugged
2252 info address, iamain -- Describe where symbol SYM is stored.
2253 alias iamain = info address main
2254 info all-registers -- List of all registers and their contents,
2255 for selected stack frame.
2257 show, info set -- Generic command for showing things
2260 Type "help" followed by command name for full
2262 Command name abbreviations are allowed if unambiguous.
2266 @item help @var{command}
2267 With a command name as @code{help} argument, @value{GDBN} displays a
2268 short paragraph on how to use that command. If that command has
2269 one or more aliases, @value{GDBN} will display a first line with
2270 the command name and all its aliases separated by commas.
2271 This first line will be followed by the full definition of all aliases
2272 having default arguments.
2273 When asking the help for an alias, the documentation for the aliased
2276 A user-defined alias can optionally be documented using the
2277 @code{document} command (@pxref{Define, document}). @value{GDBN} then
2278 considers this alias as different from the aliased command: this alias
2279 is not listed in the aliased command help output, and asking help for
2280 this alias will show the documentation provided for the alias instead of
2281 the documentation of the aliased command.
2284 @item apropos [-v] @var{regexp}
2285 The @code{apropos} command searches through all of the @value{GDBN}
2286 commands and aliases, and their documentation, for the regular expression specified in
2287 @var{args}. It prints out all matches found. The optional flag @samp{-v},
2288 which stands for @samp{verbose}, indicates to output the full documentation
2289 of the matching commands and highlight the parts of the documentation
2290 matching @var{regexp}. For example:
2301 alias -- Define a new command that is an alias of an existing command
2302 aliases -- User-defined aliases of other commands
2310 apropos -v cut.*thread apply
2314 results in the below output, where @samp{cut for 'thread apply}
2315 is highlighted if styling is enabled.
2319 taas -- Apply a command to all threads (ignoring errors
2322 shortcut for 'thread apply all -s COMMAND'
2324 tfaas -- Apply a command to all frames of all threads
2325 (ignoring errors and empty output).
2326 Usage: tfaas COMMAND
2327 shortcut for 'thread apply all -s frame apply all -s COMMAND'
2332 @item complete @var{args}
2333 The @code{complete @var{args}} command lists all the possible completions
2334 for the beginning of a command. Use @var{args} to specify the beginning of the
2335 command you want completed. For example:
2341 @noindent results in:
2352 @noindent This is intended for use by @sc{gnu} Emacs.
2355 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
2356 and @code{show} to inquire about the state of your program, or the state
2357 of @value{GDBN} itself. Each command supports many topics of inquiry; this
2358 manual introduces each of them in the appropriate context. The listings
2359 under @code{info} and under @code{show} in the Command, Variable, and
2360 Function Index point to all the sub-commands. @xref{Command and Variable
2366 @kindex i @r{(@code{info})}
2368 This command (abbreviated @code{i}) is for describing the state of your
2369 program. For example, you can show the arguments passed to a function
2370 with @code{info args}, list the registers currently in use with @code{info
2371 registers}, or list the breakpoints you have set with @code{info breakpoints}.
2372 You can get a complete list of the @code{info} sub-commands with
2373 @w{@code{help info}}.
2377 You can assign the result of an expression to an environment variable with
2378 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
2379 @code{set prompt $}.
2383 In contrast to @code{info}, @code{show} is for describing the state of
2384 @value{GDBN} itself.
2385 You can change most of the things you can @code{show}, by using the
2386 related command @code{set}; for example, you can control what number
2387 system is used for displays with @code{set radix}, or simply inquire
2388 which is currently in use with @code{show radix}.
2391 To display all the settable parameters and their current
2392 values, you can use @code{show} with no arguments; you may also use
2393 @code{info set}. Both commands produce the same display.
2394 @c FIXME: "info set" violates the rule that "info" is for state of
2395 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
2396 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
2400 Here are several miscellaneous @code{show} subcommands, all of which are
2401 exceptional in lacking corresponding @code{set} commands:
2404 @kindex show version
2405 @cindex @value{GDBN} version number
2407 Show what version of @value{GDBN} is running. You should include this
2408 information in @value{GDBN} bug-reports. If multiple versions of
2409 @value{GDBN} are in use at your site, you may need to determine which
2410 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
2411 commands are introduced, and old ones may wither away. Also, many
2412 system vendors ship variant versions of @value{GDBN}, and there are
2413 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
2414 The version number is the same as the one announced when you start
2417 @kindex show copying
2418 @kindex info copying
2419 @cindex display @value{GDBN} copyright
2422 Display information about permission for copying @value{GDBN}.
2424 @kindex show warranty
2425 @kindex info warranty
2427 @itemx info warranty
2428 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
2429 if your version of @value{GDBN} comes with one.
2431 @kindex show configuration
2432 @item show configuration
2433 Display detailed information about the way @value{GDBN} was configured
2434 when it was built. This displays the optional arguments passed to the
2435 @file{configure} script and also configuration parameters detected
2436 automatically by @command{configure}. When reporting a @value{GDBN}
2437 bug (@pxref{GDB Bugs}), it is important to include this information in
2443 @chapter Running Programs Under @value{GDBN}
2445 When you run a program under @value{GDBN}, you must first generate
2446 debugging information when you compile it.
2448 You may start @value{GDBN} with its arguments, if any, in an environment
2449 of your choice. If you are doing native debugging, you may redirect
2450 your program's input and output, debug an already running process, or
2451 kill a child process.
2454 * Compilation:: Compiling for debugging
2455 * Starting:: Starting your program
2456 * Arguments:: Your program's arguments
2457 * Environment:: Your program's environment
2459 * Working Directory:: Your program's working directory
2460 * Input/Output:: Your program's input and output
2461 * Attach:: Debugging an already-running process
2462 * Kill Process:: Killing the child process
2463 * Inferiors Connections and Programs:: Debugging multiple inferiors
2464 connections and programs
2465 * Threads:: Debugging programs with multiple threads
2466 * Forks:: Debugging forks
2467 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
2471 @section Compiling for Debugging
2473 In order to debug a program effectively, you need to generate
2474 debugging information when you compile it. This debugging information
2475 is stored in the object file; it describes the data type of each
2476 variable or function and the correspondence between source line numbers
2477 and addresses in the executable code.
2479 To request debugging information, specify the @samp{-g} option when you run
2482 Programs that are to be shipped to your customers are compiled with
2483 optimizations, using the @samp{-O} compiler option. However, some
2484 compilers are unable to handle the @samp{-g} and @samp{-O} options
2485 together. Using those compilers, you cannot generate optimized
2486 executables containing debugging information.
2488 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
2489 without @samp{-O}, making it possible to debug optimized code. We
2490 recommend that you @emph{always} use @samp{-g} whenever you compile a
2491 program. You may think your program is correct, but there is no sense
2492 in pushing your luck. For more information, see @ref{Optimized Code}.
2494 Older versions of the @sc{gnu} C compiler permitted a variant option
2495 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
2496 format; if your @sc{gnu} C compiler has this option, do not use it.
2498 @value{GDBN} knows about preprocessor macros and can show you their
2499 expansion (@pxref{Macros}). Most compilers do not include information
2500 about preprocessor macros in the debugging information if you specify
2501 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
2502 the @sc{gnu} C compiler, provides macro information if you are using
2503 the DWARF debugging format, and specify the option @option{-g3}.
2505 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
2506 gcc, Using the @sc{gnu} Compiler Collection (GCC)}, for more
2507 information on @value{NGCC} options affecting debug information.
2509 You will have the best debugging experience if you use the latest
2510 version of the DWARF debugging format that your compiler supports.
2511 DWARF is currently the most expressive and best supported debugging
2512 format in @value{GDBN}.
2516 @section Starting your Program
2522 @kindex r @r{(@code{run})}
2525 Use the @code{run} command to start your program under @value{GDBN}.
2526 You must first specify the program name with an argument to
2527 @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
2528 @value{GDBN}}), or by using the @code{file} or @code{exec-file}
2529 command (@pxref{Files, ,Commands to Specify Files}).
2533 If you are running your program in an execution environment that
2534 supports processes, @code{run} creates an inferior process and makes
2535 that process run your program. In some environments without processes,
2536 @code{run} jumps to the start of your program. Other targets,
2537 like @samp{remote}, are always running. If you get an error
2538 message like this one:
2541 The "remote" target does not support "run".
2542 Try "help target" or "continue".
2546 then use @code{continue} to run your program. You may need @code{load}
2547 first (@pxref{load}).
2549 The execution of a program is affected by certain information it
2550 receives from its superior. @value{GDBN} provides ways to specify this
2551 information, which you must do @emph{before} starting your program. (You
2552 can change it after starting your program, but such changes only affect
2553 your program the next time you start it.) This information may be
2554 divided into four categories:
2557 @item The @emph{arguments.}
2558 Specify the arguments to give your program as the arguments of the
2559 @code{run} command. If a shell is available on your target, the shell
2560 is used to pass the arguments, so that you may use normal conventions
2561 (such as wildcard expansion or variable substitution) in describing
2563 In Unix systems, you can control which shell is used with the
2564 @env{SHELL} environment variable. If you do not define @env{SHELL},
2565 @value{GDBN} uses the default shell (@file{/bin/sh}). You can disable
2566 use of any shell with the @code{set startup-with-shell} command (see
2569 @item The @emph{environment.}
2570 Your program normally inherits its environment from @value{GDBN}, but you can
2571 use the @value{GDBN} commands @code{set environment} and @code{unset
2572 environment} to change parts of the environment that affect
2573 your program. @xref{Environment, ,Your Program's Environment}.
2575 @item The @emph{working directory.}
2576 You can set your program's working directory with the command
2577 @kbd{set cwd}. If you do not set any working directory with this
2578 command, your program will inherit @value{GDBN}'s working directory if
2579 native debugging, or the remote server's working directory if remote
2580 debugging. @xref{Working Directory, ,Your Program's Working
2583 @item The @emph{standard input and output.}
2584 Your program normally uses the same device for standard input and
2585 standard output as @value{GDBN} is using. You can redirect input and output
2586 in the @code{run} command line, or you can use the @code{tty} command to
2587 set a different device for your program.
2588 @xref{Input/Output, ,Your Program's Input and Output}.
2591 @emph{Warning:} While input and output redirection work, you cannot use
2592 pipes to pass the output of the program you are debugging to another
2593 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2597 When you issue the @code{run} command, your program begins to execute
2598 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2599 of how to arrange for your program to stop. Once your program has
2600 stopped, you may call functions in your program, using the @code{print}
2601 or @code{call} commands. @xref{Data, ,Examining Data}.
2603 If the modification time of your symbol file has changed since the last
2604 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2605 table, and reads it again. When it does this, @value{GDBN} tries to retain
2606 your current breakpoints.
2611 @cindex run to main procedure
2612 The name of the main procedure can vary from language to language.
2613 With C or C@t{++}, the main procedure name is always @code{main}, but
2614 other languages such as Ada do not require a specific name for their
2615 main procedure. The debugger provides a convenient way to start the
2616 execution of the program and to stop at the beginning of the main
2617 procedure, depending on the language used.
2619 The @samp{start} command does the equivalent of setting a temporary
2620 breakpoint at the beginning of the main procedure and then invoking
2621 the @samp{run} command.
2623 @cindex elaboration phase
2624 Some programs contain an @dfn{elaboration} phase where some startup code is
2625 executed before the main procedure is called. This depends on the
2626 languages used to write your program. In C@t{++}, for instance,
2627 constructors for static and global objects are executed before
2628 @code{main} is called. It is therefore possible that the debugger stops
2629 before reaching the main procedure. However, the temporary breakpoint
2630 will remain to halt execution.
2632 Specify the arguments to give to your program as arguments to the
2633 @samp{start} command. These arguments will be given verbatim to the
2634 underlying @samp{run} command. Note that the same arguments will be
2635 reused if no argument is provided during subsequent calls to
2636 @samp{start} or @samp{run}.
2638 It is sometimes necessary to debug the program during elaboration. In
2639 these cases, using the @code{start} command would stop the execution
2640 of your program too late, as the program would have already completed
2641 the elaboration phase. Under these circumstances, either insert
2642 breakpoints in your elaboration code before running your program or
2643 use the @code{starti} command.
2647 @cindex run to first instruction
2648 The @samp{starti} command does the equivalent of setting a temporary
2649 breakpoint at the first instruction of a program's execution and then
2650 invoking the @samp{run} command. For programs containing an
2651 elaboration phase, the @code{starti} command will stop execution at
2652 the start of the elaboration phase.
2654 @anchor{set exec-wrapper}
2655 @kindex set exec-wrapper
2656 @item set exec-wrapper @var{wrapper}
2657 @itemx show exec-wrapper
2658 @itemx unset exec-wrapper
2659 When @samp{exec-wrapper} is set, the specified wrapper is used to
2660 launch programs for debugging. @value{GDBN} starts your program
2661 with a shell command of the form @kbd{exec @var{wrapper}
2662 @var{program}}. Quoting is added to @var{program} and its
2663 arguments, but not to @var{wrapper}, so you should add quotes if
2664 appropriate for your shell. The wrapper runs until it executes
2665 your program, and then @value{GDBN} takes control.
2667 You can use any program that eventually calls @code{execve} with
2668 its arguments as a wrapper. Several standard Unix utilities do
2669 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2670 with @code{exec "$@@"} will also work.
2672 For example, you can use @code{env} to pass an environment variable to
2673 the debugged program, without setting the variable in your shell's
2677 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2681 This command is available when debugging locally on most targets, excluding
2682 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2684 @kindex set startup-with-shell
2685 @anchor{set startup-with-shell}
2686 @item set startup-with-shell
2687 @itemx set startup-with-shell on
2688 @itemx set startup-with-shell off
2689 @itemx show startup-with-shell
2690 On Unix systems, by default, if a shell is available on your target,
2691 @value{GDBN}) uses it to start your program. Arguments of the
2692 @code{run} command are passed to the shell, which does variable
2693 substitution, expands wildcard characters and performs redirection of
2694 I/O. In some circumstances, it may be useful to disable such use of a
2695 shell, for example, when debugging the shell itself or diagnosing
2696 startup failures such as:
2700 Starting program: ./a.out
2701 During startup program terminated with signal SIGSEGV, Segmentation fault.
2705 which indicates the shell or the wrapper specified with
2706 @samp{exec-wrapper} crashed, not your program. Most often, this is
2707 caused by something odd in your shell's non-interactive mode
2708 initialization file---such as @file{.cshrc} for C-shell,
2709 $@file{.zshenv} for the Z shell, or the file specified in the
2710 @env{BASH_ENV} environment variable for BASH.
2712 @anchor{set auto-connect-native-target}
2713 @kindex set auto-connect-native-target
2714 @item set auto-connect-native-target
2715 @itemx set auto-connect-native-target on
2716 @itemx set auto-connect-native-target off
2717 @itemx show auto-connect-native-target
2719 By default, if the current inferior is not connected to any target yet
2720 (e.g., with @code{target remote}), the @code{run} command starts your
2721 program as a native process under @value{GDBN}, on your local machine.
2722 If you're sure you don't want to debug programs on your local machine,
2723 you can tell @value{GDBN} to not connect to the native target
2724 automatically with the @code{set auto-connect-native-target off}
2727 If @code{on}, which is the default, and if the current inferior is not
2728 connected to a target already, the @code{run} command automaticaly
2729 connects to the native target, if one is available.
2731 If @code{off}, and if the current inferior is not connected to a
2732 target already, the @code{run} command fails with an error:
2736 Don't know how to run. Try "help target".
2739 If the current inferior is already connected to a target, @value{GDBN}
2740 always uses it with the @code{run} command.
2742 In any case, you can explicitly connect to the native target with the
2743 @code{target native} command. For example,
2746 (@value{GDBP}) set auto-connect-native-target off
2748 Don't know how to run. Try "help target".
2749 (@value{GDBP}) target native
2751 Starting program: ./a.out
2752 [Inferior 1 (process 10421) exited normally]
2755 In case you connected explicitly to the @code{native} target,
2756 @value{GDBN} remains connected even if all inferiors exit, ready for
2757 the next @code{run} command. Use the @code{disconnect} command to
2760 Examples of other commands that likewise respect the
2761 @code{auto-connect-native-target} setting: @code{attach}, @code{info
2762 proc}, @code{info os}.
2764 @kindex set disable-randomization
2765 @item set disable-randomization
2766 @itemx set disable-randomization on
2767 This option (enabled by default in @value{GDBN}) will turn off the native
2768 randomization of the virtual address space of the started program. This option
2769 is useful for multiple debugging sessions to make the execution better
2770 reproducible and memory addresses reusable across debugging sessions.
2772 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2773 On @sc{gnu}/Linux you can get the same behavior using
2776 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2779 @item set disable-randomization off
2780 Leave the behavior of the started executable unchanged. Some bugs rear their
2781 ugly heads only when the program is loaded at certain addresses. If your bug
2782 disappears when you run the program under @value{GDBN}, that might be because
2783 @value{GDBN} by default disables the address randomization on platforms, such
2784 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2785 disable-randomization off} to try to reproduce such elusive bugs.
2787 On targets where it is available, virtual address space randomization
2788 protects the programs against certain kinds of security attacks. In these
2789 cases the attacker needs to know the exact location of a concrete executable
2790 code. Randomizing its location makes it impossible to inject jumps misusing
2791 a code at its expected addresses.
2793 Prelinking shared libraries provides a startup performance advantage but it
2794 makes addresses in these libraries predictable for privileged processes by
2795 having just unprivileged access at the target system. Reading the shared
2796 library binary gives enough information for assembling the malicious code
2797 misusing it. Still even a prelinked shared library can get loaded at a new
2798 random address just requiring the regular relocation process during the
2799 startup. Shared libraries not already prelinked are always loaded at
2800 a randomly chosen address.
2802 Position independent executables (PIE) contain position independent code
2803 similar to the shared libraries and therefore such executables get loaded at
2804 a randomly chosen address upon startup. PIE executables always load even
2805 already prelinked shared libraries at a random address. You can build such
2806 executable using @command{gcc -fPIE -pie}.
2808 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2809 (as long as the randomization is enabled).
2811 @item show disable-randomization
2812 Show the current setting of the explicit disable of the native randomization of
2813 the virtual address space of the started program.
2818 @section Your Program's Arguments
2820 @cindex arguments (to your program)
2821 The arguments to your program can be specified by the arguments of the
2823 They are passed to a shell, which expands wildcard characters and
2824 performs redirection of I/O, and thence to your program. Your
2825 @env{SHELL} environment variable (if it exists) specifies what shell
2826 @value{GDBN} uses. If you do not define @env{SHELL}, @value{GDBN} uses
2827 the default shell (@file{/bin/sh} on Unix).
2829 On non-Unix systems, the program is usually invoked directly by
2830 @value{GDBN}, which emulates I/O redirection via the appropriate system
2831 calls, and the wildcard characters are expanded by the startup code of
2832 the program, not by the shell.
2834 @code{run} with no arguments uses the same arguments used by the previous
2835 @code{run}, or those set by the @code{set args} command.
2840 Specify the arguments to be used the next time your program is run. If
2841 @code{set args} has no arguments, @code{run} executes your program
2842 with no arguments. Once you have run your program with arguments,
2843 using @code{set args} before the next @code{run} is the only way to run
2844 it again without arguments.
2848 Show the arguments to give your program when it is started.
2852 @section Your Program's Environment
2854 @cindex environment (of your program)
2855 The @dfn{environment} consists of a set of environment variables and
2856 their values. Environment variables conventionally record such things as
2857 your user name, your home directory, your terminal type, and your search
2858 path for programs to run. Usually you set up environment variables with
2859 the shell and they are inherited by all the other programs you run. When
2860 debugging, it can be useful to try running your program with a modified
2861 environment without having to start @value{GDBN} over again.
2865 @item path @var{directory}
2866 Add @var{directory} to the front of the @env{PATH} environment variable
2867 (the search path for executables) that will be passed to your program.
2868 The value of @env{PATH} used by @value{GDBN} does not change.
2869 You may specify several directory names, separated by whitespace or by a
2870 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2871 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2872 is moved to the front, so it is searched sooner.
2874 You can use the string @samp{$cwd} to refer to whatever is the current
2875 working directory at the time @value{GDBN} searches the path. If you
2876 use @samp{.} instead, it refers to the directory where you executed the
2877 @code{path} command. @value{GDBN} replaces @samp{.} in the
2878 @var{directory} argument (with the current path) before adding
2879 @var{directory} to the search path.
2880 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2881 @c document that, since repeating it would be a no-op.
2885 Display the list of search paths for executables (the @env{PATH}
2886 environment variable).
2888 @kindex show environment
2889 @item show environment @r{[}@var{varname}@r{]}
2890 Print the value of environment variable @var{varname} to be given to
2891 your program when it starts. If you do not supply @var{varname},
2892 print the names and values of all environment variables to be given to
2893 your program. You can abbreviate @code{environment} as @code{env}.
2895 @kindex set environment
2896 @anchor{set environment}
2897 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2898 Set environment variable @var{varname} to @var{value}. The value
2899 changes for your program (and the shell @value{GDBN} uses to launch
2900 it), not for @value{GDBN} itself. The @var{value} may be any string; the
2901 values of environment variables are just strings, and any
2902 interpretation is supplied by your program itself. The @var{value}
2903 parameter is optional; if it is eliminated, the variable is set to a
2905 @c "any string" here does not include leading, trailing
2906 @c blanks. Gnu asks: does anyone care?
2908 For example, this command:
2915 tells the debugged program, when subsequently run, that its user is named
2916 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2917 are not actually required.)
2919 Note that on Unix systems, @value{GDBN} runs your program via a shell,
2920 which also inherits the environment set with @code{set environment}.
2921 If necessary, you can avoid that by using the @samp{env} program as a
2922 wrapper instead of using @code{set environment}. @xref{set
2923 exec-wrapper}, for an example doing just that.
2925 Environment variables that are set by the user are also transmitted to
2926 @command{gdbserver} to be used when starting the remote inferior.
2927 @pxref{QEnvironmentHexEncoded}.
2929 @kindex unset environment
2930 @anchor{unset environment}
2931 @item unset environment @var{varname}
2932 Remove variable @var{varname} from the environment to be passed to your
2933 program. This is different from @samp{set env @var{varname} =};
2934 @code{unset environment} removes the variable from the environment,
2935 rather than assigning it an empty value.
2937 Environment variables that are unset by the user are also unset on
2938 @command{gdbserver} when starting the remote inferior.
2939 @pxref{QEnvironmentUnset}.
2942 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2943 the shell indicated by your @env{SHELL} environment variable if it
2944 exists (or @code{/bin/sh} if not). If your @env{SHELL} variable
2945 names a shell that runs an initialization file when started
2946 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2947 for the Z shell, or the file specified in the @env{BASH_ENV}
2948 environment variable for BASH---any variables you set in that file
2949 affect your program. You may wish to move setting of environment
2950 variables to files that are only run when you sign on, such as
2951 @file{.login} or @file{.profile}.
2953 @node Working Directory
2954 @section Your Program's Working Directory
2956 @cindex working directory (of your program)
2957 Each time you start your program with @code{run}, the inferior will be
2958 initialized with the current working directory specified by the
2959 @kbd{set cwd} command. If no directory has been specified by this
2960 command, then the inferior will inherit @value{GDBN}'s current working
2961 directory as its working directory if native debugging, or it will
2962 inherit the remote server's current working directory if remote
2967 @cindex change inferior's working directory
2968 @anchor{set cwd command}
2969 @item set cwd @r{[}@var{directory}@r{]}
2970 Set the inferior's working directory to @var{directory}, which will be
2971 @code{glob}-expanded in order to resolve tildes (@file{~}). If no
2972 argument has been specified, the command clears the setting and resets
2973 it to an empty state. This setting has no effect on @value{GDBN}'s
2974 working directory, and it only takes effect the next time you start
2975 the inferior. The @file{~} in @var{directory} is a short for the
2976 @dfn{home directory}, usually pointed to by the @env{HOME} environment
2977 variable. On MS-Windows, if @env{HOME} is not defined, @value{GDBN}
2978 uses the concatenation of @env{HOMEDRIVE} and @env{HOMEPATH} as
2981 You can also change @value{GDBN}'s current working directory by using
2982 the @code{cd} command.
2986 @cindex show inferior's working directory
2988 Show the inferior's working directory. If no directory has been
2989 specified by @kbd{set cwd}, then the default inferior's working
2990 directory is the same as @value{GDBN}'s working directory.
2993 @cindex change @value{GDBN}'s working directory
2995 @item cd @r{[}@var{directory}@r{]}
2996 Set the @value{GDBN} working directory to @var{directory}. If not
2997 given, @var{directory} uses @file{'~'}.
2999 The @value{GDBN} working directory serves as a default for the
3000 commands that specify files for @value{GDBN} to operate on.
3001 @xref{Files, ,Commands to Specify Files}.
3002 @xref{set cwd command}.
3006 Print the @value{GDBN} working directory.
3009 It is generally impossible to find the current working directory of
3010 the process being debugged (since a program can change its directory
3011 during its run). If you work on a system where @value{GDBN} supports
3012 the @code{info proc} command (@pxref{Process Information}), you can
3013 use the @code{info proc} command to find out the
3014 current working directory of the debuggee.
3017 @section Your Program's Input and Output
3022 By default, the program you run under @value{GDBN} does input and output to
3023 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
3024 to its own terminal modes to interact with you, but it records the terminal
3025 modes your program was using and switches back to them when you continue
3026 running your program.
3029 @kindex info terminal
3031 Displays information recorded by @value{GDBN} about the terminal modes your
3035 You can redirect your program's input and/or output using shell
3036 redirection with the @code{run} command. For example,
3043 starts your program, diverting its output to the file @file{outfile}.
3046 @cindex controlling terminal
3047 Another way to specify where your program should do input and output is
3048 with the @code{tty} command. This command accepts a file name as
3049 argument, and causes this file to be the default for future @code{run}
3050 commands. It also resets the controlling terminal for the child
3051 process, for future @code{run} commands. For example,
3058 directs that processes started with subsequent @code{run} commands
3059 default to do input and output on the terminal @file{/dev/ttyb} and have
3060 that as their controlling terminal.
3062 An explicit redirection in @code{run} overrides the @code{tty} command's
3063 effect on the input/output device, but not its effect on the controlling
3066 When you use the @code{tty} command or redirect input in the @code{run}
3067 command, only the input @emph{for your program} is affected. The input
3068 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
3069 for @code{set inferior-tty}.
3071 @cindex inferior tty
3072 @cindex set inferior controlling terminal
3073 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
3074 display the name of the terminal that will be used for future runs of your
3078 @item set inferior-tty [ @var{tty} ]
3079 @kindex set inferior-tty
3080 Set the tty for the program being debugged to @var{tty}. Omitting @var{tty}
3081 restores the default behavior, which is to use the same terminal as
3084 @item show inferior-tty
3085 @kindex show inferior-tty
3086 Show the current tty for the program being debugged.
3090 @section Debugging an Already-running Process
3095 @item attach @var{process-id}
3096 This command attaches to a running process---one that was started
3097 outside @value{GDBN}. (@code{info files} shows your active
3098 targets.) The command takes as argument a process ID. The usual way to
3099 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
3100 or with the @samp{jobs -l} shell command.
3102 @code{attach} does not repeat if you press @key{RET} a second time after
3103 executing the command.
3106 To use @code{attach}, your program must be running in an environment
3107 which supports processes; for example, @code{attach} does not work for
3108 programs on bare-board targets that lack an operating system. You must
3109 also have permission to send the process a signal.
3111 When you use @code{attach}, the debugger finds the program running in
3112 the process first by looking in the current working directory, then (if
3113 the program is not found) by using the source file search path
3114 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
3115 the @code{file} command to load the program. @xref{Files, ,Commands to
3118 @anchor{set exec-file-mismatch}
3119 If the debugger can determine that the executable file running in the
3120 process it is attaching to does not match the current exec-file loaded
3121 by @value{GDBN}, the option @code{exec-file-mismatch} specifies how to
3122 handle the mismatch. @value{GDBN} tries to compare the files by
3123 comparing their build IDs (@pxref{build ID}), if available.
3126 @kindex exec-file-mismatch
3127 @cindex set exec-file-mismatch
3128 @item set exec-file-mismatch @samp{ask|warn|off}
3130 Whether to detect mismatch between the current executable file loaded
3131 by @value{GDBN} and the executable file used to start the process. If
3132 @samp{ask}, the default, display a warning and ask the user whether to
3133 load the process executable file; if @samp{warn}, just display a
3134 warning; if @samp{off}, don't attempt to detect a mismatch.
3135 If the user confirms loading the process executable file, then its symbols
3136 will be loaded as well.
3138 @cindex show exec-file-mismatch
3139 @item show exec-file-mismatch
3140 Show the current value of @code{exec-file-mismatch}.
3144 The first thing @value{GDBN} does after arranging to debug the specified
3145 process is to stop it. You can examine and modify an attached process
3146 with all the @value{GDBN} commands that are ordinarily available when
3147 you start processes with @code{run}. You can insert breakpoints; you
3148 can step and continue; you can modify storage. If you would rather the
3149 process continue running, you may use the @code{continue} command after
3150 attaching @value{GDBN} to the process.
3155 When you have finished debugging the attached process, you can use the
3156 @code{detach} command to release it from @value{GDBN} control. Detaching
3157 the process continues its execution. After the @code{detach} command,
3158 that process and @value{GDBN} become completely independent once more, and you
3159 are ready to @code{attach} another process or start one with @code{run}.
3160 @code{detach} does not repeat if you press @key{RET} again after
3161 executing the command.
3164 If you exit @value{GDBN} while you have an attached process, you detach
3165 that process. If you use the @code{run} command, you kill that process.
3166 By default, @value{GDBN} asks for confirmation if you try to do either of these
3167 things; you can control whether or not you need to confirm by using the
3168 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
3172 @section Killing the Child Process
3177 Kill the child process in which your program is running under @value{GDBN}.
3180 This command is useful if you wish to debug a core dump instead of a
3181 running process. @value{GDBN} ignores any core dump file while your program
3184 On some operating systems, a program cannot be executed outside @value{GDBN}
3185 while you have breakpoints set on it inside @value{GDBN}. You can use the
3186 @code{kill} command in this situation to permit running your program
3187 outside the debugger.
3189 The @code{kill} command is also useful if you wish to recompile and
3190 relink your program, since on many systems it is impossible to modify an
3191 executable file while it is running in a process. In this case, when you
3192 next type @code{run}, @value{GDBN} notices that the file has changed, and
3193 reads the symbol table again (while trying to preserve your current
3194 breakpoint settings).
3196 @node Inferiors Connections and Programs
3197 @section Debugging Multiple Inferiors Connections and Programs
3199 @value{GDBN} lets you run and debug multiple programs in a single
3200 session. In addition, @value{GDBN} on some systems may let you run
3201 several programs simultaneously (otherwise you have to exit from one
3202 before starting another). On some systems @value{GDBN} may even let
3203 you debug several programs simultaneously on different remote systems.
3204 In the most general case, you can have multiple threads of execution
3205 in each of multiple processes, launched from multiple executables,
3206 running on different machines.
3209 @value{GDBN} represents the state of each program execution with an
3210 object called an @dfn{inferior}. An inferior typically corresponds to
3211 a process, but is more general and applies also to targets that do not
3212 have processes. Inferiors may be created before a process runs, and
3213 may be retained after a process exits. Inferiors have unique
3214 identifiers that are different from process ids. Usually each
3215 inferior will also have its own distinct address space, although some
3216 embedded targets may have several inferiors running in different parts
3217 of a single address space. Each inferior may in turn have multiple
3218 threads running in it.
3221 The commands @code{info inferiors} and @code{info connections}, which will be
3222 introduced below, accept a space-separated @dfn{ID list} as their argument
3223 specifying one or more elements on which to operate. A list element can be
3224 either a single non-negative number, like @samp{5}, or an ascending range of
3225 such numbers, like @samp{5-7}. A list can consist of any combination of such
3226 elements, even duplicates or overlapping ranges are valid. E.g.@:
3227 @samp{1 4-6 5 4-4} or @samp{1 2 4-7}.
3229 To find out what inferiors exist at any moment, use @w{@code{info
3233 @kindex info inferiors [ @var{id}@dots{} ]
3234 @item info inferiors
3235 Print a list of all inferiors currently being managed by @value{GDBN}.
3236 By default all inferiors are printed, but the ID list @var{id}@dots{} can be
3237 used to limit the display to just the requested inferiors.
3239 @value{GDBN} displays for each inferior (in this order):
3243 the inferior number assigned by @value{GDBN}
3246 the target system's inferior identifier
3249 the target connection the inferior is bound to, including the unique
3250 connection number assigned by @value{GDBN}, and the protocol used by
3254 the name of the executable the inferior is running.
3259 An asterisk @samp{*} preceding the @value{GDBN} inferior number
3260 indicates the current inferior.
3264 @c end table here to get a little more width for example
3267 (@value{GDBP}) info inferiors
3268 Num Description Connection Executable
3269 * 1 process 3401 1 (native) goodbye
3270 2 process 2307 2 (extended-remote host:10000) hello
3273 To get information about the current inferior, use @code{inferior}:
3278 Shows information about the current inferior.
3282 @c end table here to get a little more width for example
3285 (@value{GDBP}) inferior
3286 [Current inferior is 1 [process 3401] (helloworld)]
3289 To find out what open target connections exist at any moment, use
3290 @w{@code{info connections}}:
3293 @kindex info connections [ @var{id}@dots{} ]
3294 @item info connections
3295 Print a list of all open target connections currently being managed by
3296 @value{GDBN}. By default all connections are printed, but the ID list
3297 @var{id}@dots{} can be used to limit the display to just the requested
3300 @value{GDBN} displays for each connection (in this order):
3304 the connection number assigned by @value{GDBN}.
3307 the protocol used by the connection.
3310 a textual description of the protocol used by the connection.
3315 An asterisk @samp{*} preceding the connection number indicates the
3316 connection of the current inferior.
3320 @c end table here to get a little more width for example
3323 (@value{GDBP}) info connections
3324 Num What Description
3325 * 1 extended-remote host:10000 Extended remote serial target in gdb-specific protocol
3326 2 native Native process
3327 3 core Local core dump file
3330 To switch focus between inferiors, use the @code{inferior} command:
3333 @kindex inferior @var{infno}
3334 @item inferior @var{infno}
3335 Make inferior number @var{infno} the current inferior. The argument
3336 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
3337 in the first field of the @samp{info inferiors} display.
3340 @vindex $_inferior@r{, convenience variable}
3341 The debugger convenience variable @samp{$_inferior} contains the
3342 number of the current inferior. You may find this useful in writing
3343 breakpoint conditional expressions, command scripts, and so forth.
3344 @xref{Convenience Vars,, Convenience Variables}, for general
3345 information on convenience variables.
3347 You can get multiple executables into a debugging session via the
3348 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
3349 systems @value{GDBN} can add inferiors to the debug session
3350 automatically by following calls to @code{fork} and @code{exec}. To
3351 remove inferiors from the debugging session use the
3352 @w{@code{remove-inferiors}} command.
3355 @anchor{add_inferior_cli}
3356 @kindex add-inferior
3357 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ] [-no-connection ]
3358 Adds @var{n} inferiors to be run using @var{executable} as the
3359 executable; @var{n} defaults to 1. If no executable is specified,
3360 the inferiors begins empty, with no program. You can still assign or
3361 change the program assigned to the inferior at any time by using the
3362 @code{file} command with the executable name as its argument.
3364 By default, the new inferior begins connected to the same target
3365 connection as the current inferior. For example, if the current
3366 inferior was connected to @code{gdbserver} with @code{target remote},
3367 then the new inferior will be connected to the same @code{gdbserver}
3368 instance. The @samp{-no-connection} option starts the new inferior
3369 with no connection yet. You can then for example use the @code{target
3370 remote} command to connect to some other @code{gdbserver} instance,
3371 use @code{run} to spawn a local program, etc.
3373 @kindex clone-inferior
3374 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
3375 Adds @var{n} inferiors ready to execute the same program as inferior
3376 @var{infno}; @var{n} defaults to 1, and @var{infno} defaults to the
3377 number of the current inferior. This command copies the values of the
3378 @var{args}, @w{@var{inferior-tty}} and @var{cwd} properties from the
3379 current inferior to the new one. It also propagates changes the user
3380 made to environment variables using the @w{@code{set environment}} and
3381 @w{@code{unset environment}} commands. This is a convenient command
3382 when you want to run another instance of the inferior you are debugging.
3385 (@value{GDBP}) info inferiors
3386 Num Description Connection Executable
3387 * 1 process 29964 1 (native) helloworld
3388 (@value{GDBP}) clone-inferior
3391 (@value{GDBP}) info inferiors
3392 Num Description Connection Executable
3393 * 1 process 29964 1 (native) helloworld
3394 2 <null> 1 (native) helloworld
3397 You can now simply switch focus to inferior 2 and run it.
3399 @anchor{remove_inferiors_cli}
3400 @kindex remove-inferiors
3401 @item remove-inferiors @var{infno}@dots{}
3402 Removes the inferior or inferiors @var{infno}@dots{}. It is not
3403 possible to remove an inferior that is running with this command. For
3404 those, use the @code{kill} or @code{detach} command first.
3408 To quit debugging one of the running inferiors that is not the current
3409 inferior, you can either detach from it by using the @w{@code{detach
3410 inferior}} command (allowing it to run independently), or kill it
3411 using the @w{@code{kill inferiors}} command:
3414 @kindex detach inferiors @var{infno}@dots{}
3415 @item detach inferior @var{infno}@dots{}
3416 Detach from the inferior or inferiors identified by @value{GDBN}
3417 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
3418 still stays on the list of inferiors shown by @code{info inferiors},
3419 but its Description will show @samp{<null>}.
3421 @kindex kill inferiors @var{infno}@dots{}
3422 @item kill inferiors @var{infno}@dots{}
3423 Kill the inferior or inferiors identified by @value{GDBN} inferior
3424 number(s) @var{infno}@dots{}. Note that the inferior's entry still
3425 stays on the list of inferiors shown by @code{info inferiors}, but its
3426 Description will show @samp{<null>}.
3429 After the successful completion of a command such as @code{detach},
3430 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
3431 a normal process exit, the inferior is still valid and listed with
3432 @code{info inferiors}, ready to be restarted.
3435 To be notified when inferiors are started or exit under @value{GDBN}'s
3436 control use @w{@code{set print inferior-events}}:
3439 @kindex set print inferior-events
3440 @cindex print messages on inferior start and exit
3441 @item set print inferior-events
3442 @itemx set print inferior-events on
3443 @itemx set print inferior-events off
3444 The @code{set print inferior-events} command allows you to enable or
3445 disable printing of messages when @value{GDBN} notices that new
3446 inferiors have started or that inferiors have exited or have been
3447 detached. By default, these messages will be printed.
3449 @kindex show print inferior-events
3450 @item show print inferior-events
3451 Show whether messages will be printed when @value{GDBN} detects that
3452 inferiors have started, exited or have been detached.
3455 Many commands will work the same with multiple programs as with a
3456 single program: e.g., @code{print myglobal} will simply display the
3457 value of @code{myglobal} in the current inferior.
3460 Occasionally, when debugging @value{GDBN} itself, it may be useful to
3461 get more info about the relationship of inferiors, programs, address
3462 spaces in a debug session. You can do that with the @w{@code{maint
3463 info program-spaces}} command.
3466 @kindex maint info program-spaces
3467 @item maint info program-spaces
3468 Print a list of all program spaces currently being managed by
3471 @value{GDBN} displays for each program space (in this order):
3475 the program space number assigned by @value{GDBN}
3478 the name of the executable loaded into the program space, with e.g.,
3479 the @code{file} command.
3482 the name of the core file loaded into the program space, with e.g.,
3483 the @code{core-file} command.
3488 An asterisk @samp{*} preceding the @value{GDBN} program space number
3489 indicates the current program space.
3491 In addition, below each program space line, @value{GDBN} prints extra
3492 information that isn't suitable to display in tabular form. For
3493 example, the list of inferiors bound to the program space.
3496 (@value{GDBP}) maint info program-spaces
3497 Id Executable Core File
3500 Bound inferiors: ID 1 (process 21561)
3503 Here we can see that no inferior is running the program @code{hello},
3504 while @code{process 21561} is running the program @code{goodbye}. On
3505 some targets, it is possible that multiple inferiors are bound to the
3506 same program space. The most common example is that of debugging both
3507 the parent and child processes of a @code{vfork} call. For example,
3510 (@value{GDBP}) maint info program-spaces
3511 Id Executable Core File
3513 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
3516 Here, both inferior 2 and inferior 1 are running in the same program
3517 space as a result of inferior 1 having executed a @code{vfork} call.
3521 * Inferior-Specific Breakpoints:: Controlling breakpoints
3524 @node Inferior-Specific Breakpoints
3525 @subsection Inferior-Specific Breakpoints
3527 When debugging multiple inferiors, you can choose whether to set
3528 breakpoints for all inferiors, or for a particular inferior.
3531 @cindex breakpoints and inferiors
3532 @cindex inferior-specific breakpoints
3533 @kindex break @dots{} inferior @var{inferior-id}
3534 @item break @var{locspec} inferior @var{inferior-id}
3535 @itemx break @var{locspec} inferior @var{inferior-id} if @dots{}
3536 @var{locspec} specifies a code location or locations in your program.
3537 @xref{Location Specifications}, for details.
3539 Use the qualifier @samp{inferior @var{inferior-id}} with a breakpoint
3540 command to specify that you only want @value{GDBN} to stop when a
3541 particular inferior reaches this breakpoint. The @var{inferior-id}
3542 specifier is one of the inferior identifiers assigned by @value{GDBN},
3543 shown in the first column of the @samp{info inferiors} output.
3545 If you do not specify @samp{inferior @var{inferior-id}} when you set a
3546 breakpoint, the breakpoint applies to @emph{all} inferiors of your
3549 You can use the @code{inferior} qualifier on conditional breakpoints as
3550 well; in this case, place @samp{inferior @var{inferior-id}} before or
3551 after the breakpoint condition, like this:
3554 (@value{GDBP}) break frik.c:13 inferior 2 if bartab > lim
3558 Inferior-specific breakpoints are automatically deleted when the
3559 corresponding inferior is removed from @value{GDBN}. For example:
3562 (@value{GDBP}) remove-inferiors 2
3563 Inferior-specific breakpoint 3 deleted - inferior 2 has been removed.
3566 A breakpoint can't be both inferior-specific and thread-specific
3567 (@pxref{Thread-Specific Breakpoints}), or task-specific (@pxref{Ada
3568 Tasks}); using more than one of the @code{inferior}, @code{thread}, or
3569 @code{task} keywords when creating a breakpoint will give an error.
3572 @section Debugging Programs with Multiple Threads
3574 @cindex threads of execution
3575 @cindex multiple threads
3576 @cindex switching threads
3577 In some operating systems, such as GNU/Linux and Solaris, a single program
3578 may have more than one @dfn{thread} of execution. The precise semantics
3579 of threads differ from one operating system to another, but in general
3580 the threads of a single program are akin to multiple processes---except
3581 that they share one address space (that is, they can all examine and
3582 modify the same variables). On the other hand, each thread has its own
3583 registers and execution stack, and perhaps private memory.
3585 @value{GDBN} provides these facilities for debugging multi-thread
3589 @item automatic notification of new threads
3590 @item @samp{thread @var{thread-id}}, a command to switch among threads
3591 @item @samp{info threads}, a command to inquire about existing threads
3592 @item @samp{thread apply [@var{thread-id-list} | all] @var{args}},
3593 a command to apply a command to a list of threads
3594 @item thread-specific breakpoints
3595 @item @samp{set print thread-events}, which controls printing of
3596 messages on thread start and exit.
3597 @item @samp{set libthread-db-search-path @var{path}}, which lets
3598 the user specify which @code{libthread_db} to use if the default choice
3599 isn't compatible with the program.
3602 @cindex focus of debugging
3603 @cindex current thread
3604 The @value{GDBN} thread debugging facility allows you to observe all
3605 threads while your program runs---but whenever @value{GDBN} takes
3606 control, one thread in particular is always the focus of debugging.
3607 This thread is called the @dfn{current thread}. Debugging commands show
3608 program information from the perspective of the current thread.
3610 @cindex @code{New} @var{systag} message
3611 @cindex thread identifier (system)
3612 @c FIXME-implementors!! It would be more helpful if the [New...] message
3613 @c included GDB's numeric thread handle, so you could just go to that
3614 @c thread without first checking `info threads'.
3615 Whenever @value{GDBN} detects a new thread in your program, it displays
3616 the target system's identification for the thread with a message in the
3617 form @samp{[New @var{systag}]}, where @var{systag} is a thread identifier
3618 whose form varies depending on the particular system. For example, on
3619 @sc{gnu}/Linux, you might see
3622 [New Thread 0x41e02940 (LWP 25582)]
3626 when @value{GDBN} notices a new thread. In contrast, on other systems,
3627 the @var{systag} is simply something like @samp{process 368}, with no
3630 @c FIXME!! (1) Does the [New...] message appear even for the very first
3631 @c thread of a program, or does it only appear for the
3632 @c second---i.e.@: when it becomes obvious we have a multithread
3634 @c (2) *Is* there necessarily a first thread always? Or do some
3635 @c multithread systems permit starting a program with multiple
3636 @c threads ab initio?
3638 @anchor{thread numbers}
3639 @cindex thread number, per inferior
3640 @cindex thread identifier (GDB)
3641 For debugging purposes, @value{GDBN} associates its own thread number
3642 ---always a single integer---with each thread of an inferior. This
3643 number is unique between all threads of an inferior, but not unique
3644 between threads of different inferiors.
3646 @cindex qualified thread ID
3647 You can refer to a given thread in an inferior using the qualified
3648 @var{inferior-num}.@var{thread-num} syntax, also known as
3649 @dfn{qualified thread ID}, with @var{inferior-num} being the inferior
3650 number and @var{thread-num} being the thread number of the given
3651 inferior. For example, thread @code{2.3} refers to thread number 3 of
3652 inferior 2. If you omit @var{inferior-num} (e.g., @code{thread 3}),
3653 then @value{GDBN} infers you're referring to a thread of the current
3656 Until you create a second inferior, @value{GDBN} does not show the
3657 @var{inferior-num} part of thread IDs, even though you can always use
3658 the full @var{inferior-num}.@var{thread-num} form to refer to threads
3659 of inferior 1, the initial inferior.
3661 @anchor{thread ID lists}
3662 @cindex thread ID lists
3663 Some commands accept a space-separated @dfn{thread ID list} as
3664 argument. A list element can be:
3668 A thread ID as shown in the first field of the @samp{info threads}
3669 display, with or without an inferior qualifier. E.g., @samp{2.1} or
3673 A range of thread numbers, again with or without an inferior
3674 qualifier, as in @var{inf}.@var{thr1}-@var{thr2} or
3675 @var{thr1}-@var{thr2}. E.g., @samp{1.2-4} or @samp{2-4}.
3678 All threads of an inferior, specified with a star wildcard, with or
3679 without an inferior qualifier, as in @var{inf}.@code{*} (e.g.,
3680 @samp{1.*}) or @code{*}. The former refers to all threads of the
3681 given inferior, and the latter form without an inferior qualifier
3682 refers to all threads of the current inferior.
3686 For example, if the current inferior is 1, and inferior 7 has one
3687 thread with ID 7.1, the thread list @samp{1 2-3 4.5 6.7-9 7.*}
3688 includes threads 1 to 3 of inferior 1, thread 5 of inferior 4, threads
3689 7 to 9 of inferior 6 and all threads of inferior 7. That is, in
3690 expanded qualified form, the same as @samp{1.1 1.2 1.3 4.5 6.7 6.8 6.9
3694 @anchor{global thread numbers}
3695 @cindex global thread number
3696 @cindex global thread identifier (GDB)
3697 In addition to a @emph{per-inferior} number, each thread is also
3698 assigned a unique @emph{global} number, also known as @dfn{global
3699 thread ID}, a single integer. Unlike the thread number component of
3700 the thread ID, no two threads have the same global ID, even when
3701 you're debugging multiple inferiors.
3703 From @value{GDBN}'s perspective, a process always has at least one
3704 thread. In other words, @value{GDBN} assigns a thread number to the
3705 program's ``main thread'' even if the program is not multi-threaded.
3707 @vindex $_thread@r{, convenience variable}
3708 @vindex $_gthread@r{, convenience variable}
3709 @vindex $_inferior_thread_count@r{, convenience variable}
3710 The debugger convenience variables @samp{$_thread} and
3711 @samp{$_gthread} contain, respectively, the per-inferior thread number
3712 and the global thread number of the current thread. You may find this
3713 useful in writing breakpoint conditional expressions, command scripts,
3714 and so forth. The convenience variable @samp{$_inferior_thread_count}
3715 contains the number of live threads in the current inferior.
3716 @xref{Convenience Vars,, Convenience Variables}, for general
3717 information on convenience variables.
3719 When running in non-stop mode (@pxref{Non-Stop Mode}), where new
3720 threads can be created, and existing threads exit, at any time,
3721 @samp{$_inferior_thread_count} could return a different value each
3722 time it is evaluated.
3724 If @value{GDBN} detects the program is multi-threaded, it augments the
3725 usual message about stopping at a breakpoint with the ID and name of
3726 the thread that hit the breakpoint.
3729 Thread 2 "client" hit Breakpoint 1, send_message () at client.c:68
3732 Likewise when the program receives a signal:
3735 Thread 1 "main" received signal SIGINT, Interrupt.
3739 @anchor{info_threads}
3740 @kindex info threads
3741 @item info threads @r{[}-gid@r{]} @r{[}@var{thread-id-list}@r{]}
3743 Display information about one or more threads. With no arguments
3744 displays information about all threads. You can specify the list of
3745 threads that you want to display using the thread ID list syntax
3746 (@pxref{thread ID lists}).
3748 @value{GDBN} displays for each thread (in this order):
3752 the per-inferior thread number assigned by @value{GDBN}
3755 the global thread number assigned by @value{GDBN}, if the @samp{-gid}
3756 option was specified
3759 the target system's thread identifier (@var{systag})
3762 the thread's name, if one is known. A thread can either be named by
3763 the user (see @code{thread name}, below), or, in some cases, by the
3767 the current stack frame summary for that thread
3771 An asterisk @samp{*} to the left of the @value{GDBN} thread number
3772 indicates the current thread.
3776 @c end table here to get a little more width for example
3779 (@value{GDBP}) info threads
3781 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3782 2 process 35 thread 23 0x34e5 in sigpause ()
3783 3 process 35 thread 27 0x34e5 in sigpause ()
3787 If you're debugging multiple inferiors, @value{GDBN} displays thread
3788 IDs using the qualified @var{inferior-num}.@var{thread-num} format.
3789 Otherwise, only @var{thread-num} is shown.
3791 If you specify the @samp{-gid} option, @value{GDBN} displays a column
3792 indicating each thread's global thread ID:
3795 (@value{GDBP}) info threads
3796 Id GId Target Id Frame
3797 1.1 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3798 1.2 3 process 35 thread 23 0x34e5 in sigpause ()
3799 1.3 4 process 35 thread 27 0x34e5 in sigpause ()
3800 * 2.1 2 process 65 thread 1 main (argc=1, argv=0x7ffffff8)
3803 On Solaris, you can display more information about user threads with a
3804 Solaris-specific command:
3807 @item maint info sol-threads
3808 @kindex maint info sol-threads
3809 @cindex thread info (Solaris)
3810 Display info on Solaris user threads.
3814 @kindex thread @var{thread-id}
3815 @item thread @var{thread-id}
3816 Make thread ID @var{thread-id} the current thread. The command
3817 argument @var{thread-id} is the @value{GDBN} thread ID, as shown in
3818 the first field of the @samp{info threads} display, with or without an
3819 inferior qualifier (e.g., @samp{2.1} or @samp{1}).
3821 @value{GDBN} responds by displaying the system identifier of the
3822 thread you selected, and its current stack frame summary:
3825 (@value{GDBP}) thread 2
3826 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
3827 #0 some_function (ignore=0x0) at example.c:8
3828 8 printf ("hello\n");
3832 As with the @samp{[New @dots{}]} message, the form of the text after
3833 @samp{Switching to} depends on your system's conventions for identifying
3836 @anchor{thread apply all}
3837 @kindex thread apply
3838 @cindex apply command to several threads
3839 @item thread apply [@var{thread-id-list} | all [-ascending]] [@var{flag}]@dots{} @var{command}
3840 The @code{thread apply} command allows you to apply the named
3841 @var{command} to one or more threads. Specify the threads that you
3842 want affected using the thread ID list syntax (@pxref{thread ID
3843 lists}), or specify @code{all} to apply to all threads. To apply a
3844 command to all threads in descending order, type @kbd{thread apply all
3845 @var{command}}. To apply a command to all threads in ascending order,
3846 type @kbd{thread apply all -ascending @var{command}}.
3848 The @var{flag} arguments control what output to produce and how to handle
3849 errors raised when applying @var{command} to a thread. @var{flag}
3850 must start with a @code{-} directly followed by one letter in
3851 @code{qcs}. If several flags are provided, they must be given
3852 individually, such as @code{-c -q}.
3854 By default, @value{GDBN} displays some thread information before the
3855 output produced by @var{command}, and an error raised during the
3856 execution of a @var{command} will abort @code{thread apply}. The
3857 following flags can be used to fine-tune this behavior:
3861 The flag @code{-c}, which stands for @samp{continue}, causes any
3862 errors in @var{command} to be displayed, and the execution of
3863 @code{thread apply} then continues.
3865 The flag @code{-s}, which stands for @samp{silent}, causes any errors
3866 or empty output produced by a @var{command} to be silently ignored.
3867 That is, the execution continues, but the thread information and errors
3870 The flag @code{-q} (@samp{quiet}) disables printing the thread
3874 Flags @code{-c} and @code{-s} cannot be used together.
3877 @cindex apply command to all threads (ignoring errors and empty output)
3878 @item taas [@var{option}]@dots{} @var{command}
3879 Shortcut for @code{thread apply all -s [@var{option}]@dots{} @var{command}}.
3880 Applies @var{command} on all threads, ignoring errors and empty output.
3882 The @code{taas} command accepts the same options as the @code{thread
3883 apply all} command. @xref{thread apply all}.
3886 @cindex apply a command to all frames of all threads (ignoring errors and empty output)
3887 @item tfaas [@var{option}]@dots{} @var{command}
3888 Shortcut for @code{thread apply all -s -- frame apply all -s [@var{option}]@dots{} @var{command}}.
3889 Applies @var{command} on all frames of all threads, ignoring errors
3890 and empty output. Note that the flag @code{-s} is specified twice:
3891 The first @code{-s} ensures that @code{thread apply} only shows the thread
3892 information of the threads for which @code{frame apply} produces
3893 some output. The second @code{-s} is needed to ensure that @code{frame
3894 apply} shows the frame information of a frame only if the
3895 @var{command} successfully produced some output.
3897 It can for example be used to print a local variable or a function
3898 argument without knowing the thread or frame where this variable or argument
3901 (@value{GDBP}) tfaas p some_local_var_i_do_not_remember_where_it_is
3904 The @code{tfaas} command accepts the same options as the @code{frame
3905 apply} command. @xref{Frame Apply,,frame apply}.
3908 @cindex name a thread
3909 @item thread name [@var{name}]
3910 This command assigns a name to the current thread. If no argument is
3911 given, any existing user-specified name is removed. The thread name
3912 appears in the @samp{info threads} display.
3914 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
3915 determine the name of the thread as given by the OS. On these
3916 systems, a name specified with @samp{thread name} will override the
3917 system-give name, and removing the user-specified name will cause
3918 @value{GDBN} to once again display the system-specified name.
3921 @cindex search for a thread
3922 @item thread find [@var{regexp}]
3923 Search for and display thread ids whose name or @var{systag}
3924 matches the supplied regular expression.
3926 As well as being the complement to the @samp{thread name} command,
3927 this command also allows you to identify a thread by its target
3928 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
3932 (@value{GDBP}) thread find 26688
3933 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
3934 (@value{GDBP}) info thread 4
3936 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
3939 @kindex set print thread-events
3940 @cindex print messages on thread start and exit
3941 @item set print thread-events
3942 @itemx set print thread-events on
3943 @itemx set print thread-events off
3944 The @code{set print thread-events} command allows you to enable or
3945 disable printing of messages when @value{GDBN} notices that new threads have
3946 started or that threads have exited. By default, these messages will
3947 be printed if detection of these events is supported by the target.
3948 Note that these messages cannot be disabled on all targets.
3950 @kindex show print thread-events
3951 @item show print thread-events
3952 Show whether messages will be printed when @value{GDBN} detects that threads
3953 have started and exited.
3956 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
3957 more information about how @value{GDBN} behaves when you stop and start
3958 programs with multiple threads.
3960 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
3961 watchpoints in programs with multiple threads.
3963 @anchor{set libthread-db-search-path}
3965 @kindex set libthread-db-search-path
3966 @cindex search path for @code{libthread_db}
3967 @item set libthread-db-search-path @r{[}@var{path}@r{]}
3968 If this variable is set, @var{path} is a colon-separated list of
3969 directories @value{GDBN} will use to search for @code{libthread_db}.
3970 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
3971 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
3972 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
3975 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
3976 @code{libthread_db} library to obtain information about threads in the
3977 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
3978 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
3979 specific thread debugging library loading is enabled
3980 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
3982 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
3983 refers to the default system directories that are
3984 normally searched for loading shared libraries. The @samp{$sdir} entry
3985 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
3986 (@pxref{libthread_db.so.1 file}).
3988 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
3989 refers to the directory from which @code{libpthread}
3990 was loaded in the inferior process.
3992 For any @code{libthread_db} library @value{GDBN} finds in above directories,
3993 @value{GDBN} attempts to initialize it with the current inferior process.
3994 If this initialization fails (which could happen because of a version
3995 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
3996 will unload @code{libthread_db}, and continue with the next directory.
3997 If none of @code{libthread_db} libraries initialize successfully,
3998 @value{GDBN} will issue a warning and thread debugging will be disabled.
4000 Setting @code{libthread-db-search-path} is currently implemented
4001 only on some platforms.
4003 @kindex show libthread-db-search-path
4004 @item show libthread-db-search-path
4005 Display current libthread_db search path.
4007 @kindex set debug libthread-db
4008 @kindex show debug libthread-db
4009 @cindex debugging @code{libthread_db}
4010 @item set debug libthread-db
4011 @itemx show debug libthread-db
4012 Turns on or off display of @code{libthread_db}-related events.
4013 Use @code{1} to enable, @code{0} to disable.
4015 @kindex set debug threads
4016 @kindex show debug threads
4017 @cindex debugging @code{threads}
4018 @item set debug threads @r{[}on@r{|}off@r{]}
4019 @itemx show debug threads
4020 When @samp{on} @value{GDBN} will print additional messages when
4021 threads are created and deleted.
4025 @section Debugging Forks
4027 @cindex fork, debugging programs which call
4028 @cindex multiple processes
4029 @cindex processes, multiple
4030 On most systems, @value{GDBN} has no special support for debugging
4031 programs which create additional processes using the @code{fork}
4032 function. When a program forks, @value{GDBN} will continue to debug the
4033 parent process and the child process will run unimpeded. If you have
4034 set a breakpoint in any code which the child then executes, the child
4035 will get a @code{SIGTRAP} signal which (unless it catches the signal)
4036 will cause it to terminate.
4038 However, if you want to debug the child process there is a workaround
4039 which isn't too painful. Put a call to @code{sleep} in the code which
4040 the child process executes after the fork. It may be useful to sleep
4041 only if a certain environment variable is set, or a certain file exists,
4042 so that the delay need not occur when you don't want to run @value{GDBN}
4043 on the child. While the child is sleeping, use the @code{ps} program to
4044 get its process ID. Then tell @value{GDBN} (a new invocation of
4045 @value{GDBN} if you are also debugging the parent process) to attach to
4046 the child process (@pxref{Attach}). From that point on you can debug
4047 the child process just like any other process which you attached to.
4049 On some systems, @value{GDBN} provides support for debugging programs
4050 that create additional processes using the @code{fork} or @code{vfork}
4051 functions. On @sc{gnu}/Linux platforms, this feature is supported
4052 with kernel version 2.5.46 and later.
4054 The fork debugging commands are supported in native mode and when
4055 connected to @code{gdbserver} in either @code{target remote} mode or
4056 @code{target extended-remote} mode.
4058 By default, when a program forks, @value{GDBN} will continue to debug
4059 the parent process and the child process will run unimpeded.
4061 If you want to follow the child process instead of the parent process,
4062 use the command @w{@code{set follow-fork-mode}}.
4065 @kindex set follow-fork-mode
4066 @item set follow-fork-mode @var{mode}
4067 Set the debugger response to a program call of @code{fork} or
4068 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
4069 process. The @var{mode} argument can be:
4073 The original process is debugged after a fork. The child process runs
4074 unimpeded. This is the default.
4077 The new process is debugged after a fork. The parent process runs
4082 @kindex show follow-fork-mode
4083 @item show follow-fork-mode
4084 Display the current debugger response to a @code{fork} or @code{vfork} call.
4087 @cindex debugging multiple processes
4088 On Linux, if you want to debug both the parent and child processes, use the
4089 command @w{@code{set detach-on-fork}}.
4092 @kindex set detach-on-fork
4093 @item set detach-on-fork @var{mode}
4094 Tells gdb whether to detach one of the processes after a fork, or
4095 retain debugger control over them both.
4099 The child process (or parent process, depending on the value of
4100 @code{follow-fork-mode}) will be detached and allowed to run
4101 independently. This is the default.
4104 Both processes will be held under the control of @value{GDBN}.
4105 One process (child or parent, depending on the value of
4106 @code{follow-fork-mode}) is debugged as usual, while the other
4111 @kindex show detach-on-fork
4112 @item show detach-on-fork
4113 Show whether detach-on-fork mode is on/off.
4116 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
4117 will retain control of all forked processes (including nested forks).
4118 You can list the forked processes under the control of @value{GDBN} by
4119 using the @w{@code{info inferiors}} command, and switch from one fork
4120 to another by using the @code{inferior} command (@pxref{Inferiors Connections and
4121 Programs, ,Debugging Multiple Inferiors Connections and Programs}).
4123 To quit debugging one of the forked processes, you can either detach
4124 from it by using the @w{@code{detach inferiors}} command (allowing it
4125 to run independently), or kill it using the @w{@code{kill inferiors}}
4126 command. @xref{Inferiors Connections and Programs, ,Debugging
4127 Multiple Inferiors Connections and Programs}.
4129 If you ask to debug a child process and a @code{vfork} is followed by an
4130 @code{exec}, @value{GDBN} executes the new target up to the first
4131 breakpoint in the new target. If you have a breakpoint set on
4132 @code{main} in your original program, the breakpoint will also be set on
4133 the child process's @code{main}.
4135 On some systems, when a child process is spawned by @code{vfork}, you
4136 cannot debug the child or parent until an @code{exec} call completes.
4138 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
4139 call executes, the new target restarts. To restart the parent
4140 process, use the @code{file} command with the parent executable name
4141 as its argument. By default, after an @code{exec} call executes,
4142 @value{GDBN} discards the symbols of the previous executable image.
4143 You can change this behaviour with the @w{@code{set follow-exec-mode}}
4147 @kindex set follow-exec-mode
4148 @item set follow-exec-mode @var{mode}
4150 Set debugger response to a program call of @code{exec}. An
4151 @code{exec} call replaces the program image of a process.
4153 @code{follow-exec-mode} can be:
4157 @value{GDBN} creates a new inferior and rebinds the process to this
4158 new inferior. The program the process was running before the
4159 @code{exec} call can be restarted afterwards by restarting the
4165 (@value{GDBP}) info inferiors
4166 (@value{GDBP}) info inferior
4167 Id Description Executable
4170 process 12020 is executing new program: prog2
4171 Program exited normally.
4172 (@value{GDBP}) info inferiors
4173 Id Description Executable
4179 @value{GDBN} keeps the process bound to the same inferior. The new
4180 executable image replaces the previous executable loaded in the
4181 inferior. Restarting the inferior after the @code{exec} call, with
4182 e.g., the @code{run} command, restarts the executable the process was
4183 running after the @code{exec} call. This is the default mode.
4188 (@value{GDBP}) info inferiors
4189 Id Description Executable
4192 process 12020 is executing new program: prog2
4193 Program exited normally.
4194 (@value{GDBP}) info inferiors
4195 Id Description Executable
4202 @code{follow-exec-mode} is supported in native mode and
4203 @code{target extended-remote} mode.
4205 You can use the @code{catch} command to make @value{GDBN} stop whenever
4206 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
4207 Catchpoints, ,Setting Catchpoints}.
4209 @node Checkpoint/Restart
4210 @section Setting a @emph{Bookmark} to Return to Later
4215 @cindex snapshot of a process
4216 @cindex rewind program state
4218 On certain operating systems@footnote{Currently, only
4219 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
4220 program's state, called a @dfn{checkpoint}, and come back to it
4223 Returning to a checkpoint effectively undoes everything that has
4224 happened in the program since the @code{checkpoint} was saved. This
4225 includes changes in memory, registers, and even (within some limits)
4226 system state. Effectively, it is like going back in time to the
4227 moment when the checkpoint was saved.
4229 Thus, if you're stepping thru a program and you think you're
4230 getting close to the point where things go wrong, you can save
4231 a checkpoint. Then, if you accidentally go too far and miss
4232 the critical statement, instead of having to restart your program
4233 from the beginning, you can just go back to the checkpoint and
4234 start again from there.
4236 This can be especially useful if it takes a lot of time or
4237 steps to reach the point where you think the bug occurs.
4239 To use the @code{checkpoint}/@code{restart} method of debugging:
4244 Save a snapshot of the debugged program's current execution state.
4245 The @code{checkpoint} command takes no arguments, but each checkpoint
4246 is assigned a small integer id, similar to a breakpoint id.
4248 @kindex info checkpoints
4249 @item info checkpoints
4250 List the checkpoints that have been saved in the current debugging
4251 session. For each checkpoint, the following information will be
4258 @item Source line, or label
4261 @kindex restart @var{checkpoint-id}
4262 @item restart @var{checkpoint-id}
4263 Restore the program state that was saved as checkpoint number
4264 @var{checkpoint-id}. All program variables, registers, stack frames
4265 etc.@: will be returned to the values that they had when the checkpoint
4266 was saved. In essence, gdb will ``wind back the clock'' to the point
4267 in time when the checkpoint was saved.
4269 Note that breakpoints, @value{GDBN} variables, command history etc.
4270 are not affected by restoring a checkpoint. In general, a checkpoint
4271 only restores things that reside in the program being debugged, not in
4274 @kindex delete checkpoint @var{checkpoint-id}
4275 @item delete checkpoint @var{checkpoint-id}
4276 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
4280 Returning to a previously saved checkpoint will restore the user state
4281 of the program being debugged, plus a significant subset of the system
4282 (OS) state, including file pointers. It won't ``un-write'' data from
4283 a file, but it will rewind the file pointer to the previous location,
4284 so that the previously written data can be overwritten. For files
4285 opened in read mode, the pointer will also be restored so that the
4286 previously read data can be read again.
4288 Of course, characters that have been sent to a printer (or other
4289 external device) cannot be ``snatched back'', and characters received
4290 from eg.@: a serial device can be removed from internal program buffers,
4291 but they cannot be ``pushed back'' into the serial pipeline, ready to
4292 be received again. Similarly, the actual contents of files that have
4293 been changed cannot be restored (at this time).
4295 However, within those constraints, you actually can ``rewind'' your
4296 program to a previously saved point in time, and begin debugging it
4297 again --- and you can change the course of events so as to debug a
4298 different execution path this time.
4300 @cindex checkpoints and process id
4301 Finally, there is one bit of internal program state that will be
4302 different when you return to a checkpoint --- the program's process
4303 id. Each checkpoint will have a unique process id (or @var{pid}),
4304 and each will be different from the program's original @var{pid}.
4305 If your program has saved a local copy of its process id, this could
4306 potentially pose a problem.
4308 @subsection A Non-obvious Benefit of Using Checkpoints
4310 On some systems such as @sc{gnu}/Linux, address space randomization
4311 is performed on new processes for security reasons. This makes it
4312 difficult or impossible to set a breakpoint, or watchpoint, on an
4313 absolute address if you have to restart the program, since the
4314 absolute location of a symbol will change from one execution to the
4317 A checkpoint, however, is an @emph{identical} copy of a process.
4318 Therefore if you create a checkpoint at (eg.@:) the start of main,
4319 and simply return to that checkpoint instead of restarting the
4320 process, you can avoid the effects of address randomization and
4321 your symbols will all stay in the same place.
4324 @chapter Stopping and Continuing
4326 The principal purposes of using a debugger are so that you can stop your
4327 program before it terminates; or so that, if your program runs into
4328 trouble, you can investigate and find out why.
4330 Inside @value{GDBN}, your program may stop for any of several reasons,
4331 such as a signal, a breakpoint, or reaching a new line after a
4332 @value{GDBN} command such as @code{step}. You may then examine and
4333 change variables, set new breakpoints or remove old ones, and then
4334 continue execution. Usually, the messages shown by @value{GDBN} provide
4335 ample explanation of the status of your program---but you can also
4336 explicitly request this information at any time.
4339 @kindex info program
4341 Display information about the status of your program: whether it is
4342 running or not, what process it is, and why it stopped.
4346 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
4347 * Continuing and Stepping:: Resuming execution
4348 * Skipping Over Functions and Files::
4349 Skipping over functions and files
4351 * Thread Stops:: Stopping and starting multi-thread programs
4355 @section Breakpoints, Watchpoints, and Catchpoints
4358 A @dfn{breakpoint} makes your program stop whenever a certain point in
4359 the program is reached. For each breakpoint, you can add conditions to
4360 control in finer detail whether your program stops. You can set
4361 breakpoints with the @code{break} command and its variants (@pxref{Set
4362 Breaks, ,Setting Breakpoints}), to specify the place where your program
4363 should stop by line number, function name or exact address in the
4366 On some systems, you can set breakpoints in shared libraries before
4367 the executable is run.
4370 @cindex data breakpoints
4371 @cindex memory tracing
4372 @cindex breakpoint on memory address
4373 @cindex breakpoint on variable modification
4374 A @dfn{watchpoint} is a special breakpoint that stops your program
4375 when the value of an expression changes. The expression may be a value
4376 of a variable, or it could involve values of one or more variables
4377 combined by operators, such as @samp{a + b}. This is sometimes called
4378 @dfn{data breakpoints}. You must use a different command to set
4379 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
4380 from that, you can manage a watchpoint like any other breakpoint: you
4381 enable, disable, and delete both breakpoints and watchpoints using the
4384 You can arrange to have values from your program displayed automatically
4385 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
4389 @cindex breakpoint on events
4390 A @dfn{catchpoint} is another special breakpoint that stops your program
4391 when a certain kind of event occurs, such as the throwing of a C@t{++}
4392 exception or the loading of a library. As with watchpoints, you use a
4393 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
4394 Catchpoints}), but aside from that, you can manage a catchpoint like any
4395 other breakpoint. (To stop when your program receives a signal, use the
4396 @code{handle} command; see @ref{Signals, ,Signals}.)
4398 @cindex breakpoint numbers
4399 @cindex numbers for breakpoints
4400 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
4401 catchpoint when you create it; these numbers are successive integers
4402 starting with one. In many of the commands for controlling various
4403 features of breakpoints you use the breakpoint number to say which
4404 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
4405 @dfn{disabled}; if disabled, it has no effect on your program until you
4408 @cindex breakpoint ranges
4409 @cindex breakpoint lists
4410 @cindex ranges of breakpoints
4411 @cindex lists of breakpoints
4412 Some @value{GDBN} commands accept a space-separated list of breakpoints
4413 on which to operate. A list element can be either a single breakpoint number,
4414 like @samp{5}, or a range of such numbers, like @samp{5-7}.
4415 When a breakpoint list is given to a command, all breakpoints in that list
4419 * Set Breaks:: Setting breakpoints
4420 * Set Watchpoints:: Setting watchpoints
4421 * Set Catchpoints:: Setting catchpoints
4422 * Delete Breaks:: Deleting breakpoints
4423 * Disabling:: Disabling breakpoints
4424 * Conditions:: Break conditions
4425 * Break Commands:: Breakpoint command lists
4426 * Dynamic Printf:: Dynamic printf
4427 * Save Breakpoints:: How to save breakpoints in a file
4428 * Static Probe Points:: Listing static probe points
4429 * Error in Breakpoints:: ``Cannot insert breakpoints''
4430 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
4434 @subsection Setting Breakpoints
4436 @c FIXME LMB what does GDB do if no code on line of breakpt?
4437 @c consider in particular declaration with/without initialization.
4439 @c FIXME 2 is there stuff on this already? break at fun start, already init?
4442 @kindex b @r{(@code{break})}
4443 @vindex $bpnum@r{, convenience variable}
4444 @cindex latest breakpoint
4445 Breakpoints are set with the @code{break} command (abbreviated
4446 @code{b}). The debugger convenience variable @samp{$bpnum} records the
4447 number of the breakpoint you've set most recently:
4450 Breakpoint 1 at 0x11c6: file zeoes.c, line 24.
4455 A breakpoint may be mapped to multiple code locations for example with
4456 inlined functions, Ada generics, C@t{++} templates or overloaded function names.
4457 @value{GDBN} then indicates the number of code locations in the breakpoint
4461 Breakpoint 2 at 0x1179: some_func. (3 locations)
4467 @vindex $_hit_bpnum@r{, convenience variable}
4468 @vindex $_hit_locno@r{, convenience variable}
4469 When your program stops on a breakpoint, the convenience variables
4470 @samp{$_hit_bpnum} and @samp{$_hit_locno} are respectively set to the number of
4471 the encountered breakpoint and the number of the breakpoint's code location:
4473 Thread 1 "zeoes" hit Breakpoint 2.1, some_func () at zeoes.c:8
4474 8 printf("some func\n");
4482 Note that @samp{$_hit_bpnum} and @samp{$bpnum} are not equivalent:
4483 @samp{$_hit_bpnum} is set to the breakpoint number @b{last hit}, while
4484 @samp{$bpnum} is set to the breakpoint number @b{last set}.
4487 If the encountered breakpoint has only one code location, @samp{$_hit_locno}
4490 Breakpoint 1, main (argc=1, argv=0x7fffffffe018) at zeoes.c:24
4499 The @samp{$_hit_bpnum} and @samp{$_hit_locno} variables can typically be used
4500 in a breakpoint command list.
4501 (@pxref{Break Commands, ,Breakpoint Command Lists}). For example, as
4502 part of the breakpoint command list, you can disable completely the
4503 encountered breakpoint using @kbd{disable $_hit_bpnum} or disable the
4504 specific encountered breakpoint location using
4505 @kbd{disable $_hit_bpnum.$_hit_locno}.
4506 If a breakpoint has only one location, @samp{$_hit_locno} is set to 1
4507 and the commands @kbd{disable $_hit_bpnum} and
4508 @kbd{disable $_hit_bpnum.$_hit_locno} both disable the breakpoint.
4510 You can also define aliases to easily disable the last hit location or
4511 last hit breakpoint:
4513 (gdb) alias lld = disable $_hit_bpnum.$_hit_locno
4514 (gdb) alias lbd = disable $_hit_bpnum
4518 @item break @var{locspec}
4519 Set a breakpoint at all the code locations in your program that result
4520 from resolving the given @var{locspec}. @var{locspec} can specify a
4521 function name, a line number, an address of an instruction, and more.
4522 @xref{Location Specifications}, for the various forms of
4523 @var{locspec}. The breakpoint will stop your program just before it
4524 executes the instruction at the address of any of the breakpoint's
4527 When using source languages that permit overloading of symbols, such
4528 as C@t{++}, a function name may refer to more than one symbol, and
4529 thus more than one place to break. @xref{Ambiguous
4530 Expressions,,Ambiguous Expressions}, for a discussion of that
4533 It is also possible to insert a breakpoint that will stop the program
4534 only if a specific thread (@pxref{Thread-Specific Breakpoints}),
4535 specific inferior (@pxref{Inferior-Specific Breakpoints}), or a
4536 specific task (@pxref{Ada Tasks}) hits that breakpoint.
4539 When called without any arguments, @code{break} sets a breakpoint at
4540 the next instruction to be executed in the selected stack frame
4541 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
4542 innermost, this makes your program stop as soon as control
4543 returns to that frame. This is similar to the effect of a
4544 @code{finish} command in the frame inside the selected frame---except
4545 that @code{finish} does not leave an active breakpoint. If you use
4546 @code{break} without an argument in the innermost frame, @value{GDBN} stops
4547 the next time it reaches the current location; this may be useful
4550 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
4551 least one instruction has been executed. If it did not do this, you
4552 would be unable to proceed past a breakpoint without first disabling the
4553 breakpoint. This rule applies whether or not the breakpoint already
4554 existed when your program stopped.
4556 @item break @dots{} if @var{cond}
4557 Set a breakpoint with condition @var{cond}; evaluate the expression
4558 @var{cond} each time the breakpoint is reached, and stop only if the
4559 value is nonzero---that is, if @var{cond} evaluates as true.
4560 @samp{@dots{}} stands for one of the possible arguments described
4561 above (or no argument) specifying where to break. @xref{Conditions,
4562 ,Break Conditions}, for more information on breakpoint conditions.
4564 The breakpoint may be mapped to multiple locations. If the breakpoint
4565 condition @var{cond} is invalid at some but not all of the locations,
4566 the locations for which the condition is invalid are disabled. For
4567 example, @value{GDBN} reports below that two of the three locations
4571 (@value{GDBP}) break func if a == 10
4572 warning: failed to validate condition at location 0x11ce, disabling:
4573 No symbol "a" in current context.
4574 warning: failed to validate condition at location 0x11b6, disabling:
4575 No symbol "a" in current context.
4576 Breakpoint 1 at 0x11b6: func. (3 locations)
4579 Locations that are disabled because of the condition are denoted by an
4580 uppercase @code{N} in the output of the @code{info breakpoints}
4584 (@value{GDBP}) info breakpoints
4585 Num Type Disp Enb Address What
4586 1 breakpoint keep y <MULTIPLE>
4587 stop only if a == 10
4588 1.1 N* 0x00000000000011b6 in ...
4589 1.2 y 0x00000000000011c2 in ...
4590 1.3 N* 0x00000000000011ce in ...
4591 (*): Breakpoint condition is invalid at this location.
4594 If the breakpoint condition @var{cond} is invalid in the context of
4595 @emph{all} the locations of the breakpoint, @value{GDBN} refuses to
4596 define the breakpoint. For example, if variable @code{foo} is an
4600 (@value{GDBP}) break func if foo
4601 No symbol "foo" in current context.
4604 @item break @dots{} -force-condition if @var{cond}
4605 There may be cases where the condition @var{cond} is invalid at all
4606 the current locations, but the user knows that it will be valid at a
4607 future location; for example, because of a library load. In such
4608 cases, by using the @code{-force-condition} keyword before @samp{if},
4609 @value{GDBN} can be forced to define the breakpoint with the given
4610 condition expression instead of refusing it.
4613 (@value{GDBP}) break func -force-condition if foo
4614 warning: failed to validate condition at location 1, disabling:
4615 No symbol "foo" in current context.
4616 warning: failed to validate condition at location 2, disabling:
4617 No symbol "foo" in current context.
4618 warning: failed to validate condition at location 3, disabling:
4619 No symbol "foo" in current context.
4620 Breakpoint 1 at 0x1158: test.c:18. (3 locations)
4623 This causes all the present locations where the breakpoint would
4624 otherwise be inserted, to be disabled, as seen in the example above.
4625 However, if there exist locations at which the condition is valid, the
4626 @code{-force-condition} keyword has no effect.
4629 @item tbreak @var{args}
4630 Set a breakpoint enabled only for one stop. The @var{args} are the
4631 same as for the @code{break} command, and the breakpoint is set in the same
4632 way, but the breakpoint is automatically deleted after the first time your
4633 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
4636 @cindex hardware breakpoints
4637 @item hbreak @var{args}
4638 Set a hardware-assisted breakpoint. The @var{args} are the same as for the
4639 @code{break} command and the breakpoint is set in the same way, but the
4640 breakpoint requires hardware support and some target hardware may not
4641 have this support. The main purpose of this is EPROM/ROM code
4642 debugging, so you can set a breakpoint at an instruction without
4643 changing the instruction. This can be used with the new trap-generation
4644 provided by SPARClite DSU and most x86-based targets. These targets
4645 will generate traps when a program accesses some data or instruction
4646 address that is assigned to the debug registers. However the hardware
4647 breakpoint registers can take a limited number of breakpoints. For
4648 example, on the DSU, only two data breakpoints can be set at a time, and
4649 @value{GDBN} will reject this command if more than two are used. Delete
4650 or disable unused hardware breakpoints before setting new ones
4651 (@pxref{Disabling, ,Disabling Breakpoints}).
4652 @xref{Conditions, ,Break Conditions}.
4653 For remote targets, you can restrict the number of hardware
4654 breakpoints @value{GDBN} will use, see @ref{set remote
4655 hardware-breakpoint-limit}.
4658 @item thbreak @var{args}
4659 Set a hardware-assisted breakpoint enabled only for one stop. The @var{args}
4660 are the same as for the @code{hbreak} command and the breakpoint is set in
4661 the same way. However, like the @code{tbreak} command,
4662 the breakpoint is automatically deleted after the
4663 first time your program stops there. Also, like the @code{hbreak}
4664 command, the breakpoint requires hardware support and some target hardware
4665 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
4666 See also @ref{Conditions, ,Break Conditions}.
4669 @cindex regular expression
4670 @cindex breakpoints at functions matching a regexp
4671 @cindex set breakpoints in many functions
4672 @item rbreak @var{regex}
4673 Set breakpoints on all functions matching the regular expression
4674 @var{regex}. This command sets an unconditional breakpoint on all
4675 matches, printing a list of all breakpoints it set. Once these
4676 breakpoints are set, they are treated just like the breakpoints set with
4677 the @code{break} command. You can delete them, disable them, or make
4678 them conditional the same way as any other breakpoint.
4680 In programs using different languages, @value{GDBN} chooses the syntax
4681 to print the list of all breakpoints it sets according to the
4682 @samp{set language} value: using @samp{set language auto}
4683 (see @ref{Automatically, ,Set Language Automatically}) means to use the
4684 language of the breakpoint's function, other values mean to use
4685 the manually specified language (see @ref{Manually, ,Set Language Manually}).
4687 The syntax of the regular expression is the standard one used with tools
4688 like @file{grep}. Note that this is different from the syntax used by
4689 shells, so for instance @code{foo*} matches all functions that include
4690 an @code{fo} followed by zero or more @code{o}s. There is an implicit
4691 @code{.*} leading and trailing the regular expression you supply, so to
4692 match only functions that begin with @code{foo}, use @code{^foo}.
4694 @cindex non-member C@t{++} functions, set breakpoint in
4695 When debugging C@t{++} programs, @code{rbreak} is useful for setting
4696 breakpoints on overloaded functions that are not members of any special
4699 @cindex set breakpoints on all functions
4700 The @code{rbreak} command can be used to set breakpoints in
4701 @strong{all} the functions in a program, like this:
4704 (@value{GDBP}) rbreak .
4707 @item rbreak @var{file}:@var{regex}
4708 If @code{rbreak} is called with a filename qualification, it limits
4709 the search for functions matching the given regular expression to the
4710 specified @var{file}. This can be used, for example, to set breakpoints on
4711 every function in a given file:
4714 (@value{GDBP}) rbreak file.c:.
4717 The colon separating the filename qualifier from the regex may
4718 optionally be surrounded by spaces.
4720 @kindex info breakpoints
4721 @cindex @code{$_} and @code{info breakpoints}
4722 @item info breakpoints @r{[}@var{list}@dots{}@r{]}
4723 @itemx info break @r{[}@var{list}@dots{}@r{]}
4724 Print a table of all breakpoints, watchpoints, and catchpoints set and
4725 not deleted. Optional argument @var{n} means print information only
4726 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
4727 For each breakpoint, following columns are printed:
4730 @item Breakpoint Numbers
4732 Breakpoint, watchpoint, or catchpoint.
4734 Whether the breakpoint is marked to be disabled or deleted when hit.
4735 @item Enabled or Disabled
4736 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
4737 that are not enabled.
4739 Where the breakpoint is in your program, as a memory address. For a
4740 pending breakpoint whose address is not yet known, this field will
4741 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
4742 library that has the symbol or line referred by breakpoint is loaded.
4743 See below for details. A breakpoint with several locations will
4744 have @samp{<MULTIPLE>} in this field---see below for details.
4746 Where the breakpoint is in the source for your program, as a file and
4747 line number. For a pending breakpoint, the original string passed to
4748 the breakpoint command will be listed as it cannot be resolved until
4749 the appropriate shared library is loaded in the future.
4753 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
4754 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
4755 @value{GDBN} on the host's side. If it is ``target'', then the condition
4756 is evaluated by the target. The @code{info break} command shows
4757 the condition on the line following the affected breakpoint, together with
4758 its condition evaluation mode in between parentheses.
4760 Breakpoint commands, if any, are listed after that. A pending breakpoint is
4761 allowed to have a condition specified for it. The condition is not parsed for
4762 validity until a shared library is loaded that allows the pending
4763 breakpoint to resolve to a valid location.
4766 @code{info break} with a breakpoint
4767 number @var{n} as argument lists only that breakpoint. The
4768 convenience variable @code{$_} and the default examining-address for
4769 the @code{x} command are set to the address of the last breakpoint
4770 listed (@pxref{Memory, ,Examining Memory}).
4773 @code{info break} displays a count of the number of times the breakpoint
4774 has been hit. This is especially useful in conjunction with the
4775 @code{ignore} command. You can ignore a large number of breakpoint
4776 hits, look at the breakpoint info to see how many times the breakpoint
4777 was hit, and then run again, ignoring one less than that number. This
4778 will get you quickly to the last hit of that breakpoint.
4781 For a breakpoints with an enable count (xref) greater than 1,
4782 @code{info break} also displays that count.
4786 @value{GDBN} allows you to set any number of breakpoints at the same place in
4787 your program. There is nothing silly or meaningless about this. When
4788 the breakpoints are conditional, this is even useful
4789 (@pxref{Conditions, ,Break Conditions}).
4791 @cindex multiple locations, breakpoints
4792 @cindex breakpoints, multiple locations
4793 It is possible that a single logical breakpoint is set at several code
4794 locations in your program. @xref{Location Specifications}, for
4797 A breakpoint with multiple code locations is displayed in the
4798 breakpoint table using several rows---one header row, followed by one
4799 row for each code location. The header row has @samp{<MULTIPLE>} in
4800 the address column. Each code location row contains the actual
4801 address, source file, source line and function of its code location.
4802 The number column for a code location is of the form
4803 @var{breakpoint-number}.@var{location-number}.
4808 Num Type Disp Enb Address What
4809 1 breakpoint keep y <MULTIPLE>
4811 breakpoint already hit 1 time
4812 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
4813 1.2 y 0x080486ca in void foo<double>() at t.cc:8
4816 You cannot delete the individual locations from a breakpoint. However,
4817 each location can be individually enabled or disabled by passing
4818 @var{breakpoint-number}.@var{location-number} as argument to the
4819 @code{enable} and @code{disable} commands. It's also possible to
4820 @code{enable} and @code{disable} a range of @var{location-number}
4821 locations using a @var{breakpoint-number} and two @var{location-number}s,
4822 in increasing order, separated by a hyphen, like
4823 @kbd{@var{breakpoint-number}.@var{location-number1}-@var{location-number2}},
4824 in which case @value{GDBN} acts on all the locations in the range (inclusive).
4825 Disabling or enabling the parent breakpoint (@pxref{Disabling}) affects
4826 all of the locations that belong to that breakpoint.
4828 Locations that are enabled while their parent breakpoint is disabled
4829 won't trigger a break, and are denoted by @code{y-} in the @code{Enb}
4830 column. For example:
4833 (@value{GDBP}) info breakpoints
4834 Num Type Disp Enb Address What
4835 1 breakpoint keep n <MULTIPLE>
4836 1.1 y- 0x00000000000011b6 in ...
4837 1.2 y- 0x00000000000011c2 in ...
4838 1.3 n 0x00000000000011ce in ...
4841 @cindex pending breakpoints
4842 It's quite common to have a breakpoint inside a shared library.
4843 Shared libraries can be loaded and unloaded explicitly,
4844 and possibly repeatedly, as the program is executed. To support
4845 this use case, @value{GDBN} updates breakpoint locations whenever
4846 any shared library is loaded or unloaded. Typically, you would
4847 set a breakpoint in a shared library at the beginning of your
4848 debugging session, when the library is not loaded, and when the
4849 symbols from the library are not available. When you try to set
4850 breakpoint, @value{GDBN} will ask you if you want to set
4851 a so called @dfn{pending breakpoint}---breakpoint whose address
4852 is not yet resolved.
4854 After the program is run, whenever a new shared library is loaded,
4855 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
4856 shared library contains the symbol or line referred to by some
4857 pending breakpoint, that breakpoint is resolved and becomes an
4858 ordinary breakpoint. When a library is unloaded, all breakpoints
4859 that refer to its symbols or source lines become pending again.
4861 This logic works for breakpoints with multiple locations, too. For
4862 example, if you have a breakpoint in a C@t{++} template function, and
4863 a newly loaded shared library has an instantiation of that template,
4864 a new location is added to the list of locations for the breakpoint.
4866 Except for having unresolved address, pending breakpoints do not
4867 differ from regular breakpoints. You can set conditions or commands,
4868 enable and disable them and perform other breakpoint operations.
4870 @value{GDBN} provides some additional commands for controlling what
4871 happens when the @samp{break} command cannot resolve the location spec
4872 to any code location in your program (@pxref{Location
4875 @kindex set breakpoint pending
4876 @kindex show breakpoint pending
4878 @item set breakpoint pending auto
4879 This is the default behavior. When @value{GDBN} cannot resolve the
4880 location spec, it queries you whether a pending breakpoint should be
4883 @item set breakpoint pending on
4884 This indicates that when @value{GDBN} cannot resolve the location
4885 spec, it should create a pending breakpoint without confirmation.
4887 @item set breakpoint pending off
4888 This indicates that pending breakpoints are not to be created. If
4889 @value{GDBN} cannot resolve the location spec, it aborts the
4890 breakpoint creation with an error. This setting does not affect any
4891 pending breakpoints previously created.
4893 @item show breakpoint pending
4894 Show the current behavior setting for creating pending breakpoints.
4897 The settings above only affect the @code{break} command and its
4898 variants. Once a breakpoint is set, it will be automatically updated
4899 as shared libraries are loaded and unloaded.
4901 @cindex automatic hardware breakpoints
4902 For some targets, @value{GDBN} can automatically decide if hardware or
4903 software breakpoints should be used, depending on whether the
4904 breakpoint address is read-only or read-write. This applies to
4905 breakpoints set with the @code{break} command as well as to internal
4906 breakpoints set by commands like @code{next} and @code{finish}. For
4907 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
4910 You can control this automatic behaviour with the following commands:
4912 @kindex set breakpoint auto-hw
4913 @kindex show breakpoint auto-hw
4915 @item set breakpoint auto-hw on
4916 This is the default behavior. When @value{GDBN} sets a breakpoint, it
4917 will try to use the target memory map to decide if software or hardware
4918 breakpoint must be used.
4920 @item set breakpoint auto-hw off
4921 This indicates @value{GDBN} should not automatically select breakpoint
4922 type. If the target provides a memory map, @value{GDBN} will warn when
4923 trying to set software breakpoint at a read-only address.
4926 @value{GDBN} normally implements breakpoints by replacing the program code
4927 at the breakpoint address with a special instruction, which, when
4928 executed, given control to the debugger. By default, the program
4929 code is so modified only when the program is resumed. As soon as
4930 the program stops, @value{GDBN} restores the original instructions. This
4931 behaviour guards against leaving breakpoints inserted in the
4932 target should gdb abrubptly disconnect. However, with slow remote
4933 targets, inserting and removing breakpoint can reduce the performance.
4934 This behavior can be controlled with the following commands::
4936 @kindex set breakpoint always-inserted
4937 @kindex show breakpoint always-inserted
4939 @item set breakpoint always-inserted off
4940 All breakpoints, including newly added by the user, are inserted in
4941 the target only when the target is resumed. All breakpoints are
4942 removed from the target when it stops. This is the default mode.
4944 @item set breakpoint always-inserted on
4945 Causes all breakpoints to be inserted in the target at all times. If
4946 the user adds a new breakpoint, or changes an existing breakpoint, the
4947 breakpoints in the target are updated immediately. A breakpoint is
4948 removed from the target only when breakpoint itself is deleted.
4951 @value{GDBN} handles conditional breakpoints by evaluating these conditions
4952 when a breakpoint breaks. If the condition is true, then the process being
4953 debugged stops, otherwise the process is resumed.
4955 If the target supports evaluating conditions on its end, @value{GDBN} may
4956 download the breakpoint, together with its conditions, to it.
4958 This feature can be controlled via the following commands:
4960 @kindex set breakpoint condition-evaluation
4961 @kindex show breakpoint condition-evaluation
4963 @item set breakpoint condition-evaluation host
4964 This option commands @value{GDBN} to evaluate the breakpoint
4965 conditions on the host's side. Unconditional breakpoints are sent to
4966 the target which in turn receives the triggers and reports them back to GDB
4967 for condition evaluation. This is the standard evaluation mode.
4969 @item set breakpoint condition-evaluation target
4970 This option commands @value{GDBN} to download breakpoint conditions
4971 to the target at the moment of their insertion. The target
4972 is responsible for evaluating the conditional expression and reporting
4973 breakpoint stop events back to @value{GDBN} whenever the condition
4974 is true. Due to limitations of target-side evaluation, some conditions
4975 cannot be evaluated there, e.g., conditions that depend on local data
4976 that is only known to the host. Examples include
4977 conditional expressions involving convenience variables, complex types
4978 that cannot be handled by the agent expression parser and expressions
4979 that are too long to be sent over to the target, specially when the
4980 target is a remote system. In these cases, the conditions will be
4981 evaluated by @value{GDBN}.
4983 @item set breakpoint condition-evaluation auto
4984 This is the default mode. If the target supports evaluating breakpoint
4985 conditions on its end, @value{GDBN} will download breakpoint conditions to
4986 the target (limitations mentioned previously apply). If the target does
4987 not support breakpoint condition evaluation, then @value{GDBN} will fallback
4988 to evaluating all these conditions on the host's side.
4992 @cindex negative breakpoint numbers
4993 @cindex internal @value{GDBN} breakpoints
4994 @value{GDBN} itself sometimes sets breakpoints in your program for
4995 special purposes, such as proper handling of @code{longjmp} (in C
4996 programs). These internal breakpoints are assigned negative numbers,
4997 starting with @code{-1}; @samp{info breakpoints} does not display them.
4998 You can see these breakpoints with the @value{GDBN} maintenance command
4999 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
5002 @node Set Watchpoints
5003 @subsection Setting Watchpoints
5005 @cindex setting watchpoints
5006 You can use a watchpoint to stop execution whenever the value of an
5007 expression changes, without having to predict a particular place where
5008 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
5009 The expression may be as simple as the value of a single variable, or
5010 as complex as many variables combined by operators. Examples include:
5014 A reference to the value of a single variable.
5017 An address cast to an appropriate data type. For example,
5018 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
5019 address (assuming an @code{int} occupies 4 bytes).
5022 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
5023 expression can use any operators valid in the program's native
5024 language (@pxref{Languages}).
5027 You can set a watchpoint on an expression even if the expression can
5028 not be evaluated yet. For instance, you can set a watchpoint on
5029 @samp{*global_ptr} before @samp{global_ptr} is initialized.
5030 @value{GDBN} will stop when your program sets @samp{global_ptr} and
5031 the expression produces a valid value. If the expression becomes
5032 valid in some other way than changing a variable (e.g.@: if the memory
5033 pointed to by @samp{*global_ptr} becomes readable as the result of a
5034 @code{malloc} call), @value{GDBN} may not stop until the next time
5035 the expression changes.
5037 @cindex software watchpoints
5038 @cindex hardware watchpoints
5039 Depending on your system, watchpoints may be implemented in software or
5040 hardware. @value{GDBN} does software watchpointing by single-stepping your
5041 program and testing the variable's value each time, which is hundreds of
5042 times slower than normal execution. (But this may still be worth it, to
5043 catch errors where you have no clue what part of your program is the
5046 On some systems, such as most PowerPC or x86-based targets,
5047 @value{GDBN} includes support for hardware watchpoints, which do not
5048 slow down the running of your program.
5052 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]} @r{[}task @var{task-id}@r{]}
5053 Set a watchpoint for an expression. @value{GDBN} will break when the
5054 expression @var{expr} is written into by the program and its value
5055 changes. The simplest (and the most popular) use of this command is
5056 to watch the value of a single variable:
5059 (@value{GDBP}) watch foo
5062 If the command includes a @code{@r{[}thread @var{thread-id}@r{]}}
5063 argument, @value{GDBN} breaks only when the thread identified by
5064 @var{thread-id} changes the value of @var{expr}. If any other threads
5065 change the value of @var{expr}, @value{GDBN} will not break. Note
5066 that watchpoints restricted to a single thread in this way only work
5067 with Hardware Watchpoints.
5069 Similarly, if the @code{task} argument is given, then the watchpoint
5070 will be specific to the indicated Ada task (@pxref{Ada Tasks}).
5072 Ordinarily a watchpoint respects the scope of variables in @var{expr}
5073 (see below). The @code{-location} argument tells @value{GDBN} to
5074 instead watch the memory referred to by @var{expr}. In this case,
5075 @value{GDBN} will evaluate @var{expr}, take the address of the result,
5076 and watch the memory at that address. The type of the result is used
5077 to determine the size of the watched memory. If the expression's
5078 result does not have an address, then @value{GDBN} will print an
5081 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
5082 of masked watchpoints, if the current architecture supports this
5083 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
5084 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
5085 to an address to watch. The mask specifies that some bits of an address
5086 (the bits which are reset in the mask) should be ignored when matching
5087 the address accessed by the inferior against the watchpoint address.
5088 Thus, a masked watchpoint watches many addresses simultaneously---those
5089 addresses whose unmasked bits are identical to the unmasked bits in the
5090 watchpoint address. The @code{mask} argument implies @code{-location}.
5094 (@value{GDBP}) watch foo mask 0xffff00ff
5095 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
5099 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
5100 Set a watchpoint that will break when the value of @var{expr} is read
5104 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
5105 Set a watchpoint that will break when @var{expr} is either read from
5106 or written into by the program.
5108 @kindex info watchpoints @r{[}@var{list}@dots{}@r{]}
5109 @item info watchpoints @r{[}@var{list}@dots{}@r{]}
5110 This command prints a list of watchpoints, using the same format as
5111 @code{info break} (@pxref{Set Breaks}).
5114 If you watch for a change in a numerically entered address you need to
5115 dereference it, as the address itself is just a constant number which will
5116 never change. @value{GDBN} refuses to create a watchpoint that watches
5117 a never-changing value:
5120 (@value{GDBP}) watch 0x600850
5121 Cannot watch constant value 0x600850.
5122 (@value{GDBP}) watch *(int *) 0x600850
5123 Watchpoint 1: *(int *) 6293584
5126 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
5127 watchpoints execute very quickly, and the debugger reports a change in
5128 value at the exact instruction where the change occurs. If @value{GDBN}
5129 cannot set a hardware watchpoint, it sets a software watchpoint, which
5130 executes more slowly and reports the change in value at the next
5131 @emph{statement}, not the instruction, after the change occurs.
5133 @cindex use only software watchpoints
5134 You can force @value{GDBN} to use only software watchpoints with the
5135 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
5136 zero, @value{GDBN} will never try to use hardware watchpoints, even if
5137 the underlying system supports them. (Note that hardware-assisted
5138 watchpoints that were set @emph{before} setting
5139 @code{can-use-hw-watchpoints} to zero will still use the hardware
5140 mechanism of watching expression values.)
5143 @item set can-use-hw-watchpoints
5144 @kindex set can-use-hw-watchpoints
5145 Set whether or not to use hardware watchpoints.
5147 @item show can-use-hw-watchpoints
5148 @kindex show can-use-hw-watchpoints
5149 Show the current mode of using hardware watchpoints.
5152 For remote targets, you can restrict the number of hardware
5153 watchpoints @value{GDBN} will use, see @ref{set remote
5154 hardware-breakpoint-limit}.
5156 When you issue the @code{watch} command, @value{GDBN} reports
5159 Hardware watchpoint @var{num}: @var{expr}
5163 if it was able to set a hardware watchpoint.
5165 Currently, the @code{awatch} and @code{rwatch} commands can only set
5166 hardware watchpoints, because accesses to data that don't change the
5167 value of the watched expression cannot be detected without examining
5168 every instruction as it is being executed, and @value{GDBN} does not do
5169 that currently. If @value{GDBN} finds that it is unable to set a
5170 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
5171 will print a message like this:
5174 Expression cannot be implemented with read/access watchpoint.
5177 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
5178 data type of the watched expression is wider than what a hardware
5179 watchpoint on the target machine can handle. For example, some systems
5180 can only watch regions that are up to 4 bytes wide; on such systems you
5181 cannot set hardware watchpoints for an expression that yields a
5182 double-precision floating-point number (which is typically 8 bytes
5183 wide). As a work-around, it might be possible to break the large region
5184 into a series of smaller ones and watch them with separate watchpoints.
5186 If you set too many hardware watchpoints, @value{GDBN} might be unable
5187 to insert all of them when you resume the execution of your program.
5188 Since the precise number of active watchpoints is unknown until such
5189 time as the program is about to be resumed, @value{GDBN} might not be
5190 able to warn you about this when you set the watchpoints, and the
5191 warning will be printed only when the program is resumed:
5194 Hardware watchpoint @var{num}: Could not insert watchpoint
5198 If this happens, delete or disable some of the watchpoints.
5200 Watching complex expressions that reference many variables can also
5201 exhaust the resources available for hardware-assisted watchpoints.
5202 That's because @value{GDBN} needs to watch every variable in the
5203 expression with separately allocated resources.
5205 If you call a function interactively using @code{print} or @code{call},
5206 any watchpoints you have set will be inactive until @value{GDBN} reaches another
5207 kind of breakpoint or the call completes.
5209 @value{GDBN} automatically deletes watchpoints that watch local
5210 (automatic) variables, or expressions that involve such variables, when
5211 they go out of scope, that is, when the execution leaves the block in
5212 which these variables were defined. In particular, when the program
5213 being debugged terminates, @emph{all} local variables go out of scope,
5214 and so only watchpoints that watch global variables remain set. If you
5215 rerun the program, you will need to set all such watchpoints again. One
5216 way of doing that would be to set a code breakpoint at the entry to the
5217 @code{main} function and when it breaks, set all the watchpoints.
5219 @cindex watchpoints and threads
5220 @cindex threads and watchpoints
5221 In multi-threaded programs, watchpoints will detect changes to the
5222 watched expression from every thread.
5225 @emph{Warning:} In multi-threaded programs, software watchpoints
5226 have only limited usefulness. If @value{GDBN} creates a software
5227 watchpoint, it can only watch the value of an expression @emph{in a
5228 single thread}. If you are confident that the expression can only
5229 change due to the current thread's activity (and if you are also
5230 confident that no other thread can become current), then you can use
5231 software watchpoints as usual. However, @value{GDBN} may not notice
5232 when a non-current thread's activity changes the expression. (Hardware
5233 watchpoints, in contrast, watch an expression in all threads.)
5236 @xref{set remote hardware-watchpoint-limit}.
5238 @node Set Catchpoints
5239 @subsection Setting Catchpoints
5240 @cindex catchpoints, setting
5241 @cindex exception handlers
5242 @cindex event handling
5244 You can use @dfn{catchpoints} to cause the debugger to stop for certain
5245 kinds of program events, such as C@t{++} exceptions or the loading of a
5246 shared library. Use the @code{catch} command to set a catchpoint.
5250 @item catch @var{event}
5251 Stop when @var{event} occurs. The @var{event} can be any of the following:
5254 @item throw @r{[}@var{regexp}@r{]}
5255 @itemx rethrow @r{[}@var{regexp}@r{]}
5256 @itemx catch @r{[}@var{regexp}@r{]}
5258 @kindex catch rethrow
5260 @cindex stop on C@t{++} exceptions
5261 The throwing, re-throwing, or catching of a C@t{++} exception.
5263 If @var{regexp} is given, then only exceptions whose type matches the
5264 regular expression will be caught.
5266 @vindex $_exception@r{, convenience variable}
5267 The convenience variable @code{$_exception} is available at an
5268 exception-related catchpoint, on some systems. This holds the
5269 exception being thrown.
5271 There are currently some limitations to C@t{++} exception handling in
5276 The support for these commands is system-dependent. Currently, only
5277 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
5281 The regular expression feature and the @code{$_exception} convenience
5282 variable rely on the presence of some SDT probes in @code{libstdc++}.
5283 If these probes are not present, then these features cannot be used.
5284 These probes were first available in the GCC 4.8 release, but whether
5285 or not they are available in your GCC also depends on how it was
5289 The @code{$_exception} convenience variable is only valid at the
5290 instruction at which an exception-related catchpoint is set.
5293 When an exception-related catchpoint is hit, @value{GDBN} stops at a
5294 location in the system library which implements runtime exception
5295 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
5296 (@pxref{Selection}) to get to your code.
5299 If you call a function interactively, @value{GDBN} normally returns
5300 control to you when the function has finished executing. If the call
5301 raises an exception, however, the call may bypass the mechanism that
5302 returns control to you and cause your program either to abort or to
5303 simply continue running until it hits a breakpoint, catches a signal
5304 that @value{GDBN} is listening for, or exits. This is the case even if
5305 you set a catchpoint for the exception; catchpoints on exceptions are
5306 disabled within interactive calls. @xref{Calling}, for information on
5307 controlling this with @code{set unwind-on-terminating-exception}.
5310 You cannot raise an exception interactively.
5313 You cannot install an exception handler interactively.
5316 @item exception @r{[}@var{name}@r{]}
5317 @kindex catch exception
5318 @cindex Ada exception catching
5319 @cindex catch Ada exceptions
5320 An Ada exception being raised. If an exception name is specified
5321 at the end of the command (eg @code{catch exception Program_Error}),
5322 the debugger will stop only when this specific exception is raised.
5323 Otherwise, the debugger stops execution when any Ada exception is raised.
5325 When inserting an exception catchpoint on a user-defined exception whose
5326 name is identical to one of the exceptions defined by the language, the
5327 fully qualified name must be used as the exception name. Otherwise,
5328 @value{GDBN} will assume that it should stop on the pre-defined exception
5329 rather than the user-defined one. For instance, assuming an exception
5330 called @code{Constraint_Error} is defined in package @code{Pck}, then
5331 the command to use to catch such exceptions is @kbd{catch exception
5332 Pck.Constraint_Error}.
5334 @vindex $_ada_exception@r{, convenience variable}
5335 The convenience variable @code{$_ada_exception} holds the address of
5336 the exception being thrown. This can be useful when setting a
5337 condition for such a catchpoint.
5339 @item exception unhandled
5340 @kindex catch exception unhandled
5341 An exception that was raised but is not handled by the program. The
5342 convenience variable @code{$_ada_exception} is set as for @code{catch
5345 @item handlers @r{[}@var{name}@r{]}
5346 @kindex catch handlers
5347 @cindex Ada exception handlers catching
5348 @cindex catch Ada exceptions when handled
5349 An Ada exception being handled. If an exception name is
5350 specified at the end of the command
5351 (eg @kbd{catch handlers Program_Error}), the debugger will stop
5352 only when this specific exception is handled.
5353 Otherwise, the debugger stops execution when any Ada exception is handled.
5355 When inserting a handlers catchpoint on a user-defined
5356 exception whose name is identical to one of the exceptions
5357 defined by the language, the fully qualified name must be used
5358 as the exception name. Otherwise, @value{GDBN} will assume that it
5359 should stop on the pre-defined exception rather than the
5360 user-defined one. For instance, assuming an exception called
5361 @code{Constraint_Error} is defined in package @code{Pck}, then the
5362 command to use to catch such exceptions handling is
5363 @kbd{catch handlers Pck.Constraint_Error}.
5365 The convenience variable @code{$_ada_exception} is set as for
5366 @code{catch exception}.
5369 @kindex catch assert
5370 A failed Ada assertion. Note that the convenience variable
5371 @code{$_ada_exception} is @emph{not} set by this catchpoint.
5375 @cindex break on fork/exec
5376 A call to @code{exec}.
5378 @anchor{catch syscall}
5380 @itemx syscall @r{[}@var{name} @r{|} @var{number} @r{|} @r{group:}@var{groupname} @r{|} @r{g:}@var{groupname}@r{]} @dots{}
5381 @kindex catch syscall
5382 @cindex break on a system call.
5383 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
5384 syscall is a mechanism for application programs to request a service
5385 from the operating system (OS) or one of the OS system services.
5386 @value{GDBN} can catch some or all of the syscalls issued by the
5387 debuggee, and show the related information for each syscall. If no
5388 argument is specified, calls to and returns from all system calls
5391 @var{name} can be any system call name that is valid for the
5392 underlying OS. Just what syscalls are valid depends on the OS. On
5393 GNU and Unix systems, you can find the full list of valid syscall
5394 names on @file{/usr/include/asm/unistd.h}.
5396 @c For MS-Windows, the syscall names and the corresponding numbers
5397 @c can be found, e.g., on this URL:
5398 @c http://www.metasploit.com/users/opcode/syscalls.html
5399 @c but we don't support Windows syscalls yet.
5401 Normally, @value{GDBN} knows in advance which syscalls are valid for
5402 each OS, so you can use the @value{GDBN} command-line completion
5403 facilities (@pxref{Completion,, command completion}) to list the
5406 You may also specify the system call numerically. A syscall's
5407 number is the value passed to the OS's syscall dispatcher to
5408 identify the requested service. When you specify the syscall by its
5409 name, @value{GDBN} uses its database of syscalls to convert the name
5410 into the corresponding numeric code, but using the number directly
5411 may be useful if @value{GDBN}'s database does not have the complete
5412 list of syscalls on your system (e.g., because @value{GDBN} lags
5413 behind the OS upgrades).
5415 You may specify a group of related syscalls to be caught at once using
5416 the @code{group:} syntax (@code{g:} is a shorter equivalent). For
5417 instance, on some platforms @value{GDBN} allows you to catch all
5418 network related syscalls, by passing the argument @code{group:network}
5419 to @code{catch syscall}. Note that not all syscall groups are
5420 available in every system. You can use the command completion
5421 facilities (@pxref{Completion,, command completion}) to list the
5422 syscall groups available on your environment.
5424 The example below illustrates how this command works if you don't provide
5428 (@value{GDBP}) catch syscall
5429 Catchpoint 1 (syscall)
5431 Starting program: /tmp/catch-syscall
5433 Catchpoint 1 (call to syscall 'close'), \
5434 0xffffe424 in __kernel_vsyscall ()
5438 Catchpoint 1 (returned from syscall 'close'), \
5439 0xffffe424 in __kernel_vsyscall ()
5443 Here is an example of catching a system call by name:
5446 (@value{GDBP}) catch syscall chroot
5447 Catchpoint 1 (syscall 'chroot' [61])
5449 Starting program: /tmp/catch-syscall
5451 Catchpoint 1 (call to syscall 'chroot'), \
5452 0xffffe424 in __kernel_vsyscall ()
5456 Catchpoint 1 (returned from syscall 'chroot'), \
5457 0xffffe424 in __kernel_vsyscall ()
5461 An example of specifying a system call numerically. In the case
5462 below, the syscall number has a corresponding entry in the XML
5463 file, so @value{GDBN} finds its name and prints it:
5466 (@value{GDBP}) catch syscall 252
5467 Catchpoint 1 (syscall(s) 'exit_group')
5469 Starting program: /tmp/catch-syscall
5471 Catchpoint 1 (call to syscall 'exit_group'), \
5472 0xffffe424 in __kernel_vsyscall ()
5476 Program exited normally.
5480 Here is an example of catching a syscall group:
5483 (@value{GDBP}) catch syscall group:process
5484 Catchpoint 1 (syscalls 'exit' [1] 'fork' [2] 'waitpid' [7]
5485 'execve' [11] 'wait4' [114] 'clone' [120] 'vfork' [190]
5486 'exit_group' [252] 'waitid' [284] 'unshare' [310])
5488 Starting program: /tmp/catch-syscall
5490 Catchpoint 1 (call to syscall fork), 0x00007ffff7df4e27 in open64 ()
5491 from /lib64/ld-linux-x86-64.so.2
5497 However, there can be situations when there is no corresponding name
5498 in XML file for that syscall number. In this case, @value{GDBN} prints
5499 a warning message saying that it was not able to find the syscall name,
5500 but the catchpoint will be set anyway. See the example below:
5503 (@value{GDBP}) catch syscall 764
5504 warning: The number '764' does not represent a known syscall.
5505 Catchpoint 2 (syscall 764)
5509 If you configure @value{GDBN} using the @samp{--without-expat} option,
5510 it will not be able to display syscall names. Also, if your
5511 architecture does not have an XML file describing its system calls,
5512 you will not be able to see the syscall names. It is important to
5513 notice that these two features are used for accessing the syscall
5514 name database. In either case, you will see a warning like this:
5517 (@value{GDBP}) catch syscall
5518 warning: Could not open "syscalls/i386-linux.xml"
5519 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
5520 GDB will not be able to display syscall names.
5521 Catchpoint 1 (syscall)
5525 Of course, the file name will change depending on your architecture and system.
5527 Still using the example above, you can also try to catch a syscall by its
5528 number. In this case, you would see something like:
5531 (@value{GDBP}) catch syscall 252
5532 Catchpoint 1 (syscall(s) 252)
5535 Again, in this case @value{GDBN} would not be able to display syscall's names.
5539 A call to @code{fork}.
5543 A call to @code{vfork}.
5545 @item load @r{[}@var{regexp}@r{]}
5546 @itemx unload @r{[}@var{regexp}@r{]}
5548 @kindex catch unload
5549 The loading or unloading of a shared library. If @var{regexp} is
5550 given, then the catchpoint will stop only if the regular expression
5551 matches one of the affected libraries.
5553 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
5554 @kindex catch signal
5555 The delivery of a signal.
5557 With no arguments, this catchpoint will catch any signal that is not
5558 used internally by @value{GDBN}, specifically, all signals except
5559 @samp{SIGTRAP} and @samp{SIGINT}.
5561 With the argument @samp{all}, all signals, including those used by
5562 @value{GDBN}, will be caught. This argument cannot be used with other
5565 Otherwise, the arguments are a list of signal names as given to
5566 @code{handle} (@pxref{Signals}). Only signals specified in this list
5569 One reason that @code{catch signal} can be more useful than
5570 @code{handle} is that you can attach commands and conditions to the
5573 When a signal is caught by a catchpoint, the signal's @code{stop} and
5574 @code{print} settings, as specified by @code{handle}, are ignored.
5575 However, whether the signal is still delivered to the inferior depends
5576 on the @code{pass} setting; this can be changed in the catchpoint's
5581 @item tcatch @var{event}
5583 Set a catchpoint that is enabled only for one stop. The catchpoint is
5584 automatically deleted after the first time the event is caught.
5588 Use the @code{info break} command to list the current catchpoints.
5592 @subsection Deleting Breakpoints
5594 @cindex clearing breakpoints, watchpoints, catchpoints
5595 @cindex deleting breakpoints, watchpoints, catchpoints
5596 It is often necessary to eliminate a breakpoint, watchpoint, or
5597 catchpoint once it has done its job and you no longer want your program
5598 to stop there. This is called @dfn{deleting} the breakpoint. A
5599 breakpoint that has been deleted no longer exists; it is forgotten.
5601 With the @code{clear} command you can delete breakpoints according to
5602 where they are in your program. With the @code{delete} command you can
5603 delete individual breakpoints, watchpoints, or catchpoints by specifying
5604 their breakpoint numbers.
5606 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
5607 automatically ignores breakpoints on the first instruction to be executed
5608 when you continue execution without changing the execution address.
5613 Delete any breakpoints at the next instruction to be executed in the
5614 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
5615 the innermost frame is selected, this is a good way to delete a
5616 breakpoint where your program just stopped.
5618 @item clear @var{locspec}
5619 Delete any breakpoint with a code location that corresponds to
5620 @var{locspec}. @xref{Location Specifications}, for the various forms
5621 of @var{locspec}. Which code locations correspond to @var{locspec}
5622 depends on the form used in the location specification @var{locspec}:
5626 @itemx @var{filename}:@var{linenum}
5627 @itemx -line @var{linenum}
5628 @itemx -source @var{filename} -line @var{linenum}
5629 If @var{locspec} specifies a line number, with or without a file name,
5630 the command deletes any breakpoint with a code location that is at or
5631 within the specified line @var{linenum} in files that match the
5632 specified @var{filename}. If @var{filename} is omitted, it defaults
5633 to the current source file.
5635 @item *@var{address}
5636 If @var{locspec} specifies an address, the command deletes any
5637 breakpoint with a code location that is at the given @var{address}.
5639 @item @var{function}
5640 @itemx -function @var{function}
5641 If @var{locspec} specifies a function, the command deletes any
5642 breakpoint with a code location that is at the entry to any function
5643 whose name matches @var{function}.
5646 Ambiguity in names of files and functions can be resolved as described
5647 in @ref{Location Specifications}.
5649 @cindex delete breakpoints
5651 @kindex d @r{(@code{delete})}
5652 @item delete @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5653 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
5654 list specified as argument. If no argument is specified, delete all
5655 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
5656 confirm off}). You can abbreviate this command as @code{d}.
5660 @subsection Disabling Breakpoints
5662 @cindex enable/disable a breakpoint
5663 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
5664 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
5665 it had been deleted, but remembers the information on the breakpoint so
5666 that you can @dfn{enable} it again later.
5668 You disable and enable breakpoints, watchpoints, and catchpoints with
5669 the @code{enable} and @code{disable} commands, optionally specifying
5670 one or more breakpoint numbers as arguments. Use @code{info break} to
5671 print a list of all breakpoints, watchpoints, and catchpoints if you
5672 do not know which numbers to use.
5674 Disabling and enabling a breakpoint that has multiple locations
5675 affects all of its locations.
5677 A breakpoint, watchpoint, or catchpoint can have any of several
5678 different states of enablement:
5682 Enabled. The breakpoint stops your program. A breakpoint set
5683 with the @code{break} command starts out in this state.
5685 Disabled. The breakpoint has no effect on your program.
5687 Enabled once. The breakpoint stops your program, but then becomes
5690 Enabled for a count. The breakpoint stops your program for the next
5691 N times, then becomes disabled.
5693 Enabled for deletion. The breakpoint stops your program, but
5694 immediately after it does so it is deleted permanently. A breakpoint
5695 set with the @code{tbreak} command starts out in this state.
5698 You can use the following commands to enable or disable breakpoints,
5699 watchpoints, and catchpoints:
5703 @kindex dis @r{(@code{disable})}
5704 @item disable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5705 Disable the specified breakpoints---or all breakpoints, if none are
5706 listed. A disabled breakpoint has no effect but is not forgotten. All
5707 options such as ignore-counts, conditions and commands are remembered in
5708 case the breakpoint is enabled again later. You may abbreviate
5709 @code{disable} as @code{dis}.
5712 @item enable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5713 Enable the specified breakpoints (or all defined breakpoints). They
5714 become effective once again in stopping your program.
5716 @item enable @r{[}breakpoints@r{]} once @var{list}@dots{}
5717 Enable the specified breakpoints temporarily. @value{GDBN} disables any
5718 of these breakpoints immediately after stopping your program.
5720 @item enable @r{[}breakpoints@r{]} count @var{count} @var{list}@dots{}
5721 Enable the specified breakpoints temporarily. @value{GDBN} records
5722 @var{count} with each of the specified breakpoints, and decrements a
5723 breakpoint's count when it is hit. When any count reaches 0,
5724 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
5725 count (@pxref{Conditions, ,Break Conditions}), that will be
5726 decremented to 0 before @var{count} is affected.
5728 @item enable @r{[}breakpoints@r{]} delete @var{list}@dots{}
5729 Enable the specified breakpoints to work once, then die. @value{GDBN}
5730 deletes any of these breakpoints as soon as your program stops there.
5731 Breakpoints set by the @code{tbreak} command start out in this state.
5734 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
5735 @c confusing: tbreak is also initially enabled.
5736 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
5737 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
5738 subsequently, they become disabled or enabled only when you use one of
5739 the commands above. (The command @code{until} can set and delete a
5740 breakpoint of its own, but it does not change the state of your other
5741 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
5745 @subsection Break Conditions
5746 @cindex conditional breakpoints
5747 @cindex breakpoint conditions
5749 @c FIXME what is scope of break condition expr? Context where wanted?
5750 @c in particular for a watchpoint?
5751 The simplest sort of breakpoint breaks every time your program reaches a
5752 specified place. You can also specify a @dfn{condition} for a
5753 breakpoint. A condition is just a Boolean expression in your
5754 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
5755 a condition evaluates the expression each time your program reaches it,
5756 and your program stops only if the condition is @emph{true}.
5758 This is the converse of using assertions for program validation; in that
5759 situation, you want to stop when the assertion is violated---that is,
5760 when the condition is false. In C, if you want to test an assertion expressed
5761 by the condition @var{assert}, you should set the condition
5762 @samp{! @var{assert}} on the appropriate breakpoint.
5764 Conditions are also accepted for watchpoints; you may not need them,
5765 since a watchpoint is inspecting the value of an expression anyhow---but
5766 it might be simpler, say, to just set a watchpoint on a variable name,
5767 and specify a condition that tests whether the new value is an interesting
5770 Break conditions can have side effects, and may even call functions in
5771 your program. This can be useful, for example, to activate functions
5772 that log program progress, or to use your own print functions to
5773 format special data structures. The effects are completely predictable
5774 unless there is another enabled breakpoint at the same address. (In
5775 that case, @value{GDBN} might see the other breakpoint first and stop your
5776 program without checking the condition of this one.) Note that
5777 breakpoint commands are usually more convenient and flexible than break
5779 purpose of performing side effects when a breakpoint is reached
5780 (@pxref{Break Commands, ,Breakpoint Command Lists}).
5782 Breakpoint conditions can also be evaluated on the target's side if
5783 the target supports it. Instead of evaluating the conditions locally,
5784 @value{GDBN} encodes the expression into an agent expression
5785 (@pxref{Agent Expressions}) suitable for execution on the target,
5786 independently of @value{GDBN}. Global variables become raw memory
5787 locations, locals become stack accesses, and so forth.
5789 In this case, @value{GDBN} will only be notified of a breakpoint trigger
5790 when its condition evaluates to true. This mechanism may provide faster
5791 response times depending on the performance characteristics of the target
5792 since it does not need to keep @value{GDBN} informed about
5793 every breakpoint trigger, even those with false conditions.
5795 Break conditions can be specified when a breakpoint is set, by using
5796 @samp{if} in the arguments to the @code{break} command. @xref{Set
5797 Breaks, ,Setting Breakpoints}. They can also be changed at any time
5798 with the @code{condition} command.
5800 You can also use the @code{if} keyword with the @code{watch} command.
5801 The @code{catch} command does not recognize the @code{if} keyword;
5802 @code{condition} is the only way to impose a further condition on a
5807 @item condition @var{bnum} @var{expression}
5808 Specify @var{expression} as the break condition for breakpoint,
5809 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
5810 breakpoint @var{bnum} stops your program only if the value of
5811 @var{expression} is true (nonzero, in C). When you use
5812 @code{condition}, @value{GDBN} checks @var{expression} immediately for
5813 syntactic correctness, and to determine whether symbols in it have
5814 referents in the context of your breakpoint. If @var{expression} uses
5815 symbols not referenced in the context of the breakpoint, @value{GDBN}
5816 prints an error message:
5819 No symbol "foo" in current context.
5824 not actually evaluate @var{expression} at the time the @code{condition}
5825 command (or a command that sets a breakpoint with a condition, like
5826 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
5828 @item condition -force @var{bnum} @var{expression}
5829 When the @code{-force} flag is used, define the condition even if
5830 @var{expression} is invalid at all the current locations of breakpoint
5831 @var{bnum}. This is similar to the @code{-force-condition} option
5832 of the @code{break} command.
5834 @item condition @var{bnum}
5835 Remove the condition from breakpoint number @var{bnum}. It becomes
5836 an ordinary unconditional breakpoint.
5839 @cindex ignore count (of breakpoint)
5840 A special case of a breakpoint condition is to stop only when the
5841 breakpoint has been reached a certain number of times. This is so
5842 useful that there is a special way to do it, using the @dfn{ignore
5843 count} of the breakpoint. Every breakpoint has an ignore count, which
5844 is an integer. Most of the time, the ignore count is zero, and
5845 therefore has no effect. But if your program reaches a breakpoint whose
5846 ignore count is positive, then instead of stopping, it just decrements
5847 the ignore count by one and continues. As a result, if the ignore count
5848 value is @var{n}, the breakpoint does not stop the next @var{n} times
5849 your program reaches it.
5853 @item ignore @var{bnum} @var{count}
5854 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
5855 The next @var{count} times the breakpoint is reached, your program's
5856 execution does not stop; other than to decrement the ignore count, @value{GDBN}
5859 To make the breakpoint stop the next time it is reached, specify
5862 When you use @code{continue} to resume execution of your program from a
5863 breakpoint, you can specify an ignore count directly as an argument to
5864 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
5865 Stepping,,Continuing and Stepping}.
5867 If a breakpoint has a positive ignore count and a condition, the
5868 condition is not checked. Once the ignore count reaches zero,
5869 @value{GDBN} resumes checking the condition.
5871 You could achieve the effect of the ignore count with a condition such
5872 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
5873 is decremented each time. @xref{Convenience Vars, ,Convenience
5877 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
5880 @node Break Commands
5881 @subsection Breakpoint Command Lists
5883 @cindex breakpoint commands
5884 You can give any breakpoint (or watchpoint or catchpoint) a series of
5885 commands to execute when your program stops due to that breakpoint. For
5886 example, you might want to print the values of certain expressions, or
5887 enable other breakpoints.
5891 @kindex end@r{ (breakpoint commands)}
5892 @item commands @r{[}@var{list}@dots{}@r{]}
5893 @itemx @dots{} @var{command-list} @dots{}
5895 Specify a list of commands for the given breakpoints. The commands
5896 themselves appear on the following lines. Type a line containing just
5897 @code{end} to terminate the commands.
5899 To remove all commands from a breakpoint, type @code{commands} and
5900 follow it immediately with @code{end}; that is, give no commands.
5902 With no argument, @code{commands} refers to the last breakpoint,
5903 watchpoint, or catchpoint set (not to the breakpoint most recently
5904 encountered). If the most recent breakpoints were set with a single
5905 command, then the @code{commands} will apply to all the breakpoints
5906 set by that command. This applies to breakpoints set by
5907 @code{rbreak}, and also applies when a single @code{break} command
5908 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
5912 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
5913 disabled within a @var{command-list}.
5915 Inside a command list, you can use the command
5916 @kbd{disable $_hit_bpnum} to disable the encountered breakpoint.
5918 If your breakpoint has several code locations, the command
5919 @kbd{disable $_hit_bpnum.$_hit_locno} will disable the specific breakpoint
5920 code location encountered. If the breakpoint has only one location,
5921 this command will disable the encountered breakpoint.
5923 You can use breakpoint commands to start your program up again. Simply
5924 use the @code{continue} command, or @code{step}, or any other command
5925 that resumes execution.
5927 Any other commands in the command list, after a command that resumes
5928 execution, are ignored. This is because any time you resume execution
5929 (even with a simple @code{next} or @code{step}), you may encounter
5930 another breakpoint---which could have its own command list, leading to
5931 ambiguities about which list to execute.
5934 If the first command you specify in a command list is @code{silent}, the
5935 usual message about stopping at a breakpoint is not printed. This may
5936 be desirable for breakpoints that are to print a specific message and
5937 then continue. If none of the remaining commands print anything, you
5938 see no sign that the breakpoint was reached. @code{silent} is
5939 meaningful only at the beginning of a breakpoint command list.
5941 The commands @code{echo}, @code{output}, and @code{printf} allow you to
5942 print precisely controlled output, and are often useful in silent
5943 breakpoints. @xref{Output, ,Commands for Controlled Output}.
5945 For example, here is how you could use breakpoint commands to print the
5946 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
5952 printf "x is %d\n",x
5957 One application for breakpoint commands is to compensate for one bug so
5958 you can test for another. Put a breakpoint just after the erroneous line
5959 of code, give it a condition to detect the case in which something
5960 erroneous has been done, and give it commands to assign correct values
5961 to any variables that need them. End with the @code{continue} command
5962 so that your program does not stop, and start with the @code{silent}
5963 command so that no output is produced. Here is an example:
5974 @node Dynamic Printf
5975 @subsection Dynamic Printf
5977 @cindex dynamic printf
5979 The dynamic printf command @code{dprintf} combines a breakpoint with
5980 formatted printing of your program's data to give you the effect of
5981 inserting @code{printf} calls into your program on-the-fly, without
5982 having to recompile it.
5984 In its most basic form, the output goes to the GDB console. However,
5985 you can set the variable @code{dprintf-style} for alternate handling.
5986 For instance, you can ask to format the output by calling your
5987 program's @code{printf} function. This has the advantage that the
5988 characters go to the program's output device, so they can recorded in
5989 redirects to files and so forth.
5991 If you are doing remote debugging with a stub or agent, you can also
5992 ask to have the printf handled by the remote agent. In addition to
5993 ensuring that the output goes to the remote program's device along
5994 with any other output the program might produce, you can also ask that
5995 the dprintf remain active even after disconnecting from the remote
5996 target. Using the stub/agent is also more efficient, as it can do
5997 everything without needing to communicate with @value{GDBN}.
6001 @item dprintf @var{locspec},@var{template},@var{expression}[,@var{expression}@dots{}]
6002 Whenever execution reaches a code location that results from resolving
6003 @var{locspec}, print the values of one or more @var{expressions} under
6004 the control of the string @var{template}. To print several values,
6005 separate them with commas.
6007 @item set dprintf-style @var{style}
6008 Set the dprintf output to be handled in one of several different
6009 styles enumerated below. A change of style affects all existing
6010 dynamic printfs immediately. (If you need individual control over the
6011 print commands, simply define normal breakpoints with
6012 explicitly-supplied command lists.)
6016 @kindex dprintf-style gdb
6017 Handle the output using the @value{GDBN} @code{printf} command. When
6018 using this style, it is possible to use the @samp{%V} format specifier
6019 (@pxref{%V Format Specifier}).
6022 @kindex dprintf-style call
6023 Handle the output by calling a function in your program (normally
6024 @code{printf}). When using this style the supported format specifiers
6025 depend entirely on the function being called.
6027 Most of @value{GDBN}'s format specifiers align with those supported by
6028 the @code{printf} function, however, @value{GDBN}'s @samp{%V} format
6029 specifier extension is not supported by @code{printf}. When using
6030 @samp{call} style dprintf, care should be taken to ensure that only
6031 format specifiers supported by the output function are used, otherwise
6032 the results will be undefined.
6035 @kindex dprintf-style agent
6036 Have the remote debugging agent (such as @code{gdbserver}) handle the
6037 output itself. This style is only available for agents that support
6038 running commands on the target. This style does not support the
6039 @samp{%V} format specifier.
6042 @item set dprintf-function @var{function}
6043 Set the function to call if the dprintf style is @code{call}. By
6044 default its value is @code{printf}. You may set it to any expression
6045 that @value{GDBN} can evaluate to a function, as per the @code{call}
6048 @item set dprintf-channel @var{channel}
6049 Set a ``channel'' for dprintf. If set to a non-empty value,
6050 @value{GDBN} will evaluate it as an expression and pass the result as
6051 a first argument to the @code{dprintf-function}, in the manner of
6052 @code{fprintf} and similar functions. Otherwise, the dprintf format
6053 string will be the first argument, in the manner of @code{printf}.
6055 As an example, if you wanted @code{dprintf} output to go to a logfile
6056 that is a standard I/O stream assigned to the variable @code{mylog},
6057 you could do the following:
6060 (@value{GDBP}) set dprintf-style call
6061 (@value{GDBP}) set dprintf-function fprintf
6062 (@value{GDBP}) set dprintf-channel mylog
6063 (@value{GDBP}) dprintf 25,"at line 25, glob=%d\n",glob
6064 Dprintf 1 at 0x123456: file main.c, line 25.
6065 (@value{GDBP}) info break
6066 1 dprintf keep y 0x00123456 in main at main.c:25
6067 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
6072 Note that the @code{info break} displays the dynamic printf commands
6073 as normal breakpoint commands; you can thus easily see the effect of
6074 the variable settings.
6076 @item set disconnected-dprintf on
6077 @itemx set disconnected-dprintf off
6078 @kindex set disconnected-dprintf
6079 Choose whether @code{dprintf} commands should continue to run if
6080 @value{GDBN} has disconnected from the target. This only applies
6081 if the @code{dprintf-style} is @code{agent}.
6083 @item show disconnected-dprintf off
6084 @kindex show disconnected-dprintf
6085 Show the current choice for disconnected @code{dprintf}.
6089 @value{GDBN} does not check the validity of function and channel,
6090 relying on you to supply values that are meaningful for the contexts
6091 in which they are being used. For instance, the function and channel
6092 may be the values of local variables, but if that is the case, then
6093 all enabled dynamic prints must be at locations within the scope of
6094 those locals. If evaluation fails, @value{GDBN} will report an error.
6096 @node Save Breakpoints
6097 @subsection How to save breakpoints to a file
6099 To save breakpoint definitions to a file use the @w{@code{save
6100 breakpoints}} command.
6103 @kindex save breakpoints
6104 @cindex save breakpoints to a file for future sessions
6105 @item save breakpoints [@var{filename}]
6106 This command saves all current breakpoint definitions together with
6107 their commands and ignore counts, into a file @file{@var{filename}}
6108 suitable for use in a later debugging session. This includes all
6109 types of breakpoints (breakpoints, watchpoints, catchpoints,
6110 tracepoints). To read the saved breakpoint definitions, use the
6111 @code{source} command (@pxref{Command Files}). Note that watchpoints
6112 with expressions involving local variables may fail to be recreated
6113 because it may not be possible to access the context where the
6114 watchpoint is valid anymore. Because the saved breakpoint definitions
6115 are simply a sequence of @value{GDBN} commands that recreate the
6116 breakpoints, you can edit the file in your favorite editing program,
6117 and remove the breakpoint definitions you're not interested in, or
6118 that can no longer be recreated.
6121 @node Static Probe Points
6122 @subsection Static Probe Points
6124 @cindex static probe point, SystemTap
6125 @cindex static probe point, DTrace
6126 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
6127 for Statically Defined Tracing, and the probes are designed to have a tiny
6128 runtime code and data footprint, and no dynamic relocations.
6130 Currently, the following types of probes are supported on
6131 ELF-compatible systems:
6135 @item @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
6136 @acronym{SDT} probes@footnote{See
6137 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
6138 for more information on how to add @code{SystemTap} @acronym{SDT}
6139 probes in your applications.}. @code{SystemTap} probes are usable
6140 from assembly, C and C@t{++} languages@footnote{See
6141 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
6142 for a good reference on how the @acronym{SDT} probes are implemented.}.
6144 @item @code{DTrace} (@uref{http://oss.oracle.com/projects/DTrace})
6145 @acronym{USDT} probes. @code{DTrace} probes are usable from C and
6149 @cindex semaphores on static probe points
6150 Some @code{SystemTap} probes have an associated semaphore variable;
6151 for instance, this happens automatically if you defined your probe
6152 using a DTrace-style @file{.d} file. If your probe has a semaphore,
6153 @value{GDBN} will automatically enable it when you specify a
6154 breakpoint using the @samp{-probe-stap} notation. But, if you put a
6155 breakpoint at a probe's location by some other method (e.g.,
6156 @code{break file:line}), then @value{GDBN} will not automatically set
6157 the semaphore. @code{DTrace} probes do not support semaphores.
6159 You can examine the available static static probes using @code{info
6160 probes}, with optional arguments:
6164 @item info probes @r{[}@var{type}@r{]} @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
6165 If given, @var{type} is either @code{stap} for listing
6166 @code{SystemTap} probes or @code{dtrace} for listing @code{DTrace}
6167 probes. If omitted all probes are listed regardless of their types.
6169 If given, @var{provider} is a regular expression used to match against provider
6170 names when selecting which probes to list. If omitted, probes by all
6171 probes from all providers are listed.
6173 If given, @var{name} is a regular expression to match against probe names
6174 when selecting which probes to list. If omitted, probe names are not
6175 considered when deciding whether to display them.
6177 If given, @var{objfile} is a regular expression used to select which
6178 object files (executable or shared libraries) to examine. If not
6179 given, all object files are considered.
6181 @item info probes all
6182 List the available static probes, from all types.
6185 @cindex enabling and disabling probes
6186 Some probe points can be enabled and/or disabled. The effect of
6187 enabling or disabling a probe depends on the type of probe being
6188 handled. Some @code{DTrace} probes can be enabled or
6189 disabled, but @code{SystemTap} probes cannot be disabled.
6191 You can enable (or disable) one or more probes using the following
6192 commands, with optional arguments:
6194 @anchor{enable probes}
6196 @kindex enable probes
6197 @item enable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
6198 If given, @var{provider} is a regular expression used to match against
6199 provider names when selecting which probes to enable. If omitted,
6200 all probes from all providers are enabled.
6202 If given, @var{name} is a regular expression to match against probe
6203 names when selecting which probes to enable. If omitted, probe names
6204 are not considered when deciding whether to enable them.
6206 If given, @var{objfile} is a regular expression used to select which
6207 object files (executable or shared libraries) to examine. If not
6208 given, all object files are considered.
6210 @kindex disable probes
6211 @item disable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
6212 See the @code{enable probes} command above for a description of the
6213 optional arguments accepted by this command.
6216 @vindex $_probe_arg@r{, convenience variable}
6217 A probe may specify up to twelve arguments. These are available at the
6218 point at which the probe is defined---that is, when the current PC is
6219 at the probe's location. The arguments are available using the
6220 convenience variables (@pxref{Convenience Vars})
6221 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. In @code{SystemTap}
6222 probes each probe argument is an integer of the appropriate size;
6223 types are not preserved. In @code{DTrace} probes types are preserved
6224 provided that they are recognized as such by @value{GDBN}; otherwise
6225 the value of the probe argument will be a long integer. The
6226 convenience variable @code{$_probe_argc} holds the number of arguments
6227 at the current probe point.
6229 These variables are always available, but attempts to access them at
6230 any location other than a probe point will cause @value{GDBN} to give
6234 @c @ifclear BARETARGET
6235 @node Error in Breakpoints
6236 @subsection ``Cannot insert breakpoints''
6238 If you request too many active hardware-assisted breakpoints and
6239 watchpoints, you will see this error message:
6241 @c FIXME: the precise wording of this message may change; the relevant
6242 @c source change is not committed yet (Sep 3, 1999).
6244 Stopped; cannot insert breakpoints.
6245 You may have requested too many hardware breakpoints and watchpoints.
6249 This message is printed when you attempt to resume the program, since
6250 only then @value{GDBN} knows exactly how many hardware breakpoints and
6251 watchpoints it needs to insert.
6253 When this message is printed, you need to disable or remove some of the
6254 hardware-assisted breakpoints and watchpoints, and then continue.
6256 @node Breakpoint-related Warnings
6257 @subsection ``Breakpoint address adjusted...''
6258 @cindex breakpoint address adjusted
6260 Some processor architectures place constraints on the addresses at
6261 which breakpoints may be placed. For architectures thus constrained,
6262 @value{GDBN} will attempt to adjust the breakpoint's address to comply
6263 with the constraints dictated by the architecture.
6265 One example of such an architecture is the Fujitsu FR-V. The FR-V is
6266 a VLIW architecture in which a number of RISC-like instructions may be
6267 bundled together for parallel execution. The FR-V architecture
6268 constrains the location of a breakpoint instruction within such a
6269 bundle to the instruction with the lowest address. @value{GDBN}
6270 honors this constraint by adjusting a breakpoint's address to the
6271 first in the bundle.
6273 It is not uncommon for optimized code to have bundles which contain
6274 instructions from different source statements, thus it may happen that
6275 a breakpoint's address will be adjusted from one source statement to
6276 another. Since this adjustment may significantly alter @value{GDBN}'s
6277 breakpoint related behavior from what the user expects, a warning is
6278 printed when the breakpoint is first set and also when the breakpoint
6281 A warning like the one below is printed when setting a breakpoint
6282 that's been subject to address adjustment:
6285 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
6288 Such warnings are printed both for user settable and @value{GDBN}'s
6289 internal breakpoints. If you see one of these warnings, you should
6290 verify that a breakpoint set at the adjusted address will have the
6291 desired affect. If not, the breakpoint in question may be removed and
6292 other breakpoints may be set which will have the desired behavior.
6293 E.g., it may be sufficient to place the breakpoint at a later
6294 instruction. A conditional breakpoint may also be useful in some
6295 cases to prevent the breakpoint from triggering too often.
6297 @value{GDBN} will also issue a warning when stopping at one of these
6298 adjusted breakpoints:
6301 warning: Breakpoint 1 address previously adjusted from 0x00010414
6305 When this warning is encountered, it may be too late to take remedial
6306 action except in cases where the breakpoint is hit earlier or more
6307 frequently than expected.
6309 @node Continuing and Stepping
6310 @section Continuing and Stepping
6314 @cindex resuming execution
6315 @dfn{Continuing} means resuming program execution until your program
6316 completes normally. In contrast, @dfn{stepping} means executing just
6317 one more ``step'' of your program, where ``step'' may mean either one
6318 line of source code, or one machine instruction (depending on what
6319 particular command you use). Either when continuing or when stepping,
6320 your program may stop even sooner, due to a breakpoint or a signal. (If
6321 it stops due to a signal, you may want to use @code{handle}, or use
6322 @samp{signal 0} to resume execution (@pxref{Signals, ,Signals}),
6323 or you may step into the signal's handler (@pxref{stepping and signal
6328 @kindex c @r{(@code{continue})}
6329 @kindex fg @r{(resume foreground execution)}
6330 @item continue @r{[}@var{ignore-count}@r{]}
6331 @itemx c @r{[}@var{ignore-count}@r{]}
6332 @itemx fg @r{[}@var{ignore-count}@r{]}
6333 Resume program execution, at the address where your program last stopped;
6334 any breakpoints set at that address are bypassed. The optional argument
6335 @var{ignore-count} allows you to specify a further number of times to
6336 ignore a breakpoint at this location; its effect is like that of
6337 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
6339 The argument @var{ignore-count} is meaningful only when your program
6340 stopped due to a breakpoint. At other times, the argument to
6341 @code{continue} is ignored.
6343 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
6344 debugged program is deemed to be the foreground program) are provided
6345 purely for convenience, and have exactly the same behavior as
6349 To resume execution at a different place, you can use @code{return}
6350 (@pxref{Returning, ,Returning from a Function}) to go back to the
6351 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
6352 Different Address}) to go to an arbitrary location in your program.
6354 A typical technique for using stepping is to set a breakpoint
6355 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
6356 beginning of the function or the section of your program where a problem
6357 is believed to lie, run your program until it stops at that breakpoint,
6358 and then step through the suspect area, examining the variables that are
6359 interesting, until you see the problem happen.
6363 @kindex s @r{(@code{step})}
6365 Continue running your program until control reaches a different source
6366 line, then stop it and return control to @value{GDBN}. This command is
6367 abbreviated @code{s}.
6370 @c "without debugging information" is imprecise; actually "without line
6371 @c numbers in the debugging information". (gcc -g1 has debugging info but
6372 @c not line numbers). But it seems complex to try to make that
6373 @c distinction here.
6374 @emph{Warning:} If you use the @code{step} command while control is
6375 within a function that was compiled without debugging information,
6376 execution proceeds until control reaches a function that does have
6377 debugging information. Likewise, it will not step into a function which
6378 is compiled without debugging information. To step through functions
6379 without debugging information, use the @code{stepi} command, described
6383 The @code{step} command only stops at the first instruction of a source
6384 line. This prevents the multiple stops that could otherwise occur in
6385 @code{switch} statements, @code{for} loops, etc. @code{step} continues
6386 to stop if a function that has debugging information is called within
6387 the line. In other words, @code{step} @emph{steps inside} any functions
6388 called within the line.
6390 Also, the @code{step} command only enters a function if there is line
6391 number information for the function. Otherwise it acts like the
6392 @code{next} command. This avoids problems when using @code{cc -gl}
6393 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
6394 was any debugging information about the routine.
6396 @item step @var{count}
6397 Continue running as in @code{step}, but do so @var{count} times. If a
6398 breakpoint is reached, or a signal not related to stepping occurs before
6399 @var{count} steps, stepping stops right away.
6402 @kindex n @r{(@code{next})}
6403 @item next @r{[}@var{count}@r{]}
6404 Continue to the next source line in the current (innermost) stack frame.
6405 This is similar to @code{step}, but function calls that appear within
6406 the line of code are executed without stopping. Execution stops when
6407 control reaches a different line of code at the original stack level
6408 that was executing when you gave the @code{next} command. This command
6409 is abbreviated @code{n}.
6411 An argument @var{count} is a repeat count, as for @code{step}.
6414 @c FIX ME!! Do we delete this, or is there a way it fits in with
6415 @c the following paragraph? --- Vctoria
6417 @c @code{next} within a function that lacks debugging information acts like
6418 @c @code{step}, but any function calls appearing within the code of the
6419 @c function are executed without stopping.
6421 The @code{next} command only stops at the first instruction of a
6422 source line. This prevents multiple stops that could otherwise occur in
6423 @code{switch} statements, @code{for} loops, etc.
6425 @kindex set step-mode
6427 @cindex functions without line info, and stepping
6428 @cindex stepping into functions with no line info
6429 @itemx set step-mode on
6430 The @code{set step-mode on} command causes the @code{step} command to
6431 stop at the first instruction of a function which contains no debug line
6432 information rather than stepping over it.
6434 This is useful in cases where you may be interested in inspecting the
6435 machine instructions of a function which has no symbolic info and do not
6436 want @value{GDBN} to automatically skip over this function.
6438 @item set step-mode off
6439 Causes the @code{step} command to step over any functions which contains no
6440 debug information. This is the default.
6442 @item show step-mode
6443 Show whether @value{GDBN} will stop in or step over functions without
6444 source line debug information.
6447 @kindex fin @r{(@code{finish})}
6449 Continue running until just after function in the selected stack frame
6450 returns. Print the returned value (if any). This command can be
6451 abbreviated as @code{fin}.
6453 Contrast this with the @code{return} command (@pxref{Returning,
6454 ,Returning from a Function}).
6456 @kindex set print finish
6457 @kindex show print finish
6458 @item set print finish @r{[}on|off@r{]}
6459 @itemx show print finish
6460 By default the @code{finish} command will show the value that is
6461 returned by the function. This can be disabled using @code{set print
6462 finish off}. When disabled, the value is still entered into the value
6463 history (@pxref{Value History}), but not displayed.
6466 @kindex u @r{(@code{until})}
6467 @cindex run until specified location
6470 Continue running until a source line past the current line, in the
6471 current stack frame, is reached. This command is used to avoid single
6472 stepping through a loop more than once. It is like the @code{next}
6473 command, except that when @code{until} encounters a jump, it
6474 automatically continues execution until the program counter is greater
6475 than the address of the jump.
6477 This means that when you reach the end of a loop after single stepping
6478 though it, @code{until} makes your program continue execution until it
6479 exits the loop. In contrast, a @code{next} command at the end of a loop
6480 simply steps back to the beginning of the loop, which forces you to step
6481 through the next iteration.
6483 @code{until} always stops your program if it attempts to exit the current
6486 @code{until} may produce somewhat counterintuitive results if the order
6487 of machine code does not match the order of the source lines. For
6488 example, in the following excerpt from a debugging session, the @code{f}
6489 (@code{frame}) command shows that execution is stopped at line
6490 @code{206}; yet when we use @code{until}, we get to line @code{195}:
6494 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
6496 (@value{GDBP}) until
6497 195 for ( ; argc > 0; NEXTARG) @{
6500 This happened because, for execution efficiency, the compiler had
6501 generated code for the loop closure test at the end, rather than the
6502 start, of the loop---even though the test in a C @code{for}-loop is
6503 written before the body of the loop. The @code{until} command appeared
6504 to step back to the beginning of the loop when it advanced to this
6505 expression; however, it has not really gone to an earlier
6506 statement---not in terms of the actual machine code.
6508 @code{until} with no argument works by means of single
6509 instruction stepping, and hence is slower than @code{until} with an
6512 @item until @var{locspec}
6513 @itemx u @var{locspec}
6514 Continue running your program until either it reaches a code location
6515 that results from resolving @var{locspec}, or the current stack frame
6516 returns. @var{locspec} is any of the forms described in @ref{Location
6518 This form of the command uses temporary breakpoints, and
6519 hence is quicker than @code{until} without an argument. The specified
6520 location is actually reached only if it is in the current frame. This
6521 implies that @code{until} can be used to skip over recursive function
6522 invocations. For instance in the code below, if the current location is
6523 line @code{96}, issuing @code{until 99} will execute the program up to
6524 line @code{99} in the same invocation of factorial, i.e., after the inner
6525 invocations have returned.
6528 94 int factorial (int value)
6530 96 if (value > 1) @{
6531 97 value *= factorial (value - 1);
6538 @kindex advance @var{locspec}
6539 @item advance @var{locspec}
6540 Continue running your program until either it reaches a code location
6541 that results from resolving @var{locspec}, or the current stack frame
6542 returns. @var{locspec} is any of the forms described in @ref{Location
6543 Specifications}. This command is similar to @code{until}, but
6544 @code{advance} will not skip over recursive function calls, and the
6545 target code location doesn't have to be in the same frame as the
6550 @kindex si @r{(@code{stepi})}
6552 @itemx stepi @var{arg}
6554 Execute one machine instruction, then stop and return to the debugger.
6556 It is often useful to do @samp{display/i $pc} when stepping by machine
6557 instructions. This makes @value{GDBN} automatically display the next
6558 instruction to be executed, each time your program stops. @xref{Auto
6559 Display,, Automatic Display}.
6561 An argument is a repeat count, as in @code{step}.
6565 @kindex ni @r{(@code{nexti})}
6567 @itemx nexti @var{arg}
6569 Execute one machine instruction, but if it is a function call,
6570 proceed until the function returns.
6572 An argument is a repeat count, as in @code{next}.
6576 @anchor{range stepping}
6577 @cindex range stepping
6578 @cindex target-assisted range stepping
6579 By default, and if available, @value{GDBN} makes use of
6580 target-assisted @dfn{range stepping}. In other words, whenever you
6581 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
6582 tells the target to step the corresponding range of instruction
6583 addresses instead of issuing multiple single-steps. This speeds up
6584 line stepping, particularly for remote targets. Ideally, there should
6585 be no reason you would want to turn range stepping off. However, it's
6586 possible that a bug in the debug info, a bug in the remote stub (for
6587 remote targets), or even a bug in @value{GDBN} could make line
6588 stepping behave incorrectly when target-assisted range stepping is
6589 enabled. You can use the following command to turn off range stepping
6593 @kindex set range-stepping
6594 @kindex show range-stepping
6595 @item set range-stepping
6596 @itemx show range-stepping
6597 Control whether range stepping is enabled.
6599 If @code{on}, and the target supports it, @value{GDBN} tells the
6600 target to step a range of addresses itself, instead of issuing
6601 multiple single-steps. If @code{off}, @value{GDBN} always issues
6602 single-steps, even if range stepping is supported by the target. The
6603 default is @code{on}.
6607 @node Skipping Over Functions and Files
6608 @section Skipping Over Functions and Files
6609 @cindex skipping over functions and files
6611 The program you are debugging may contain some functions which are
6612 uninteresting to debug. The @code{skip} command lets you tell @value{GDBN} to
6613 skip a function, all functions in a file or a particular function in
6614 a particular file when stepping.
6616 For example, consider the following C function:
6627 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
6628 are not interested in stepping through @code{boring}. If you run @code{step}
6629 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
6630 step over both @code{foo} and @code{boring}!
6632 One solution is to @code{step} into @code{boring} and use the @code{finish}
6633 command to immediately exit it. But this can become tedious if @code{boring}
6634 is called from many places.
6636 A more flexible solution is to execute @kbd{skip boring}. This instructs
6637 @value{GDBN} never to step into @code{boring}. Now when you execute
6638 @code{step} at line 103, you'll step over @code{boring} and directly into
6641 Functions may be skipped by providing either a function name, linespec
6642 (@pxref{Location Specifications}), regular expression that matches the function's
6643 name, file name or a @code{glob}-style pattern that matches the file name.
6645 On Posix systems the form of the regular expression is
6646 ``Extended Regular Expressions''. See for example @samp{man 7 regex}
6647 on @sc{gnu}/Linux systems. On non-Posix systems the form of the regular
6648 expression is whatever is provided by the @code{regcomp} function of
6649 the underlying system.
6650 See for example @samp{man 7 glob} on @sc{gnu}/Linux systems for a
6651 description of @code{glob}-style patterns.
6655 @item skip @r{[}@var{options}@r{]}
6656 The basic form of the @code{skip} command takes zero or more options
6657 that specify what to skip.
6658 The @var{options} argument is any useful combination of the following:
6661 @item -file @var{file}
6662 @itemx -fi @var{file}
6663 Functions in @var{file} will be skipped over when stepping.
6665 @item -gfile @var{file-glob-pattern}
6666 @itemx -gfi @var{file-glob-pattern}
6667 @cindex skipping over files via glob-style patterns
6668 Functions in files matching @var{file-glob-pattern} will be skipped
6672 (@value{GDBP}) skip -gfi utils/*.c
6675 @item -function @var{linespec}
6676 @itemx -fu @var{linespec}
6677 Functions named by @var{linespec} or the function containing the line
6678 named by @var{linespec} will be skipped over when stepping.
6679 @xref{Location Specifications}.
6681 @item -rfunction @var{regexp}
6682 @itemx -rfu @var{regexp}
6683 @cindex skipping over functions via regular expressions
6684 Functions whose name matches @var{regexp} will be skipped over when stepping.
6686 This form is useful for complex function names.
6687 For example, there is generally no need to step into C@t{++} @code{std::string}
6688 constructors or destructors. Plus with C@t{++} templates it can be hard to
6689 write out the full name of the function, and often it doesn't matter what
6690 the template arguments are. Specifying the function to be skipped as a
6691 regular expression makes this easier.
6694 (@value{GDBP}) skip -rfu ^std::(allocator|basic_string)<.*>::~?\1 *\(
6697 If you want to skip every templated C@t{++} constructor and destructor
6698 in the @code{std} namespace you can do:
6701 (@value{GDBP}) skip -rfu ^std::([a-zA-z0-9_]+)<.*>::~?\1 *\(
6705 If no options are specified, the function you're currently debugging
6708 @kindex skip function
6709 @item skip function @r{[}@var{linespec}@r{]}
6710 After running this command, the function named by @var{linespec} or the
6711 function containing the line named by @var{linespec} will be skipped over when
6712 stepping. @xref{Location Specifications}.
6714 If you do not specify @var{linespec}, the function you're currently debugging
6717 (If you have a function called @code{file} that you want to skip, use
6718 @kbd{skip function file}.)
6721 @item skip file @r{[}@var{filename}@r{]}
6722 After running this command, any function whose source lives in @var{filename}
6723 will be skipped over when stepping.
6726 (@value{GDBP}) skip file boring.c
6727 File boring.c will be skipped when stepping.
6730 If you do not specify @var{filename}, functions whose source lives in the file
6731 you're currently debugging will be skipped.
6734 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
6735 These are the commands for managing your list of skips:
6739 @item info skip @r{[}@var{range}@r{]}
6740 Print details about the specified skip(s). If @var{range} is not specified,
6741 print a table with details about all functions and files marked for skipping.
6742 @code{info skip} prints the following information about each skip:
6746 A number identifying this skip.
6747 @item Enabled or Disabled
6748 Enabled skips are marked with @samp{y}.
6749 Disabled skips are marked with @samp{n}.
6751 If the file name is a @samp{glob} pattern this is @samp{y}.
6752 Otherwise it is @samp{n}.
6754 The name or @samp{glob} pattern of the file to be skipped.
6755 If no file is specified this is @samp{<none>}.
6757 If the function name is a @samp{regular expression} this is @samp{y}.
6758 Otherwise it is @samp{n}.
6760 The name or regular expression of the function to skip.
6761 If no function is specified this is @samp{<none>}.
6765 @item skip delete @r{[}@var{range}@r{]}
6766 Delete the specified skip(s). If @var{range} is not specified, delete all
6770 @item skip enable @r{[}@var{range}@r{]}
6771 Enable the specified skip(s). If @var{range} is not specified, enable all
6774 @kindex skip disable
6775 @item skip disable @r{[}@var{range}@r{]}
6776 Disable the specified skip(s). If @var{range} is not specified, disable all
6779 @kindex set debug skip
6780 @item set debug skip @r{[}on|off@r{]}
6781 Set whether to print the debug output about skipping files and functions.
6783 @kindex show debug skip
6784 @item show debug skip
6785 Show whether the debug output about skipping files and functions is printed.
6793 A signal is an asynchronous event that can happen in a program. The
6794 operating system defines the possible kinds of signals, and gives each
6795 kind a name and a number. For example, in Unix @code{SIGINT} is the
6796 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
6797 @code{SIGSEGV} is the signal a program gets from referencing a place in
6798 memory far away from all the areas in use; @code{SIGALRM} occurs when
6799 the alarm clock timer goes off (which happens only if your program has
6800 requested an alarm).
6802 @cindex fatal signals
6803 Some signals, including @code{SIGALRM}, are a normal part of the
6804 functioning of your program. Others, such as @code{SIGSEGV}, indicate
6805 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
6806 program has not specified in advance some other way to handle the signal.
6807 @code{SIGINT} does not indicate an error in your program, but it is normally
6808 fatal so it can carry out the purpose of the interrupt: to kill the program.
6810 @value{GDBN} has the ability to detect any occurrence of a signal in your
6811 program. You can tell @value{GDBN} in advance what to do for each kind of
6814 @cindex handling signals
6815 Normally, @value{GDBN} is set up to let the non-erroneous signals like
6816 @code{SIGALRM} be silently passed to your program
6817 (so as not to interfere with their role in the program's functioning)
6818 but to stop your program immediately whenever an error signal happens.
6819 You can change these settings with the @code{handle} command.
6822 @kindex info signals
6826 Print a table of all the kinds of signals and how @value{GDBN} has been told to
6827 handle each one. You can use this to see the signal numbers of all
6828 the defined types of signals.
6830 @item info signals @var{sig}
6831 Similar, but print information only about the specified signal number.
6833 @code{info handle} is an alias for @code{info signals}.
6835 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
6836 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
6837 for details about this command.
6840 @item handle @var{signal} @r{[} @var{signal} @dots{} @r{]} @r{[}@var{keywords}@dots{}@r{]}
6841 Change the way @value{GDBN} handles each @var{signal}. Each
6842 @var{signal} can be the number of a signal or its name (with or
6843 without the @samp{SIG} at the beginning); a list of signal numbers of
6844 the form @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning
6845 all the known signals, except @code{SIGINT} and @code{SIGTRAP}, which
6846 are used by @value{GDBN}. Optional argument @var{keywords}, described
6847 below, say what changes to make to all of the specified signals.
6851 The keywords allowed by the @code{handle} command can be abbreviated.
6852 Their full names are:
6856 @value{GDBN} should not stop your program when this signal happens. It may
6857 still print a message telling you that the signal has come in.
6860 @value{GDBN} should stop your program when this signal happens. This implies
6861 the @code{print} keyword as well.
6864 @value{GDBN} should print a message when this signal happens.
6867 @value{GDBN} should not mention the occurrence of the signal at all. This
6868 implies the @code{nostop} keyword as well.
6872 @value{GDBN} should allow your program to see this signal; your program
6873 can handle the signal, or else it may terminate if the signal is fatal
6874 and not handled. @code{pass} and @code{noignore} are synonyms.
6878 @value{GDBN} should not allow your program to see this signal.
6879 @code{nopass} and @code{ignore} are synonyms.
6883 When a signal stops your program, the signal is not visible to the
6885 continue. Your program sees the signal then, if @code{pass} is in
6886 effect for the signal in question @emph{at that time}. In other words,
6887 after @value{GDBN} reports a signal, you can use the @code{handle}
6888 command with @code{pass} or @code{nopass} to control whether your
6889 program sees that signal when you continue.
6891 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
6892 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
6893 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
6896 You can also use the @code{signal} command to prevent your program from
6897 seeing a signal, or cause it to see a signal it normally would not see,
6898 or to give it any signal at any time. For example, if your program stopped
6899 due to some sort of memory reference error, you might store correct
6900 values into the erroneous variables and continue, hoping to see more
6901 execution; but your program would probably terminate immediately as
6902 a result of the fatal signal once it saw the signal. To prevent this,
6903 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
6906 @cindex stepping and signal handlers
6907 @anchor{stepping and signal handlers}
6909 @value{GDBN} optimizes for stepping the mainline code. If a signal
6910 that has @code{handle nostop} and @code{handle pass} set arrives while
6911 a stepping command (e.g., @code{stepi}, @code{step}, @code{next}) is
6912 in progress, @value{GDBN} lets the signal handler run and then resumes
6913 stepping the mainline code once the signal handler returns. In other
6914 words, @value{GDBN} steps over the signal handler. This prevents
6915 signals that you've specified as not interesting (with @code{handle
6916 nostop}) from changing the focus of debugging unexpectedly. Note that
6917 the signal handler itself may still hit a breakpoint, stop for another
6918 signal that has @code{handle stop} in effect, or for any other event
6919 that normally results in stopping the stepping command sooner. Also
6920 note that @value{GDBN} still informs you that the program received a
6921 signal if @code{handle print} is set.
6923 @anchor{stepping into signal handlers}
6925 If you set @code{handle pass} for a signal, and your program sets up a
6926 handler for it, then issuing a stepping command, such as @code{step}
6927 or @code{stepi}, when your program is stopped due to the signal will
6928 step @emph{into} the signal handler (if the target supports that).
6930 Likewise, if you use the @code{queue-signal} command to queue a signal
6931 to be delivered to the current thread when execution of the thread
6932 resumes (@pxref{Signaling, ,Giving your Program a Signal}), then a
6933 stepping command will step into the signal handler.
6935 Here's an example, using @code{stepi} to step to the first instruction
6936 of @code{SIGUSR1}'s handler:
6939 (@value{GDBP}) handle SIGUSR1
6940 Signal Stop Print Pass to program Description
6941 SIGUSR1 Yes Yes Yes User defined signal 1
6945 Program received signal SIGUSR1, User defined signal 1.
6946 main () sigusr1.c:28
6949 sigusr1_handler () at sigusr1.c:9
6953 The same, but using @code{queue-signal} instead of waiting for the
6954 program to receive the signal first:
6959 (@value{GDBP}) queue-signal SIGUSR1
6961 sigusr1_handler () at sigusr1.c:9
6966 @cindex extra signal information
6967 @anchor{extra signal information}
6969 On some targets, @value{GDBN} can inspect extra signal information
6970 associated with the intercepted signal, before it is actually
6971 delivered to the program being debugged. This information is exported
6972 by the convenience variable @code{$_siginfo}, and consists of data
6973 that is passed by the kernel to the signal handler at the time of the
6974 receipt of a signal. The data type of the information itself is
6975 target dependent. You can see the data type using the @code{ptype
6976 $_siginfo} command. On Unix systems, it typically corresponds to the
6977 standard @code{siginfo_t} type, as defined in the @file{signal.h}
6980 Here's an example, on a @sc{gnu}/Linux system, printing the stray
6981 referenced address that raised a segmentation fault.
6985 (@value{GDBP}) continue
6986 Program received signal SIGSEGV, Segmentation fault.
6987 0x0000000000400766 in main ()
6989 (@value{GDBP}) ptype $_siginfo
6996 struct @{...@} _kill;
6997 struct @{...@} _timer;
6999 struct @{...@} _sigchld;
7000 struct @{...@} _sigfault;
7001 struct @{...@} _sigpoll;
7004 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
7008 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
7009 $1 = (void *) 0x7ffff7ff7000
7013 Depending on target support, @code{$_siginfo} may also be writable.
7015 @cindex Intel MPX boundary violations
7016 @cindex boundary violations, Intel MPX
7017 On some targets, a @code{SIGSEGV} can be caused by a boundary
7018 violation, i.e., accessing an address outside of the allowed range.
7019 In those cases @value{GDBN} may displays additional information,
7020 depending on how @value{GDBN} has been told to handle the signal.
7021 With @code{handle stop SIGSEGV}, @value{GDBN} displays the violation
7022 kind: "Upper" or "Lower", the memory address accessed and the
7023 bounds, while with @code{handle nostop SIGSEGV} no additional
7024 information is displayed.
7026 The usual output of a segfault is:
7028 Program received signal SIGSEGV, Segmentation fault
7029 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
7030 68 value = *(p + len);
7033 While a bound violation is presented as:
7035 Program received signal SIGSEGV, Segmentation fault
7036 Upper bound violation while accessing address 0x7fffffffc3b3
7037 Bounds: [lower = 0x7fffffffc390, upper = 0x7fffffffc3a3]
7038 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
7039 68 value = *(p + len);
7043 @section Stopping and Starting Multi-thread Programs
7045 @cindex stopped threads
7046 @cindex threads, stopped
7048 @cindex continuing threads
7049 @cindex threads, continuing
7051 @value{GDBN} supports debugging programs with multiple threads
7052 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
7053 are two modes of controlling execution of your program within the
7054 debugger. In the default mode, referred to as @dfn{all-stop mode},
7055 when any thread in your program stops (for example, at a breakpoint
7056 or while being stepped), all other threads in the program are also stopped by
7057 @value{GDBN}. On some targets, @value{GDBN} also supports
7058 @dfn{non-stop mode}, in which other threads can continue to run freely while
7059 you examine the stopped thread in the debugger.
7062 * All-Stop Mode:: All threads stop when GDB takes control
7063 * Non-Stop Mode:: Other threads continue to execute
7064 * Background Execution:: Running your program asynchronously
7065 * Thread-Specific Breakpoints:: Controlling breakpoints
7066 * Interrupted System Calls:: GDB may interfere with system calls
7067 * Observer Mode:: GDB does not alter program behavior
7071 @subsection All-Stop Mode
7073 @cindex all-stop mode
7075 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
7076 @emph{all} threads of execution stop, not just the current thread. This
7077 allows you to examine the overall state of the program, including
7078 switching between threads, without worrying that things may change
7081 Conversely, whenever you restart the program, @emph{all} threads start
7082 executing. @emph{This is true even when single-stepping} with commands
7083 like @code{step} or @code{next}.
7085 In particular, @value{GDBN} cannot single-step all threads in lockstep.
7086 Since thread scheduling is up to your debugging target's operating
7087 system (not controlled by @value{GDBN}), other threads may
7088 execute more than one statement while the current thread completes a
7089 single step. Moreover, in general other threads stop in the middle of a
7090 statement, rather than at a clean statement boundary, when the program
7093 You might even find your program stopped in another thread after
7094 continuing or even single-stepping. This happens whenever some other
7095 thread runs into a breakpoint, a signal, or an exception before the
7096 first thread completes whatever you requested.
7098 @cindex automatic thread selection
7099 @cindex switching threads automatically
7100 @cindex threads, automatic switching
7101 Whenever @value{GDBN} stops your program, due to a breakpoint or a
7102 signal, it automatically selects the thread where that breakpoint or
7103 signal happened. @value{GDBN} alerts you to the context switch with a
7104 message such as @samp{[Switching to Thread @var{n}]} to identify the
7107 @anchor{set scheduler-locking}
7109 On some OSes, you can modify @value{GDBN}'s default behavior by
7110 locking the OS scheduler to allow only a single thread to run.
7113 @item set scheduler-locking @var{mode}
7114 @cindex scheduler-locking
7115 @cindex scheduler locking mode
7116 @cindex lock scheduler
7117 Set the scheduler locking mode. It applies to normal execution,
7118 record mode, and replay mode. @var{mode} can be one of
7123 There is no locking and any thread may run at any time.
7126 Only the current thread may run when the inferior is resumed.
7129 Behaves like @code{on} when stepping, and @code{off} otherwise.
7130 Threads other than the current never get a chance to run when you
7131 step, and they are completely free to run when you use commands like
7132 @samp{continue}, @samp{until}, or @samp{finish}.
7134 This mode optimizes for single-stepping; it prevents other threads
7135 from preempting the current thread while you are stepping, so that the
7136 focus of debugging does not change unexpectedly. However, unless
7137 another thread hits a breakpoint during its timeslice, @value{GDBN}
7138 does not change the current thread away from the thread that you are
7142 Behaves like @code{on} in replay mode, and @code{off} in either record
7143 mode or during normal execution. This is the default mode.
7146 @item show scheduler-locking
7147 Display the current scheduler locking mode.
7150 @cindex resume threads of multiple processes simultaneously
7151 By default, when you issue one of the execution commands such as
7152 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
7153 threads of the current inferior to run. For example, if @value{GDBN}
7154 is attached to two inferiors, each with two threads, the
7155 @code{continue} command resumes only the two threads of the current
7156 inferior. This is useful, for example, when you debug a program that
7157 forks and you want to hold the parent stopped (so that, for instance,
7158 it doesn't run to exit), while you debug the child. In other
7159 situations, you may not be interested in inspecting the current state
7160 of any of the processes @value{GDBN} is attached to, and you may want
7161 to resume them all until some breakpoint is hit. In the latter case,
7162 you can instruct @value{GDBN} to allow all threads of all the
7163 inferiors to run with the @w{@code{set schedule-multiple}} command.
7166 @kindex set schedule-multiple
7167 @item set schedule-multiple
7168 Set the mode for allowing threads of multiple processes to be resumed
7169 when an execution command is issued. When @code{on}, all threads of
7170 all processes are allowed to run. When @code{off}, only the threads
7171 of the current process are resumed. The default is @code{off}. The
7172 @code{scheduler-locking} mode takes precedence when set to @code{on},
7173 or while you are stepping and set to @code{step}.
7175 @item show schedule-multiple
7176 Display the current mode for resuming the execution of threads of
7181 @subsection Non-Stop Mode
7183 @cindex non-stop mode
7185 @c This section is really only a place-holder, and needs to be expanded
7186 @c with more details.
7188 For some multi-threaded targets, @value{GDBN} supports an optional
7189 mode of operation in which you can examine stopped program threads in
7190 the debugger while other threads continue to execute freely. This
7191 minimizes intrusion when debugging live systems, such as programs
7192 where some threads have real-time constraints or must continue to
7193 respond to external events. This is referred to as @dfn{non-stop} mode.
7195 In non-stop mode, when a thread stops to report a debugging event,
7196 @emph{only} that thread is stopped; @value{GDBN} does not stop other
7197 threads as well, in contrast to the all-stop mode behavior. Additionally,
7198 execution commands such as @code{continue} and @code{step} apply by default
7199 only to the current thread in non-stop mode, rather than all threads as
7200 in all-stop mode. This allows you to control threads explicitly in
7201 ways that are not possible in all-stop mode --- for example, stepping
7202 one thread while allowing others to run freely, stepping
7203 one thread while holding all others stopped, or stepping several threads
7204 independently and simultaneously.
7206 To enter non-stop mode, use this sequence of commands before you run
7207 or attach to your program:
7210 # If using the CLI, pagination breaks non-stop.
7213 # Finally, turn it on!
7217 You can use these commands to manipulate the non-stop mode setting:
7220 @kindex set non-stop
7221 @item set non-stop on
7222 Enable selection of non-stop mode.
7223 @item set non-stop off
7224 Disable selection of non-stop mode.
7225 @kindex show non-stop
7227 Show the current non-stop enablement setting.
7230 Note these commands only reflect whether non-stop mode is enabled,
7231 not whether the currently-executing program is being run in non-stop mode.
7232 In particular, the @code{set non-stop} preference is only consulted when
7233 @value{GDBN} starts or connects to the target program, and it is generally
7234 not possible to switch modes once debugging has started. Furthermore,
7235 since not all targets support non-stop mode, even when you have enabled
7236 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
7239 In non-stop mode, all execution commands apply only to the current thread
7240 by default. That is, @code{continue} only continues one thread.
7241 To continue all threads, issue @code{continue -a} or @code{c -a}.
7243 You can use @value{GDBN}'s background execution commands
7244 (@pxref{Background Execution}) to run some threads in the background
7245 while you continue to examine or step others from @value{GDBN}.
7246 The MI execution commands (@pxref{GDB/MI Program Execution}) are
7247 always executed asynchronously in non-stop mode.
7249 Suspending execution is done with the @code{interrupt} command when
7250 running in the background, or @kbd{Ctrl-c} during foreground execution.
7251 In all-stop mode, this stops the whole process;
7252 but in non-stop mode the interrupt applies only to the current thread.
7253 To stop the whole program, use @code{interrupt -a}.
7255 Other execution commands do not currently support the @code{-a} option.
7257 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
7258 that thread current, as it does in all-stop mode. This is because the
7259 thread stop notifications are asynchronous with respect to @value{GDBN}'s
7260 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
7261 changed to a different thread just as you entered a command to operate on the
7262 previously current thread.
7264 @node Background Execution
7265 @subsection Background Execution
7267 @cindex foreground execution
7268 @cindex background execution
7269 @cindex asynchronous execution
7270 @cindex execution, foreground, background and asynchronous
7272 @value{GDBN}'s execution commands have two variants: the normal
7273 foreground (synchronous) behavior, and a background
7274 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
7275 the program to report that some thread has stopped before prompting for
7276 another command. In background execution, @value{GDBN} immediately gives
7277 a command prompt so that you can issue other commands while your program runs.
7279 If the target doesn't support async mode, @value{GDBN} issues an error
7280 message if you attempt to use the background execution commands.
7282 @cindex @code{&}, background execution of commands
7283 To specify background execution, add a @code{&} to the command. For example,
7284 the background form of the @code{continue} command is @code{continue&}, or
7285 just @code{c&}. The execution commands that accept background execution
7291 @xref{Starting, , Starting your Program}.
7295 @xref{Attach, , Debugging an Already-running Process}.
7299 @xref{Continuing and Stepping, step}.
7303 @xref{Continuing and Stepping, stepi}.
7307 @xref{Continuing and Stepping, next}.
7311 @xref{Continuing and Stepping, nexti}.
7315 @xref{Continuing and Stepping, continue}.
7319 @xref{Continuing and Stepping, finish}.
7323 @xref{Continuing and Stepping, until}.
7327 Background execution is especially useful in conjunction with non-stop
7328 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
7329 However, you can also use these commands in the normal all-stop mode with
7330 the restriction that you cannot issue another execution command until the
7331 previous one finishes. Examples of commands that are valid in all-stop
7332 mode while the program is running include @code{help} and @code{info break}.
7334 You can interrupt your program while it is running in the background by
7335 using the @code{interrupt} command.
7342 Suspend execution of the running program. In all-stop mode,
7343 @code{interrupt} stops the whole process, but in non-stop mode, it stops
7344 only the current thread. To stop the whole program in non-stop mode,
7345 use @code{interrupt -a}.
7348 @node Thread-Specific Breakpoints
7349 @subsection Thread-Specific Breakpoints
7351 When your program has multiple threads (@pxref{Threads,, Debugging
7352 Programs with Multiple Threads}), you can choose whether to set
7353 breakpoints on all threads, or on a particular thread.
7356 @cindex breakpoints and threads
7357 @cindex thread breakpoints
7358 @kindex break @dots{} thread @var{thread-id}
7359 @item break @var{locspec} thread @var{thread-id}
7360 @itemx break @var{locspec} thread @var{thread-id} if @dots{}
7361 @var{locspec} specifies a code location or locations in your program.
7362 @xref{Location Specifications}, for details.
7364 Use the qualifier @samp{thread @var{thread-id}} with a breakpoint command
7365 to specify that you only want @value{GDBN} to stop the program when a
7366 particular thread reaches this breakpoint. The @var{thread-id} specifier
7367 is one of the thread identifiers assigned by @value{GDBN}, shown
7368 in the first column of the @samp{info threads} display.
7370 If you do not specify @samp{thread @var{thread-id}} when you set a
7371 breakpoint, the breakpoint applies to @emph{all} threads of your
7374 You can use the @code{thread} qualifier on conditional breakpoints as
7375 well; in this case, place @samp{thread @var{thread-id}} before or
7376 after the breakpoint condition, like this:
7379 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
7384 Thread-specific breakpoints are automatically deleted when
7385 @value{GDBN} detects the corresponding thread is no longer in the
7386 thread list. For example:
7390 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
7393 There are several ways for a thread to disappear, such as a regular
7394 thread exit, but also when you detach from the process with the
7395 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
7396 Process}), or if @value{GDBN} loses the remote connection
7397 (@pxref{Remote Debugging}), etc. Note that with some targets,
7398 @value{GDBN} is only able to detect a thread has exited when the user
7399 explicitly asks for the thread list with the @code{info threads}
7402 A breakpoint can't be both thread-specific and inferior-specific
7403 (@pxref{Inferior-Specific Breakpoints}), or task-specific (@pxref{Ada
7404 Tasks}); using more than one of the @code{thread}, @code{inferior}, or
7405 @code{task} keywords when creating a breakpoint will give an error.
7407 @node Interrupted System Calls
7408 @subsection Interrupted System Calls
7410 @cindex thread breakpoints and system calls
7411 @cindex system calls and thread breakpoints
7412 @cindex premature return from system calls
7413 There is an unfortunate side effect when using @value{GDBN} to debug
7414 multi-threaded programs. If one thread stops for a
7415 breakpoint, or for some other reason, and another thread is blocked in a
7416 system call, then the system call may return prematurely. This is a
7417 consequence of the interaction between multiple threads and the signals
7418 that @value{GDBN} uses to implement breakpoints and other events that
7421 To handle this problem, your program should check the return value of
7422 each system call and react appropriately. This is good programming
7425 For example, do not write code like this:
7431 The call to @code{sleep} will return early if a different thread stops
7432 at a breakpoint or for some other reason.
7434 Instead, write this:
7439 unslept = sleep (unslept);
7442 A system call is allowed to return early, so the system is still
7443 conforming to its specification. But @value{GDBN} does cause your
7444 multi-threaded program to behave differently than it would without
7447 Also, @value{GDBN} uses internal breakpoints in the thread library to
7448 monitor certain events such as thread creation and thread destruction.
7449 When such an event happens, a system call in another thread may return
7450 prematurely, even though your program does not appear to stop.
7453 @subsection Observer Mode
7455 If you want to build on non-stop mode and observe program behavior
7456 without any chance of disruption by @value{GDBN}, you can set
7457 variables to disable all of the debugger's attempts to modify state,
7458 whether by writing memory, inserting breakpoints, etc. These operate
7459 at a low level, intercepting operations from all commands.
7461 When all of these are set to @code{off}, then @value{GDBN} is said to
7462 be @dfn{observer mode}. As a convenience, the variable
7463 @code{observer} can be set to disable these, plus enable non-stop
7466 Note that @value{GDBN} will not prevent you from making nonsensical
7467 combinations of these settings. For instance, if you have enabled
7468 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
7469 then breakpoints that work by writing trap instructions into the code
7470 stream will still not be able to be placed.
7475 @item set observer on
7476 @itemx set observer off
7477 When set to @code{on}, this disables all the permission variables
7478 below (except for @code{insert-fast-tracepoints}), plus enables
7479 non-stop debugging. Setting this to @code{off} switches back to
7480 normal debugging, though remaining in non-stop mode.
7483 Show whether observer mode is on or off.
7485 @kindex may-write-registers
7486 @item set may-write-registers on
7487 @itemx set may-write-registers off
7488 This controls whether @value{GDBN} will attempt to alter the values of
7489 registers, such as with assignment expressions in @code{print}, or the
7490 @code{jump} command. It defaults to @code{on}.
7492 @item show may-write-registers
7493 Show the current permission to write registers.
7495 @kindex may-write-memory
7496 @item set may-write-memory on
7497 @itemx set may-write-memory off
7498 This controls whether @value{GDBN} will attempt to alter the contents
7499 of memory, such as with assignment expressions in @code{print}. It
7500 defaults to @code{on}.
7502 @item show may-write-memory
7503 Show the current permission to write memory.
7505 @kindex may-insert-breakpoints
7506 @item set may-insert-breakpoints on
7507 @itemx set may-insert-breakpoints off
7508 This controls whether @value{GDBN} will attempt to insert breakpoints.
7509 This affects all breakpoints, including internal breakpoints defined
7510 by @value{GDBN}. It defaults to @code{on}.
7512 @item show may-insert-breakpoints
7513 Show the current permission to insert breakpoints.
7515 @kindex may-insert-tracepoints
7516 @item set may-insert-tracepoints on
7517 @itemx set may-insert-tracepoints off
7518 This controls whether @value{GDBN} will attempt to insert (regular)
7519 tracepoints at the beginning of a tracing experiment. It affects only
7520 non-fast tracepoints, fast tracepoints being under the control of
7521 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
7523 @item show may-insert-tracepoints
7524 Show the current permission to insert tracepoints.
7526 @kindex may-insert-fast-tracepoints
7527 @item set may-insert-fast-tracepoints on
7528 @itemx set may-insert-fast-tracepoints off
7529 This controls whether @value{GDBN} will attempt to insert fast
7530 tracepoints at the beginning of a tracing experiment. It affects only
7531 fast tracepoints, regular (non-fast) tracepoints being under the
7532 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
7534 @item show may-insert-fast-tracepoints
7535 Show the current permission to insert fast tracepoints.
7537 @kindex may-interrupt
7538 @item set may-interrupt on
7539 @itemx set may-interrupt off
7540 This controls whether @value{GDBN} will attempt to interrupt or stop
7541 program execution. When this variable is @code{off}, the
7542 @code{interrupt} command will have no effect, nor will
7543 @kbd{Ctrl-c}. It defaults to @code{on}.
7545 @item show may-interrupt
7546 Show the current permission to interrupt or stop the program.
7550 @node Reverse Execution
7551 @chapter Running programs backward
7552 @cindex reverse execution
7553 @cindex running programs backward
7555 When you are debugging a program, it is not unusual to realize that
7556 you have gone too far, and some event of interest has already happened.
7557 If the target environment supports it, @value{GDBN} can allow you to
7558 ``rewind'' the program by running it backward.
7560 A target environment that supports reverse execution should be able
7561 to ``undo'' the changes in machine state that have taken place as the
7562 program was executing normally. Variables, registers etc.@: should
7563 revert to their previous values. Obviously this requires a great
7564 deal of sophistication on the part of the target environment; not
7565 all target environments can support reverse execution.
7567 When a program is executed in reverse, the instructions that
7568 have most recently been executed are ``un-executed'', in reverse
7569 order. The program counter runs backward, following the previous
7570 thread of execution in reverse. As each instruction is ``un-executed'',
7571 the values of memory and/or registers that were changed by that
7572 instruction are reverted to their previous states. After executing
7573 a piece of source code in reverse, all side effects of that code
7574 should be ``undone'', and all variables should be returned to their
7575 prior values@footnote{
7576 Note that some side effects are easier to undo than others. For instance,
7577 memory and registers are relatively easy, but device I/O is hard. Some
7578 targets may be able undo things like device I/O, and some may not.
7580 The contract between @value{GDBN} and the reverse executing target
7581 requires only that the target do something reasonable when
7582 @value{GDBN} tells it to execute backwards, and then report the
7583 results back to @value{GDBN}. Whatever the target reports back to
7584 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
7585 assumes that the memory and registers that the target reports are in a
7586 consistent state, but @value{GDBN} accepts whatever it is given.
7589 On some platforms, @value{GDBN} has built-in support for reverse
7590 execution, activated with the @code{record} or @code{record btrace}
7591 commands. @xref{Process Record and Replay}. Some remote targets,
7592 typically full system emulators, support reverse execution directly
7593 without requiring any special command.
7595 If you are debugging in a target environment that supports
7596 reverse execution, @value{GDBN} provides the following commands.
7599 @kindex reverse-continue
7600 @kindex rc @r{(@code{reverse-continue})}
7601 @item reverse-continue @r{[}@var{ignore-count}@r{]}
7602 @itemx rc @r{[}@var{ignore-count}@r{]}
7603 Beginning at the point where your program last stopped, start executing
7604 in reverse. Reverse execution will stop for breakpoints and synchronous
7605 exceptions (signals), just like normal execution. Behavior of
7606 asynchronous signals depends on the target environment.
7608 @kindex reverse-step
7609 @kindex rs @r{(@code{step})}
7610 @item reverse-step @r{[}@var{count}@r{]}
7611 Run the program backward until control reaches the start of a
7612 different source line; then stop it, and return control to @value{GDBN}.
7614 Like the @code{step} command, @code{reverse-step} will only stop
7615 at the beginning of a source line. It ``un-executes'' the previously
7616 executed source line. If the previous source line included calls to
7617 debuggable functions, @code{reverse-step} will step (backward) into
7618 the called function, stopping at the beginning of the @emph{last}
7619 statement in the called function (typically a return statement).
7621 Also, as with the @code{step} command, if non-debuggable functions are
7622 called, @code{reverse-step} will run thru them backward without stopping.
7624 @kindex reverse-stepi
7625 @kindex rsi @r{(@code{reverse-stepi})}
7626 @item reverse-stepi @r{[}@var{count}@r{]}
7627 Reverse-execute one machine instruction. Note that the instruction
7628 to be reverse-executed is @emph{not} the one pointed to by the program
7629 counter, but the instruction executed prior to that one. For instance,
7630 if the last instruction was a jump, @code{reverse-stepi} will take you
7631 back from the destination of the jump to the jump instruction itself.
7633 @kindex reverse-next
7634 @kindex rn @r{(@code{reverse-next})}
7635 @item reverse-next @r{[}@var{count}@r{]}
7636 Run backward to the beginning of the previous line executed in
7637 the current (innermost) stack frame. If the line contains function
7638 calls, they will be ``un-executed'' without stopping. Starting from
7639 the first line of a function, @code{reverse-next} will take you back
7640 to the caller of that function, @emph{before} the function was called,
7641 just as the normal @code{next} command would take you from the last
7642 line of a function back to its return to its caller
7643 @footnote{Unless the code is too heavily optimized.}.
7645 @kindex reverse-nexti
7646 @kindex rni @r{(@code{reverse-nexti})}
7647 @item reverse-nexti @r{[}@var{count}@r{]}
7648 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
7649 in reverse, except that called functions are ``un-executed'' atomically.
7650 That is, if the previously executed instruction was a return from
7651 another function, @code{reverse-nexti} will continue to execute
7652 in reverse until the call to that function (from the current stack
7655 @kindex reverse-finish
7656 @item reverse-finish
7657 Just as the @code{finish} command takes you to the point where the
7658 current function returns, @code{reverse-finish} takes you to the point
7659 where it was called. Instead of ending up at the end of the current
7660 function invocation, you end up at the beginning.
7662 @kindex set exec-direction
7663 @item set exec-direction
7664 Set the direction of target execution.
7665 @item set exec-direction reverse
7666 @cindex execute forward or backward in time
7667 @value{GDBN} will perform all execution commands in reverse, until the
7668 exec-direction mode is changed to ``forward''. Affected commands include
7669 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
7670 command cannot be used in reverse mode.
7671 @item set exec-direction forward
7672 @value{GDBN} will perform all execution commands in the normal fashion.
7673 This is the default.
7677 @node Process Record and Replay
7678 @chapter Recording Inferior's Execution and Replaying It
7679 @cindex process record and replay
7680 @cindex recording inferior's execution and replaying it
7682 On some platforms, @value{GDBN} provides a special @dfn{process record
7683 and replay} target that can record a log of the process execution, and
7684 replay it later with both forward and reverse execution commands.
7687 When this target is in use, if the execution log includes the record
7688 for the next instruction, @value{GDBN} will debug in @dfn{replay
7689 mode}. In the replay mode, the inferior does not really execute code
7690 instructions. Instead, all the events that normally happen during
7691 code execution are taken from the execution log. While code is not
7692 really executed in replay mode, the values of registers (including the
7693 program counter register) and the memory of the inferior are still
7694 changed as they normally would. Their contents are taken from the
7698 If the record for the next instruction is not in the execution log,
7699 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
7700 inferior executes normally, and @value{GDBN} records the execution log
7703 The process record and replay target supports reverse execution
7704 (@pxref{Reverse Execution}), even if the platform on which the
7705 inferior runs does not. However, the reverse execution is limited in
7706 this case by the range of the instructions recorded in the execution
7707 log. In other words, reverse execution on platforms that don't
7708 support it directly can only be done in the replay mode.
7710 When debugging in the reverse direction, @value{GDBN} will work in
7711 replay mode as long as the execution log includes the record for the
7712 previous instruction; otherwise, it will work in record mode, if the
7713 platform supports reverse execution, or stop if not.
7715 Currently, process record and replay is supported on ARM, Aarch64,
7716 Moxie, PowerPC, PowerPC64, S/390, and x86 (i386/amd64) running
7717 GNU/Linux. Process record and replay can be used both when native
7718 debugging, and when remote debugging via @code{gdbserver}.
7720 For architecture environments that support process record and replay,
7721 @value{GDBN} provides the following commands:
7724 @kindex target record
7725 @kindex target record-full
7726 @kindex target record-btrace
7729 @kindex record btrace
7730 @kindex record btrace bts
7731 @kindex record btrace pt
7737 @kindex rec btrace bts
7738 @kindex rec btrace pt
7741 @item record @var{method}
7742 This command starts the process record and replay target. The
7743 recording method can be specified as parameter. Without a parameter
7744 the command uses the @code{full} recording method. The following
7745 recording methods are available:
7749 Full record/replay recording using @value{GDBN}'s software record and
7750 replay implementation. This method allows replaying and reverse
7753 @item btrace @var{format}
7754 Hardware-supported instruction recording, supported on Intel
7755 processors. This method does not record data. Further, the data is
7756 collected in a ring buffer so old data will be overwritten when the
7757 buffer is full. It allows limited reverse execution. Variables and
7758 registers are not available during reverse execution. In remote
7759 debugging, recording continues on disconnect. Recorded data can be
7760 inspected after reconnecting. The recording may be stopped using
7763 The recording format can be specified as parameter. Without a parameter
7764 the command chooses the recording format. The following recording
7765 formats are available:
7769 @cindex branch trace store
7770 Use the @dfn{Branch Trace Store} (@acronym{BTS}) recording format. In
7771 this format, the processor stores a from/to record for each executed
7772 branch in the btrace ring buffer.
7775 @cindex Intel Processor Trace
7776 Use the @dfn{Intel Processor Trace} recording format. In this
7777 format, the processor stores the execution trace in a compressed form
7778 that is afterwards decoded by @value{GDBN}.
7780 The trace can be recorded with very low overhead. The compressed
7781 trace format also allows small trace buffers to already contain a big
7782 number of instructions compared to @acronym{BTS}.
7784 Decoding the recorded execution trace, on the other hand, is more
7785 expensive than decoding @acronym{BTS} trace. This is mostly due to the
7786 increased number of instructions to process. You should increase the
7787 buffer-size with care.
7790 Not all recording formats may be available on all processors.
7793 The process record and replay target can only debug a process that is
7794 already running. Therefore, you need first to start the process with
7795 the @kbd{run} or @kbd{start} commands, and then start the recording
7796 with the @kbd{record @var{method}} command.
7798 @cindex displaced stepping, and process record and replay
7799 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
7800 will be automatically disabled when process record and replay target
7801 is started. That's because the process record and replay target
7802 doesn't support displaced stepping.
7804 @cindex non-stop mode, and process record and replay
7805 @cindex asynchronous execution, and process record and replay
7806 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
7807 the asynchronous execution mode (@pxref{Background Execution}), not
7808 all recording methods are available. The @code{full} recording method
7809 does not support these two modes.
7814 Stop the process record and replay target. When process record and
7815 replay target stops, the entire execution log will be deleted and the
7816 inferior will either be terminated, or will remain in its final state.
7818 When you stop the process record and replay target in record mode (at
7819 the end of the execution log), the inferior will be stopped at the
7820 next instruction that would have been recorded. In other words, if
7821 you record for a while and then stop recording, the inferior process
7822 will be left in the same state as if the recording never happened.
7824 On the other hand, if the process record and replay target is stopped
7825 while in replay mode (that is, not at the end of the execution log,
7826 but at some earlier point), the inferior process will become ``live''
7827 at that earlier state, and it will then be possible to continue the
7828 usual ``live'' debugging of the process from that state.
7830 When the inferior process exits, or @value{GDBN} detaches from it,
7831 process record and replay target will automatically stop itself.
7835 Go to a specific location in the execution log. There are several
7836 ways to specify the location to go to:
7839 @item record goto begin
7840 @itemx record goto start
7841 Go to the beginning of the execution log.
7843 @item record goto end
7844 Go to the end of the execution log.
7846 @item record goto @var{n}
7847 Go to instruction number @var{n} in the execution log.
7851 @item record save @var{filename}
7852 Save the execution log to a file @file{@var{filename}}.
7853 Default filename is @file{gdb_record.@var{process_id}}, where
7854 @var{process_id} is the process ID of the inferior.
7856 This command may not be available for all recording methods.
7858 @kindex record restore
7859 @item record restore @var{filename}
7860 Restore the execution log from a file @file{@var{filename}}.
7861 File must have been created with @code{record save}.
7863 @kindex set record full
7864 @item set record full insn-number-max @var{limit}
7865 @itemx set record full insn-number-max unlimited
7866 Set the limit of instructions to be recorded for the @code{full}
7867 recording method. Default value is 200000.
7869 If @var{limit} is a positive number, then @value{GDBN} will start
7870 deleting instructions from the log once the number of the record
7871 instructions becomes greater than @var{limit}. For every new recorded
7872 instruction, @value{GDBN} will delete the earliest recorded
7873 instruction to keep the number of recorded instructions at the limit.
7874 (Since deleting recorded instructions loses information, @value{GDBN}
7875 lets you control what happens when the limit is reached, by means of
7876 the @code{stop-at-limit} option, described below.)
7878 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
7879 delete recorded instructions from the execution log. The number of
7880 recorded instructions is limited only by the available memory.
7882 @kindex show record full
7883 @item show record full insn-number-max
7884 Show the limit of instructions to be recorded with the @code{full}
7887 @item set record full stop-at-limit
7888 Control the behavior of the @code{full} recording method when the
7889 number of recorded instructions reaches the limit. If ON (the
7890 default), @value{GDBN} will stop when the limit is reached for the
7891 first time and ask you whether you want to stop the inferior or
7892 continue running it and recording the execution log. If you decide
7893 to continue recording, each new recorded instruction will cause the
7894 oldest one to be deleted.
7896 If this option is OFF, @value{GDBN} will automatically delete the
7897 oldest record to make room for each new one, without asking.
7899 @item show record full stop-at-limit
7900 Show the current setting of @code{stop-at-limit}.
7902 @item set record full memory-query
7903 Control the behavior when @value{GDBN} is unable to record memory
7904 changes caused by an instruction for the @code{full} recording method.
7905 If ON, @value{GDBN} will query whether to stop the inferior in that
7908 If this option is OFF (the default), @value{GDBN} will automatically
7909 ignore the effect of such instructions on memory. Later, when
7910 @value{GDBN} replays this execution log, it will mark the log of this
7911 instruction as not accessible, and it will not affect the replay
7914 @item show record full memory-query
7915 Show the current setting of @code{memory-query}.
7917 @kindex set record btrace
7918 The @code{btrace} record target does not trace data. As a
7919 convenience, when replaying, @value{GDBN} reads read-only memory off
7920 the live program directly, assuming that the addresses of the
7921 read-only areas don't change. This for example makes it possible to
7922 disassemble code while replaying, but not to print variables.
7923 In some cases, being able to inspect variables might be useful.
7924 You can use the following command for that:
7926 @item set record btrace replay-memory-access
7927 Control the behavior of the @code{btrace} recording method when
7928 accessing memory during replay. If @code{read-only} (the default),
7929 @value{GDBN} will only allow accesses to read-only memory.
7930 If @code{read-write}, @value{GDBN} will allow accesses to read-only
7931 and to read-write memory. Beware that the accessed memory corresponds
7932 to the live target and not necessarily to the current replay
7935 @item set record btrace cpu @var{identifier}
7936 Set the processor to be used for enabling workarounds for processor
7937 errata when decoding the trace.
7939 Processor errata are defects in processor operation, caused by its
7940 design or manufacture. They can cause a trace not to match the
7941 specification. This, in turn, may cause trace decode to fail.
7942 @value{GDBN} can detect erroneous trace packets and correct them, thus
7943 avoiding the decoding failures. These corrections are known as
7944 @dfn{errata workarounds}, and are enabled based on the processor on
7945 which the trace was recorded.
7947 By default, @value{GDBN} attempts to detect the processor
7948 automatically, and apply the necessary workarounds for it. However,
7949 you may need to specify the processor if @value{GDBN} does not yet
7950 support it. This command allows you to do that, and also allows to
7951 disable the workarounds.
7953 The argument @var{identifier} identifies the @sc{cpu} and is of the
7954 form: @code{@var{vendor}:@var{processor identifier}}. In addition,
7955 there are two special identifiers, @code{none} and @code{auto}
7958 The following vendor identifiers and corresponding processor
7959 identifiers are currently supported:
7961 @multitable @columnfractions .1 .9
7964 @tab @var{family}/@var{model}[/@var{stepping}]
7968 On GNU/Linux systems, the processor @var{family}, @var{model}, and
7969 @var{stepping} can be obtained from @code{/proc/cpuinfo}.
7971 If @var{identifier} is @code{auto}, enable errata workarounds for the
7972 processor on which the trace was recorded. If @var{identifier} is
7973 @code{none}, errata workarounds are disabled.
7975 For example, when using an old @value{GDBN} on a new system, decode
7976 may fail because @value{GDBN} does not support the new processor. It
7977 often suffices to specify an older processor that @value{GDBN}
7981 (@value{GDBP}) info record
7982 Active record target: record-btrace
7983 Recording format: Intel Processor Trace.
7985 Failed to configure the Intel Processor Trace decoder: unknown cpu.
7986 (@value{GDBP}) set record btrace cpu intel:6/158
7987 (@value{GDBP}) info record
7988 Active record target: record-btrace
7989 Recording format: Intel Processor Trace.
7991 Recorded 84872 instructions in 3189 functions (0 gaps) for thread 1 (...).
7994 @kindex show record btrace
7995 @item show record btrace replay-memory-access
7996 Show the current setting of @code{replay-memory-access}.
7998 @item show record btrace cpu
7999 Show the processor to be used for enabling trace decode errata
8002 @kindex set record btrace bts
8003 @item set record btrace bts buffer-size @var{size}
8004 @itemx set record btrace bts buffer-size unlimited
8005 Set the requested ring buffer size for branch tracing in @acronym{BTS}
8006 format. Default is 64KB.
8008 If @var{size} is a positive number, then @value{GDBN} will try to
8009 allocate a buffer of at least @var{size} bytes for each new thread
8010 that uses the btrace recording method and the @acronym{BTS} format.
8011 The actually obtained buffer size may differ from the requested
8012 @var{size}. Use the @code{info record} command to see the actual
8013 buffer size for each thread that uses the btrace recording method and
8014 the @acronym{BTS} format.
8016 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
8017 allocate a buffer of 4MB.
8019 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
8020 also need longer to process the branch trace data before it can be used.
8022 @item show record btrace bts buffer-size @var{size}
8023 Show the current setting of the requested ring buffer size for branch
8024 tracing in @acronym{BTS} format.
8026 @kindex set record btrace pt
8027 @item set record btrace pt buffer-size @var{size}
8028 @itemx set record btrace pt buffer-size unlimited
8029 Set the requested ring buffer size for branch tracing in Intel
8030 Processor Trace format. Default is 16KB.
8032 If @var{size} is a positive number, then @value{GDBN} will try to
8033 allocate a buffer of at least @var{size} bytes for each new thread
8034 that uses the btrace recording method and the Intel Processor Trace
8035 format. The actually obtained buffer size may differ from the
8036 requested @var{size}. Use the @code{info record} command to see the
8037 actual buffer size for each thread.
8039 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
8040 allocate a buffer of 4MB.
8042 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
8043 also need longer to process the branch trace data before it can be used.
8045 @item show record btrace pt buffer-size @var{size}
8046 Show the current setting of the requested ring buffer size for branch
8047 tracing in Intel Processor Trace format.
8051 Show various statistics about the recording depending on the recording
8056 For the @code{full} recording method, it shows the state of process
8057 record and its in-memory execution log buffer, including:
8061 Whether in record mode or replay mode.
8063 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
8065 Highest recorded instruction number.
8067 Current instruction about to be replayed (if in replay mode).
8069 Number of instructions contained in the execution log.
8071 Maximum number of instructions that may be contained in the execution log.
8075 For the @code{btrace} recording method, it shows:
8081 Number of instructions that have been recorded.
8083 Number of blocks of sequential control-flow formed by the recorded
8086 Whether in record mode or replay mode.
8089 For the @code{bts} recording format, it also shows:
8092 Size of the perf ring buffer.
8095 For the @code{pt} recording format, it also shows:
8098 Size of the perf ring buffer.
8102 @kindex record delete
8105 When record target runs in replay mode (``in the past''), delete the
8106 subsequent execution log and begin to record a new execution log starting
8107 from the current address. This means you will abandon the previously
8108 recorded ``future'' and begin recording a new ``future''.
8110 @kindex record instruction-history
8111 @kindex rec instruction-history
8112 @item record instruction-history
8113 Disassembles instructions from the recorded execution log. By
8114 default, ten instructions are disassembled. This can be changed using
8115 the @code{set record instruction-history-size} command. Instructions
8116 are printed in execution order.
8118 It can also print mixed source+disassembly if you specify the the
8119 @code{/m} or @code{/s} modifier, and print the raw instructions in hex
8120 as well as in symbolic form by specifying the @code{/r} or @code{/b}
8121 modifier. The behaviour of the @code{/m}, @code{/s}, @code{/r}, and
8122 @code{/b} modifiers are the same as for the @kbd{disassemble} command
8123 (@pxref{disassemble,,@kbd{disassemble}}).
8125 The current position marker is printed for the instruction at the
8126 current program counter value. This instruction can appear multiple
8127 times in the trace and the current position marker will be printed
8128 every time. To omit the current position marker, specify the
8131 To better align the printed instructions when the trace contains
8132 instructions from more than one function, the function name may be
8133 omitted by specifying the @code{/f} modifier.
8135 Speculatively executed instructions are prefixed with @samp{?}. This
8136 feature is not available for all recording formats.
8138 There are several ways to specify what part of the execution log to
8142 @item record instruction-history @var{insn}
8143 Disassembles ten instructions starting from instruction number
8146 @item record instruction-history @var{insn}, +/-@var{n}
8147 Disassembles @var{n} instructions around instruction number
8148 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
8149 @var{n} instructions after instruction number @var{insn}. If
8150 @var{n} is preceded with @code{-}, disassembles @var{n}
8151 instructions before instruction number @var{insn}.
8153 @item record instruction-history
8154 Disassembles ten more instructions after the last disassembly.
8156 @item record instruction-history -
8157 Disassembles ten more instructions before the last disassembly.
8159 @item record instruction-history @var{begin}, @var{end}
8160 Disassembles instructions beginning with instruction number
8161 @var{begin} until instruction number @var{end}. The instruction
8162 number @var{end} is included.
8165 This command may not be available for all recording methods.
8168 @item set record instruction-history-size @var{size}
8169 @itemx set record instruction-history-size unlimited
8170 Define how many instructions to disassemble in the @code{record
8171 instruction-history} command. The default value is 10.
8172 A @var{size} of @code{unlimited} means unlimited instructions.
8175 @item show record instruction-history-size
8176 Show how many instructions to disassemble in the @code{record
8177 instruction-history} command.
8179 @kindex record function-call-history
8180 @kindex rec function-call-history
8181 @item record function-call-history
8182 Prints the execution history at function granularity. For each sequence
8183 of instructions that belong to the same function, it prints the name of
8184 that function, the source lines for this instruction sequence (if the
8185 @code{/l} modifier is specified), and the instructions numbers that form
8186 the sequence (if the @code{/i} modifier is specified). The function names
8187 are indented to reflect the call stack depth if the @code{/c} modifier is
8188 specified. The @code{/l}, @code{/i}, and @code{/c} modifiers can be given
8192 (@value{GDBP}) @b{list 1, 10}
8203 (@value{GDBP}) @b{record function-call-history /ilc}
8204 1 bar inst 1,4 at foo.c:6,8
8205 2 foo inst 5,10 at foo.c:2,3
8206 3 bar inst 11,13 at foo.c:9,10
8209 By default, ten functions are printed. This can be changed using the
8210 @code{set record function-call-history-size} command. Functions are
8211 printed in execution order. There are several ways to specify what
8215 @item record function-call-history @var{func}
8216 Prints ten functions starting from function number @var{func}.
8218 @item record function-call-history @var{func}, +/-@var{n}
8219 Prints @var{n} functions around function number @var{func}. If
8220 @var{n} is preceded with @code{+}, prints @var{n} functions after
8221 function number @var{func}. If @var{n} is preceded with @code{-},
8222 prints @var{n} functions before function number @var{func}.
8224 @item record function-call-history
8225 Prints ten more functions after the last ten-function print.
8227 @item record function-call-history -
8228 Prints ten more functions before the last ten-function print.
8230 @item record function-call-history @var{begin}, @var{end}
8231 Prints functions beginning with function number @var{begin} until
8232 function number @var{end}. The function number @var{end} is included.
8235 This command may not be available for all recording methods.
8237 @item set record function-call-history-size @var{size}
8238 @itemx set record function-call-history-size unlimited
8239 Define how many functions to print in the
8240 @code{record function-call-history} command. The default value is 10.
8241 A size of @code{unlimited} means unlimited functions.
8243 @item show record function-call-history-size
8244 Show how many functions to print in the
8245 @code{record function-call-history} command.
8250 @chapter Examining the Stack
8252 When your program has stopped, the first thing you need to know is where it
8253 stopped and how it got there.
8256 Each time your program performs a function call, information about the call
8258 That information includes the location of the call in your program,
8259 the arguments of the call,
8260 and the local variables of the function being called.
8261 The information is saved in a block of data called a @dfn{stack frame}.
8262 The stack frames are allocated in a region of memory called the @dfn{call
8265 When your program stops, the @value{GDBN} commands for examining the
8266 stack allow you to see all of this information.
8268 @cindex selected frame
8269 One of the stack frames is @dfn{selected} by @value{GDBN} and many
8270 @value{GDBN} commands refer implicitly to the selected frame. In
8271 particular, whenever you ask @value{GDBN} for the value of a variable in
8272 your program, the value is found in the selected frame. There are
8273 special @value{GDBN} commands to select whichever frame you are
8274 interested in. @xref{Selection, ,Selecting a Frame}.
8276 When your program stops, @value{GDBN} automatically selects the
8277 currently executing frame and describes it briefly, similar to the
8278 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
8281 * Frames:: Stack frames
8282 * Backtrace:: Backtraces
8283 * Selection:: Selecting a frame
8284 * Frame Info:: Information on a frame
8285 * Frame Apply:: Applying a command to several frames
8286 * Frame Filter Management:: Managing frame filters
8291 @section Stack Frames
8293 @cindex frame, definition
8295 The call stack is divided up into contiguous pieces called @dfn{stack
8296 frames}, or @dfn{frames} for short; each frame is the data associated
8297 with one call to one function. The frame contains the arguments given
8298 to the function, the function's local variables, and the address at
8299 which the function is executing.
8301 @cindex initial frame
8302 @cindex outermost frame
8303 @cindex innermost frame
8304 When your program is started, the stack has only one frame, that of the
8305 function @code{main}. This is called the @dfn{initial} frame or the
8306 @dfn{outermost} frame. Each time a function is called, a new frame is
8307 made. Each time a function returns, the frame for that function invocation
8308 is eliminated. If a function is recursive, there can be many frames for
8309 the same function. The frame for the function in which execution is
8310 actually occurring is called the @dfn{innermost} frame. This is the most
8311 recently created of all the stack frames that still exist.
8313 @cindex frame pointer
8314 Inside your program, stack frames are identified by their addresses. A
8315 stack frame consists of many bytes, each of which has its own address; each
8316 kind of computer has a convention for choosing one byte whose
8317 address serves as the address of the frame. Usually this address is kept
8318 in a register called the @dfn{frame pointer register}
8319 (@pxref{Registers, $fp}) while execution is going on in that frame.
8322 @cindex frame number
8323 @value{GDBN} labels each existing stack frame with a @dfn{level}, a
8324 number that is zero for the innermost frame, one for the frame that
8325 called it, and so on upward. These level numbers give you a way of
8326 designating stack frames in @value{GDBN} commands. The terms
8327 @dfn{frame number} and @dfn{frame level} can be used interchangeably to
8328 describe this number.
8330 @c The -fomit-frame-pointer below perennially causes hbox overflow
8331 @c underflow problems.
8332 @cindex frameless execution
8333 Some compilers provide a way to compile functions so that they operate
8334 without stack frames. (For example, the @value{NGCC} option
8336 @samp{-fomit-frame-pointer}
8338 generates functions without a frame.)
8339 This is occasionally done with heavily used library functions to save
8340 the frame setup time. @value{GDBN} has limited facilities for dealing
8341 with these function invocations. If the innermost function invocation
8342 has no stack frame, @value{GDBN} nevertheless regards it as though
8343 it had a separate frame, which is numbered zero as usual, allowing
8344 correct tracing of the function call chain. However, @value{GDBN} has
8345 no provision for frameless functions elsewhere in the stack.
8351 @cindex call stack traces
8352 A backtrace is a summary of how your program got where it is. It shows one
8353 line per frame, for many frames, starting with the currently executing
8354 frame (frame zero), followed by its caller (frame one), and on up the
8357 @anchor{backtrace-command}
8359 @kindex bt @r{(@code{backtrace})}
8360 To print a backtrace of the entire stack, use the @code{backtrace}
8361 command, or its alias @code{bt}. This command will print one line per
8362 frame for frames in the stack. By default, all stack frames are
8363 printed. You can stop the backtrace at any time by typing the system
8364 interrupt character, normally @kbd{Ctrl-c}.
8367 @item backtrace [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
8368 @itemx bt [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
8369 Print the backtrace of the entire stack.
8371 The optional @var{count} can be one of the following:
8376 Print only the innermost @var{n} frames, where @var{n} is a positive
8381 Print only the outermost @var{n} frames, where @var{n} is a positive
8389 Print the values of the local variables also. This can be combined
8390 with the optional @var{count} to limit the number of frames shown.
8393 Do not run Python frame filters on this backtrace. @xref{Frame
8394 Filter API}, for more information. Additionally use @ref{disable
8395 frame-filter all} to turn off all frame filters. This is only
8396 relevant when @value{GDBN} has been configured with @code{Python}
8400 A Python frame filter might decide to ``elide'' some frames. Normally
8401 such elided frames are still printed, but they are indented relative
8402 to the filtered frames that cause them to be elided. The @code{-hide}
8403 option causes elided frames to not be printed at all.
8406 The @code{backtrace} command also supports a number of options that
8407 allow overriding relevant global print settings as set by @code{set
8408 backtrace} and @code{set print} subcommands:
8411 @item -past-main [@code{on}|@code{off}]
8412 Set whether backtraces should continue past @code{main}. Related setting:
8413 @ref{set backtrace past-main}.
8415 @item -past-entry [@code{on}|@code{off}]
8416 Set whether backtraces should continue past the entry point of a program.
8417 Related setting: @ref{set backtrace past-entry}.
8419 @item -entry-values @code{no}|@code{only}|@code{preferred}|@code{if-needed}|@code{both}|@code{compact}|@code{default}
8420 Set printing of function arguments at function entry.
8421 Related setting: @ref{set print entry-values}.
8423 @item -frame-arguments @code{all}|@code{scalars}|@code{none}
8424 Set printing of non-scalar frame arguments.
8425 Related setting: @ref{set print frame-arguments}.
8427 @item -raw-frame-arguments [@code{on}|@code{off}]
8428 Set whether to print frame arguments in raw form.
8429 Related setting: @ref{set print raw-frame-arguments}.
8431 @item -frame-info @code{auto}|@code{source-line}|@code{location}|@code{source-and-location}|@code{location-and-address}|@code{short-location}
8432 Set printing of frame information.
8433 Related setting: @ref{set print frame-info}.
8436 The optional @var{qualifier} is maintained for backward compatibility.
8437 It can be one of the following:
8441 Equivalent to the @code{-full} option.
8444 Equivalent to the @code{-no-filters} option.
8447 Equivalent to the @code{-hide} option.
8454 The names @code{where} and @code{info stack} (abbreviated @code{info s})
8455 are additional aliases for @code{backtrace}.
8457 @cindex multiple threads, backtrace
8458 In a multi-threaded program, @value{GDBN} by default shows the
8459 backtrace only for the current thread. To display the backtrace for
8460 several or all of the threads, use the command @code{thread apply}
8461 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
8462 apply all backtrace}, @value{GDBN} will display the backtrace for all
8463 the threads; this is handy when you debug a core dump of a
8464 multi-threaded program.
8466 Each line in the backtrace shows the frame number and the function name.
8467 The program counter value is also shown---unless you use @code{set
8468 print address off}. The backtrace also shows the source file name and
8469 line number, as well as the arguments to the function. The program
8470 counter value is omitted if it is at the beginning of the code for that
8473 Here is an example of a backtrace. It was made with the command
8474 @samp{bt 3}, so it shows the innermost three frames.
8478 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
8480 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
8481 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
8483 (More stack frames follow...)
8488 The display for frame zero does not begin with a program counter
8489 value, indicating that your program has stopped at the beginning of the
8490 code for line @code{993} of @code{builtin.c}.
8493 The value of parameter @code{data} in frame 1 has been replaced by
8494 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
8495 only if it is a scalar (integer, pointer, enumeration, etc). See command
8496 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
8497 on how to configure the way function parameter values are printed.
8498 The command @kbd{set print frame-info} (@pxref{Print Settings}) controls
8499 what frame information is printed.
8501 @cindex optimized out, in backtrace
8502 @cindex function call arguments, optimized out
8503 If your program was compiled with optimizations, some compilers will
8504 optimize away arguments passed to functions if those arguments are
8505 never used after the call. Such optimizations generate code that
8506 passes arguments through registers, but doesn't store those arguments
8507 in the stack frame. @value{GDBN} has no way of displaying such
8508 arguments in stack frames other than the innermost one. Here's what
8509 such a backtrace might look like:
8513 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
8515 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
8516 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
8518 (More stack frames follow...)
8523 The values of arguments that were not saved in their stack frames are
8524 shown as @samp{<optimized out>}.
8526 If you need to display the values of such optimized-out arguments,
8527 either deduce that from other variables whose values depend on the one
8528 you are interested in, or recompile without optimizations.
8530 @cindex backtrace beyond @code{main} function
8531 @cindex program entry point
8532 @cindex startup code, and backtrace
8533 Most programs have a standard user entry point---a place where system
8534 libraries and startup code transition into user code. For C this is
8535 @code{main}@footnote{
8536 Note that embedded programs (the so-called ``free-standing''
8537 environment) are not required to have a @code{main} function as the
8538 entry point. They could even have multiple entry points.}.
8539 When @value{GDBN} finds the entry function in a backtrace
8540 it will terminate the backtrace, to avoid tracing into highly
8541 system-specific (and generally uninteresting) code.
8543 If you need to examine the startup code, or limit the number of levels
8544 in a backtrace, you can change this behavior:
8547 @item set backtrace past-main
8548 @itemx set backtrace past-main on
8549 @anchor{set backtrace past-main}
8550 @kindex set backtrace
8551 Backtraces will continue past the user entry point.
8553 @item set backtrace past-main off
8554 Backtraces will stop when they encounter the user entry point. This is the
8557 @item show backtrace past-main
8558 @kindex show backtrace
8559 Display the current user entry point backtrace policy.
8561 @item set backtrace past-entry
8562 @itemx set backtrace past-entry on
8563 @anchor{set backtrace past-entry}
8564 Backtraces will continue past the internal entry point of an application.
8565 This entry point is encoded by the linker when the application is built,
8566 and is likely before the user entry point @code{main} (or equivalent) is called.
8568 @item set backtrace past-entry off
8569 Backtraces will stop when they encounter the internal entry point of an
8570 application. This is the default.
8572 @item show backtrace past-entry
8573 Display the current internal entry point backtrace policy.
8575 @item set backtrace limit @var{n}
8576 @itemx set backtrace limit 0
8577 @itemx set backtrace limit unlimited
8578 @anchor{set backtrace limit}
8579 @cindex backtrace limit
8580 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
8581 or zero means unlimited levels.
8583 @item show backtrace limit
8584 Display the current limit on backtrace levels.
8587 You can control how file names are displayed.
8590 @item set filename-display
8591 @itemx set filename-display relative
8592 @cindex filename-display
8593 Display file names relative to the compilation directory. This is the default.
8595 @item set filename-display basename
8596 Display only basename of a filename.
8598 @item set filename-display absolute
8599 Display an absolute filename.
8601 @item show filename-display
8602 Show the current way to display filenames.
8606 @section Selecting a Frame
8608 Most commands for examining the stack and other data in your program work on
8609 whichever stack frame is selected at the moment. Here are the commands for
8610 selecting a stack frame; all of them finish by printing a brief description
8611 of the stack frame just selected.
8614 @kindex frame@r{, selecting}
8615 @kindex f @r{(@code{frame})}
8616 @item frame @r{[} @var{frame-selection-spec} @r{]}
8617 @item f @r{[} @var{frame-selection-spec} @r{]}
8618 The @command{frame} command allows different stack frames to be
8619 selected. The @var{frame-selection-spec} can be any of the following:
8624 @item level @var{num}
8625 Select frame level @var{num}. Recall that frame zero is the innermost
8626 (currently executing) frame, frame one is the frame that called the
8627 innermost one, and so on. The highest level frame is usually the one
8630 As this is the most common method of navigating the frame stack, the
8631 string @command{level} can be omitted. For example, the following two
8632 commands are equivalent:
8635 (@value{GDBP}) frame 3
8636 (@value{GDBP}) frame level 3
8639 @kindex frame address
8640 @item address @var{stack-address}
8641 Select the frame with stack address @var{stack-address}. The
8642 @var{stack-address} for a frame can be seen in the output of
8643 @command{info frame}, for example:
8646 (@value{GDBP}) info frame
8647 Stack level 1, frame at 0x7fffffffda30:
8648 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
8649 tail call frame, caller of frame at 0x7fffffffda30
8650 source language c++.
8651 Arglist at unknown address.
8652 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
8655 The @var{stack-address} for this frame is @code{0x7fffffffda30} as
8656 indicated by the line:
8659 Stack level 1, frame at 0x7fffffffda30:
8662 @kindex frame function
8663 @item function @var{function-name}
8664 Select the stack frame for function @var{function-name}. If there are
8665 multiple stack frames for function @var{function-name} then the inner
8666 most stack frame is selected.
8669 @item view @var{stack-address} @r{[} @var{pc-addr} @r{]}
8670 View a frame that is not part of @value{GDBN}'s backtrace. The frame
8671 viewed has stack address @var{stack-addr}, and optionally, a program
8672 counter address of @var{pc-addr}.
8674 This is useful mainly if the chaining of stack frames has been
8675 damaged by a bug, making it impossible for @value{GDBN} to assign
8676 numbers properly to all frames. In addition, this can be useful
8677 when your program has multiple stacks and switches between them.
8679 When viewing a frame outside the current backtrace using
8680 @command{frame view} then you can always return to the original
8681 stack using one of the previous stack frame selection instructions,
8682 for example @command{frame level 0}.
8688 Move @var{n} frames up the stack; @var{n} defaults to 1. For positive
8689 numbers @var{n}, this advances toward the outermost frame, to higher
8690 frame numbers, to frames that have existed longer.
8693 @kindex do @r{(@code{down})}
8695 Move @var{n} frames down the stack; @var{n} defaults to 1. For
8696 positive numbers @var{n}, this advances toward the innermost frame, to
8697 lower frame numbers, to frames that were created more recently.
8698 You may abbreviate @code{down} as @code{do}.
8701 All of these commands end by printing two lines of output describing the
8702 frame. The first line shows the frame number, the function name, the
8703 arguments, and the source file and line number of execution in that
8704 frame. The second line shows the text of that source line.
8712 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
8714 10 read_input_file (argv[i]);
8718 After such a printout, the @code{list} command with no arguments
8719 prints ten lines centered on the point of execution in the frame.
8720 You can also edit the program at the point of execution with your favorite
8721 editing program by typing @code{edit}.
8722 @xref{List, ,Printing Source Lines},
8726 @kindex select-frame
8727 @item select-frame @r{[} @var{frame-selection-spec} @r{]}
8728 The @code{select-frame} command is a variant of @code{frame} that does
8729 not display the new frame after selecting it. This command is
8730 intended primarily for use in @value{GDBN} command scripts, where the
8731 output might be unnecessary and distracting. The
8732 @var{frame-selection-spec} is as for the @command{frame} command
8733 described in @ref{Selection, ,Selecting a Frame}.
8735 @kindex down-silently
8737 @item up-silently @var{n}
8738 @itemx down-silently @var{n}
8739 These two commands are variants of @code{up} and @code{down},
8740 respectively; they differ in that they do their work silently, without
8741 causing display of the new frame. They are intended primarily for use
8742 in @value{GDBN} command scripts, where the output might be unnecessary and
8747 @section Information About a Frame
8749 There are several other commands to print information about the selected
8755 When used without any argument, this command does not change which
8756 frame is selected, but prints a brief description of the currently
8757 selected stack frame. It can be abbreviated @code{f}. With an
8758 argument, this command is used to select a stack frame.
8759 @xref{Selection, ,Selecting a Frame}.
8762 @kindex info f @r{(@code{info frame})}
8765 This command prints a verbose description of the selected stack frame,
8770 the address of the frame
8772 the address of the next frame down (called by this frame)
8774 the address of the next frame up (caller of this frame)
8776 the language in which the source code corresponding to this frame is written
8778 the address of the frame's arguments
8780 the address of the frame's local variables
8782 the program counter saved in it (the address of execution in the caller frame)
8784 which registers were saved in the frame
8787 @noindent The verbose description is useful when
8788 something has gone wrong that has made the stack format fail to fit
8789 the usual conventions.
8791 @item info frame @r{[} @var{frame-selection-spec} @r{]}
8792 @itemx info f @r{[} @var{frame-selection-spec} @r{]}
8793 Print a verbose description of the frame selected by
8794 @var{frame-selection-spec}. The @var{frame-selection-spec} is the
8795 same as for the @command{frame} command (@pxref{Selection, ,Selecting
8796 a Frame}). The selected frame remains unchanged by this command.
8799 @item info args [-q]
8800 Print the arguments of the selected frame, each on a separate line.
8802 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
8803 printing header information and messages explaining why no argument
8806 @item info args [-q] [-t @var{type_regexp}] [@var{regexp}]
8807 Like @kbd{info args}, but only print the arguments selected
8808 with the provided regexp(s).
8810 If @var{regexp} is provided, print only the arguments whose names
8811 match the regular expression @var{regexp}.
8813 If @var{type_regexp} is provided, print only the arguments whose
8814 types, as printed by the @code{whatis} command, match
8815 the regular expression @var{type_regexp}.
8816 If @var{type_regexp} contains space(s), it should be enclosed in
8817 quote characters. If needed, use backslash to escape the meaning
8818 of special characters or quotes.
8820 If both @var{regexp} and @var{type_regexp} are provided, an argument
8821 is printed only if its name matches @var{regexp} and its type matches
8824 @item info locals [-q]
8826 Print the local variables of the selected frame, each on a separate
8827 line. These are all variables (declared either static or automatic)
8828 accessible at the point of execution of the selected frame.
8830 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
8831 printing header information and messages explaining why no local variables
8834 @item info locals [-q] [-t @var{type_regexp}] [@var{regexp}]
8835 Like @kbd{info locals}, but only print the local variables selected
8836 with the provided regexp(s).
8838 If @var{regexp} is provided, print only the local variables whose names
8839 match the regular expression @var{regexp}.
8841 If @var{type_regexp} is provided, print only the local variables whose
8842 types, as printed by the @code{whatis} command, match
8843 the regular expression @var{type_regexp}.
8844 If @var{type_regexp} contains space(s), it should be enclosed in
8845 quote characters. If needed, use backslash to escape the meaning
8846 of special characters or quotes.
8848 If both @var{regexp} and @var{type_regexp} are provided, a local variable
8849 is printed only if its name matches @var{regexp} and its type matches
8852 The command @kbd{info locals -q -t @var{type_regexp}} can usefully be
8853 combined with the commands @kbd{frame apply} and @kbd{thread apply}.
8854 For example, your program might use Resource Acquisition Is
8855 Initialization types (RAII) such as @code{lock_something_t}: each
8856 local variable of type @code{lock_something_t} automatically places a
8857 lock that is destroyed when the variable goes out of scope. You can
8858 then list all acquired locks in your program by doing
8860 thread apply all -s frame apply all -s info locals -q -t lock_something_t
8863 or the equivalent shorter form
8865 tfaas i lo -q -t lock_something_t
8871 @section Applying a Command to Several Frames.
8873 @cindex apply command to several frames
8875 @item frame apply [all | @var{count} | @var{-count} | level @var{level}@dots{}] [@var{option}]@dots{} @var{command}
8876 The @code{frame apply} command allows you to apply the named
8877 @var{command} to one or more frames.
8881 Specify @code{all} to apply @var{command} to all frames.
8884 Use @var{count} to apply @var{command} to the innermost @var{count}
8885 frames, where @var{count} is a positive number.
8888 Use @var{-count} to apply @var{command} to the outermost @var{count}
8889 frames, where @var{count} is a positive number.
8892 Use @code{level} to apply @var{command} to the set of frames identified
8893 by the @var{level} list. @var{level} is a frame level or a range of frame
8894 levels as @var{level1}-@var{level2}. The frame level is the number shown
8895 in the first field of the @samp{backtrace} command output.
8896 E.g., @samp{2-4 6-8 3} indicates to apply @var{command} for the frames
8897 at levels 2, 3, 4, 6, 7, 8, and then again on frame at level 3.
8901 Note that the frames on which @code{frame apply} applies a command are
8902 also influenced by the @code{set backtrace} settings such as @code{set
8903 backtrace past-main} and @code{set backtrace limit N}.
8904 @xref{Backtrace,,Backtraces}.
8906 The @code{frame apply} command also supports a number of options that
8907 allow overriding relevant @code{set backtrace} settings:
8910 @item -past-main [@code{on}|@code{off}]
8911 Whether backtraces should continue past @code{main}.
8912 Related setting: @ref{set backtrace past-main}.
8914 @item -past-entry [@code{on}|@code{off}]
8915 Whether backtraces should continue past the entry point of a program.
8916 Related setting: @ref{set backtrace past-entry}.
8919 By default, @value{GDBN} displays some frame information before the
8920 output produced by @var{command}, and an error raised during the
8921 execution of a @var{command} will abort @code{frame apply}. The
8922 following options can be used to fine-tune these behaviors:
8926 The flag @code{-c}, which stands for @samp{continue}, causes any
8927 errors in @var{command} to be displayed, and the execution of
8928 @code{frame apply} then continues.
8930 The flag @code{-s}, which stands for @samp{silent}, causes any errors
8931 or empty output produced by a @var{command} to be silently ignored.
8932 That is, the execution continues, but the frame information and errors
8935 The flag @code{-q} (@samp{quiet}) disables printing the frame
8939 The following example shows how the flags @code{-c} and @code{-s} are
8940 working when applying the command @code{p j} to all frames, where
8941 variable @code{j} can only be successfully printed in the outermost
8942 @code{#1 main} frame.
8946 (@value{GDBP}) frame apply all p j
8947 #0 some_function (i=5) at fun.c:4
8948 No symbol "j" in current context.
8949 (@value{GDBP}) frame apply all -c p j
8950 #0 some_function (i=5) at fun.c:4
8951 No symbol "j" in current context.
8952 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8954 (@value{GDBP}) frame apply all -s p j
8955 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8961 By default, @samp{frame apply}, prints the frame location
8962 information before the command output:
8966 (@value{GDBP}) frame apply all p $sp
8967 #0 some_function (i=5) at fun.c:4
8968 $4 = (void *) 0xffffd1e0
8969 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8970 $5 = (void *) 0xffffd1f0
8975 If the flag @code{-q} is given, no frame information is printed:
8978 (@value{GDBP}) frame apply all -q p $sp
8979 $12 = (void *) 0xffffd1e0
8980 $13 = (void *) 0xffffd1f0
8990 @cindex apply a command to all frames (ignoring errors and empty output)
8991 @item faas @var{command}
8992 Shortcut for @code{frame apply all -s @var{command}}.
8993 Applies @var{command} on all frames, ignoring errors and empty output.
8995 It can for example be used to print a local variable or a function
8996 argument without knowing the frame where this variable or argument
8999 (@value{GDBP}) faas p some_local_var_i_do_not_remember_where_it_is
9002 The @code{faas} command accepts the same options as the @code{frame
9003 apply} command. @xref{Frame Apply,,frame apply}.
9005 Note that the command @code{tfaas @var{command}} applies @var{command}
9006 on all frames of all threads. See @xref{Threads,,Threads}.
9010 @node Frame Filter Management
9011 @section Management of Frame Filters.
9012 @cindex managing frame filters
9014 Frame filters are Python based utilities to manage and decorate the
9015 output of frames. @xref{Frame Filter API}, for further information.
9017 Managing frame filters is performed by several commands available
9018 within @value{GDBN}, detailed here.
9021 @kindex info frame-filter
9022 @item info frame-filter
9023 Print a list of installed frame filters from all dictionaries, showing
9024 their name, priority and enabled status.
9026 @kindex disable frame-filter
9027 @anchor{disable frame-filter all}
9028 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
9029 Disable a frame filter in the dictionary matching
9030 @var{filter-dictionary} and @var{filter-name}. The
9031 @var{filter-dictionary} may be @code{all}, @code{global},
9032 @code{progspace}, or the name of the object file where the frame filter
9033 dictionary resides. When @code{all} is specified, all frame filters
9034 across all dictionaries are disabled. The @var{filter-name} is the name
9035 of the frame filter and is used when @code{all} is not the option for
9036 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
9037 may be enabled again later.
9039 @kindex enable frame-filter
9040 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
9041 Enable a frame filter in the dictionary matching
9042 @var{filter-dictionary} and @var{filter-name}. The
9043 @var{filter-dictionary} may be @code{all}, @code{global},
9044 @code{progspace} or the name of the object file where the frame filter
9045 dictionary resides. When @code{all} is specified, all frame filters across
9046 all dictionaries are enabled. The @var{filter-name} is the name of the frame
9047 filter and is used when @code{all} is not the option for
9048 @var{filter-dictionary}.
9053 (@value{GDBP}) info frame-filter
9055 global frame-filters:
9056 Priority Enabled Name
9057 1000 No PrimaryFunctionFilter
9060 progspace /build/test frame-filters:
9061 Priority Enabled Name
9062 100 Yes ProgspaceFilter
9064 objfile /build/test frame-filters:
9065 Priority Enabled Name
9066 999 Yes BuildProgramFilter
9068 (@value{GDBP}) disable frame-filter /build/test BuildProgramFilter
9069 (@value{GDBP}) info frame-filter
9071 global frame-filters:
9072 Priority Enabled Name
9073 1000 No PrimaryFunctionFilter
9076 progspace /build/test frame-filters:
9077 Priority Enabled Name
9078 100 Yes ProgspaceFilter
9080 objfile /build/test frame-filters:
9081 Priority Enabled Name
9082 999 No BuildProgramFilter
9084 (@value{GDBP}) enable frame-filter global PrimaryFunctionFilter
9085 (@value{GDBP}) info frame-filter
9087 global frame-filters:
9088 Priority Enabled Name
9089 1000 Yes PrimaryFunctionFilter
9092 progspace /build/test frame-filters:
9093 Priority Enabled Name
9094 100 Yes ProgspaceFilter
9096 objfile /build/test frame-filters:
9097 Priority Enabled Name
9098 999 No BuildProgramFilter
9101 @kindex set frame-filter priority
9102 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
9103 Set the @var{priority} of a frame filter in the dictionary matching
9104 @var{filter-dictionary}, and the frame filter name matching
9105 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
9106 @code{progspace} or the name of the object file where the frame filter
9107 dictionary resides. The @var{priority} is an integer.
9109 @kindex show frame-filter priority
9110 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
9111 Show the @var{priority} of a frame filter in the dictionary matching
9112 @var{filter-dictionary}, and the frame filter name matching
9113 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
9114 @code{progspace} or the name of the object file where the frame filter
9120 (@value{GDBP}) info frame-filter
9122 global frame-filters:
9123 Priority Enabled Name
9124 1000 Yes PrimaryFunctionFilter
9127 progspace /build/test frame-filters:
9128 Priority Enabled Name
9129 100 Yes ProgspaceFilter
9131 objfile /build/test frame-filters:
9132 Priority Enabled Name
9133 999 No BuildProgramFilter
9135 (@value{GDBP}) set frame-filter priority global Reverse 50
9136 (@value{GDBP}) info frame-filter
9138 global frame-filters:
9139 Priority Enabled Name
9140 1000 Yes PrimaryFunctionFilter
9143 progspace /build/test frame-filters:
9144 Priority Enabled Name
9145 100 Yes ProgspaceFilter
9147 objfile /build/test frame-filters:
9148 Priority Enabled Name
9149 999 No BuildProgramFilter
9154 @chapter Examining Source Files
9156 @value{GDBN} can print parts of your program's source, since the debugging
9157 information recorded in the program tells @value{GDBN} what source files were
9158 used to build it. When your program stops, @value{GDBN} spontaneously prints
9159 the line where it stopped. Likewise, when you select a stack frame
9160 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
9161 execution in that frame has stopped. You can print other portions of
9162 source files by explicit command.
9164 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
9165 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
9166 @value{GDBN} under @sc{gnu} Emacs}.
9169 * List:: Printing source lines
9170 * Location Specifications:: How to specify code locations
9171 * Edit:: Editing source files
9172 * Search:: Searching source files
9173 * Source Path:: Specifying source directories
9174 * Machine Code:: Source and machine code
9175 * Disable Reading Source:: Disable Reading Source Code
9179 @section Printing Source Lines
9182 @kindex l @r{(@code{list})}
9183 To print lines from a source file, use the @code{list} command
9184 (abbreviated @code{l}). By default, ten lines are printed.
9185 There are several ways to specify what part of the file you want to
9186 print; see @ref{Location Specifications}, for the full list.
9188 Here are the forms of the @code{list} command most commonly used:
9191 @item list @var{linenum}
9192 Print lines centered around line number @var{linenum} in the
9193 current source file.
9195 @item list @var{function}
9196 Print lines centered around the beginning of function
9200 Print more lines. If the last lines printed were printed with a
9201 @code{list} command, this prints lines following the last lines
9202 printed; however, if the last line printed was a solitary line printed
9203 as part of displaying a stack frame (@pxref{Stack, ,Examining the
9204 Stack}), this prints lines centered around that line. If no
9205 @code{list} command has been used and no solitary line was printed,
9206 it prints the lines around the function @code{main}.
9209 Same as using with no arguments.
9212 Print lines just before the lines last printed.
9215 Print the lines surrounding the point of execution within the
9216 currently selected frame. If the inferior is not running, print lines
9217 around the start of the main function instead.
9220 @cindex @code{list}, how many lines to display
9221 By default, @value{GDBN} prints ten source lines with any of these forms of
9222 the @code{list} command. You can change this using @code{set listsize}:
9225 @kindex set listsize
9226 @item set listsize @var{count}
9227 @itemx set listsize unlimited
9228 Make the @code{list} command display @var{count} source lines (unless
9229 the @code{list} argument explicitly specifies some other number).
9230 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
9232 @kindex show listsize
9234 Display the number of lines that @code{list} prints.
9237 Repeating a @code{list} command with @key{RET} discards the argument,
9238 so it is equivalent to typing just @code{list}. This is more useful
9239 than listing the same lines again. An exception is made for an
9240 argument of @samp{-}; that argument is preserved in repetition so that
9241 each repetition moves up in the source file.
9243 In general, the @code{list} command expects you to supply zero, one or
9244 two location specs. These location specs are interpreted to resolve
9245 to source code lines; there are several ways of writing them
9246 (@pxref{Location Specifications}), but the effect is always to resolve
9247 to some source lines to display.
9249 Here is a complete description of the possible arguments for @code{list}:
9252 @item list @var{locspec}
9253 Print lines centered around the line or lines of all the code
9254 locations that result from resolving @var{locspec}.
9256 @item list @var{first},@var{last}
9257 Print lines from @var{first} to @var{last}. Both arguments are
9258 location specs. When a @code{list} command has two location specs,
9259 and the source file of the second location spec is omitted, this
9260 refers to the same source file as the first location spec. If either
9261 @var{first} or @var{last} resolve to more than one source line in the
9262 program, then the list command shows the list of resolved source
9263 lines and does not proceed with the source code listing.
9265 @item list ,@var{last}
9266 Print lines ending with @var{last}.
9268 Likewise, if @var{last} resolves to more than one source line in the
9269 program, then the list command prints the list of resolved source
9270 lines and does not proceed with the source code listing.
9272 @item list @var{first},
9273 Print lines starting with @var{first}.
9276 Print lines just after the lines last printed.
9279 Print lines just before the lines last printed.
9282 As described in the preceding table.
9285 @node Location Specifications
9286 @section Location Specifications
9287 @cindex specifying location
9289 @cindex source location
9290 @cindex code location
9292 @cindex location spec
9293 Several @value{GDBN} commands accept arguments that specify a location
9294 or locations of your program's code. Many times locations are
9295 specified using a source line number, but they can also be specified
9296 by a function name, an address, a label, etc. The different
9297 forms of specifying a location that @value{GDBN} recognizes are
9298 collectively known as forms of @dfn{location specification}, or
9299 @dfn{location spec}. This section documents the forms of specifying
9300 locations that @value{GDBN} recognizes.
9302 @cindex location resolution
9303 @cindex resolution of location spec
9304 When you specify a location, @value{GDBN} needs to find the place in
9305 your program, known as @dfn{code location}, that corresponds to the
9306 given location spec. We call this process of finding actual code
9307 locations corresponding to a location spec @dfn{location resolution}.
9309 A concrete code location in your program is uniquely identifiable by a
9310 set of several attributes: its source line number, the name of its
9311 source file, the fully-qualified and prototyped function in which it
9312 is defined, and an instruction address. Because each inferior has its
9313 own address space, the inferior number is also a necessary part of
9316 By contrast, location specs you type will many times omit some of
9317 these attributes. For example, it is customary to specify just the
9318 source line number to mean a line in the current source file, or
9319 specify just the basename of the file, omitting its directories. In
9320 other words, a location spec is usually incomplete, a kind of
9321 blueprint, and @value{GDBN} needs to complete the missing attributes
9322 by using the implied defaults, and by considering the source code and
9323 the debug information available to it. This is what location
9324 resolution is about.
9326 The resolution of an incomplete location spec can produce more than a
9327 single code location, if the spec doesn't allow distinguishing between
9328 them. Here are some examples of situations that result in a location
9329 spec matching multiple code locations in your program:
9333 The location spec specifies a function name, and there are several
9334 functions in the program which have that name. (To distinguish
9335 between them, you can specify a fully-qualified and prototyped
9336 function name, such as @code{A::func(int)} instead of just
9340 The location spec specifies a source file name, and there are several
9341 source files in the program that share the same name, for example
9342 several files with the same basename in different subdirectories. (To
9343 distinguish between them, specify enough leading directories with the
9347 For a C@t{++} constructor, the @value{NGCC} compiler generates several
9348 instances of the function body, used in different cases, but their
9349 source-level names are identical.
9352 For a C@t{++} template function, a given line in the function can
9353 correspond to any number of instantiations.
9356 For an inlined function, a given source line can correspond to several
9357 actual code locations with that function's inlined code.
9360 Resolution of a location spec can also fail to produce a complete code
9361 location, or even fail to produce any code location. Here are some
9362 examples of such situations:
9366 Some parts of the program lack detailed enough debug info, so the
9367 resolved code location lacks some attributes, like source file name
9368 and line number, leaving just the instruction address and perhaps also
9369 a function name. Such an incomplete code location is only usable in
9370 contexts that work with addresses and/or function names. Some
9371 commands can only work with complete code locations.
9374 The location spec specifies a function name, and there are no
9375 functions in the program by that name, or they only exist in a
9376 yet-unloaded shared library.
9379 The location spec specifies a source file name, and there are no
9380 source files in the program by that name, or they only exist in a
9381 yet-unloaded shared library.
9384 The location spec specifies both a source file name and a source line
9385 number, and even though there are source files in the program that
9386 match the file name, none of those files has the specified line
9390 Locations may be specified using three different formats: linespec
9391 locations, explicit locations, or address locations. The following
9392 subsections describe these formats.
9395 * Linespec Locations:: Linespec locations
9396 * Explicit Locations:: Explicit locations
9397 * Address Locations:: Address locations
9400 @node Linespec Locations
9401 @subsection Linespec Locations
9402 @cindex linespec locations
9404 A @dfn{linespec} is a colon-separated list of source location parameters such
9405 as file name, function name, etc. Here are all the different ways of
9406 specifying a linespec:
9410 Specifies the line number @var{linenum} of the current source file.
9413 @itemx +@var{offset}
9414 Specifies the line @var{offset} lines before or after the @dfn{current
9415 line}. For the @code{list} command, the current line is the last one
9416 printed; for the breakpoint commands, this is the line at which
9417 execution stopped in the currently selected @dfn{stack frame}
9418 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
9419 used as the second of the two linespecs in a @code{list} command,
9420 this specifies the line @var{offset} lines up or down from the first
9423 @item @var{filename}:@var{linenum}
9424 Specifies the line @var{linenum} in the source file @var{filename}.
9425 If @var{filename} is a relative file name, then it will match any
9426 source file name with the same trailing components. For example, if
9427 @var{filename} is @samp{gcc/expr.c}, then it will match source file
9428 name of @file{/build/trunk/gcc/expr.c}, but not
9429 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
9431 @item @var{function}
9432 Specifies the line that begins the body of the function @var{function}.
9433 For example, in C, this is the line with the open brace.
9435 By default, in C@t{++} and Ada, @var{function} is interpreted as
9436 specifying all functions named @var{function} in all scopes. For
9437 C@t{++}, this means in all namespaces and classes. For Ada, this
9438 means in all packages.
9440 For example, assuming a program with C@t{++} symbols named
9441 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
9442 func}} and @w{@kbd{break B::func}} set a breakpoint on both symbols.
9444 Commands that accept a linespec let you override this with the
9445 @code{-qualified} option. For example, @w{@kbd{break -qualified
9446 func}} sets a breakpoint on a free-function named @code{func} ignoring
9447 any C@t{++} class methods and namespace functions called @code{func}.
9449 @xref{Explicit Locations}.
9451 @item @var{function}:@var{label}
9452 Specifies the line where @var{label} appears in @var{function}.
9454 @item @var{filename}:@var{function}
9455 Specifies the line that begins the body of the function @var{function}
9456 in the file @var{filename}. You only need the file name with a
9457 function name to avoid ambiguity when there are identically named
9458 functions in different source files.
9461 Specifies the line at which the label named @var{label} appears
9462 in the function corresponding to the currently selected stack frame.
9463 If there is no current selected stack frame (for instance, if the inferior
9464 is not running), then @value{GDBN} will not search for a label.
9466 @cindex breakpoint at static probe point
9467 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
9468 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
9469 applications to embed static probes. @xref{Static Probe Points}, for more
9470 information on finding and using static probes. This form of linespec
9471 specifies the location of such a static probe.
9473 If @var{objfile} is given, only probes coming from that shared library
9474 or executable matching @var{objfile} as a regular expression are considered.
9475 If @var{provider} is given, then only probes from that provider are considered.
9476 If several probes match the spec, @value{GDBN} will insert a breakpoint at
9477 each one of those probes.
9480 @node Explicit Locations
9481 @subsection Explicit Locations
9482 @cindex explicit locations
9484 @dfn{Explicit locations} allow the user to directly specify the source
9485 location's parameters using option-value pairs.
9487 Explicit locations are useful when several functions, labels, or
9488 file names have the same name (base name for files) in the program's
9489 sources. In these cases, explicit locations point to the source
9490 line you meant more accurately and unambiguously. Also, using
9491 explicit locations might be faster in large programs.
9493 For example, the linespec @samp{foo:bar} may refer to a function @code{bar}
9494 defined in the file named @file{foo} or the label @code{bar} in a function
9495 named @code{foo}. @value{GDBN} must search either the file system or
9496 the symbol table to know.
9498 The list of valid explicit location options is summarized in the
9502 @item -source @var{filename}
9503 The value specifies the source file name. To differentiate between
9504 files with the same base name, prepend as many directories as is necessary
9505 to uniquely identify the desired file, e.g., @file{foo/bar/baz.c}. Otherwise
9506 @value{GDBN} will use the first file it finds with the given base
9507 name. This option requires the use of either @code{-function} or @code{-line}.
9509 @item -function @var{function}
9510 The value specifies the name of a function. Operations
9511 on function locations unmodified by other options (such as @code{-label}
9512 or @code{-line}) refer to the line that begins the body of the function.
9513 In C, for example, this is the line with the open brace.
9515 By default, in C@t{++} and Ada, @var{function} is interpreted as
9516 specifying all functions named @var{function} in all scopes. For
9517 C@t{++}, this means in all namespaces and classes. For Ada, this
9518 means in all packages.
9520 For example, assuming a program with C@t{++} symbols named
9521 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
9522 -function func}} and @w{@kbd{break -function B::func}} set a
9523 breakpoint on both symbols.
9525 You can use the @kbd{-qualified} flag to override this (see below).
9529 This flag makes @value{GDBN} interpret a function name specified with
9530 @kbd{-function} as a complete fully-qualified name.
9532 For example, assuming a C@t{++} program with symbols named
9533 @code{A::B::func} and @code{B::func}, the @w{@kbd{break -qualified
9534 -function B::func}} command sets a breakpoint on @code{B::func}, only.
9536 (Note: the @kbd{-qualified} option can precede a linespec as well
9537 (@pxref{Linespec Locations}), so the particular example above could be
9538 simplified as @w{@kbd{break -qualified B::func}}.)
9540 @item -label @var{label}
9541 The value specifies the name of a label. When the function
9542 name is not specified, the label is searched in the function of the currently
9543 selected stack frame.
9545 @item -line @var{number}
9546 The value specifies a line offset for the location. The offset may either
9547 be absolute (@code{-line 3}) or relative (@code{-line +3}), depending on
9548 the command. When specified without any other options, the line offset is
9549 relative to the current line.
9552 Explicit location options may be abbreviated by omitting any non-unique
9553 trailing characters from the option name, e.g., @w{@kbd{break -s main.c -li 3}}.
9555 @node Address Locations
9556 @subsection Address Locations
9557 @cindex address locations
9559 @dfn{Address locations} indicate a specific program address. They have
9560 the generalized form *@var{address}.
9562 For line-oriented commands, such as @code{list} and @code{edit}, this
9563 specifies a source line that contains @var{address}. For @code{break} and
9564 other breakpoint-oriented commands, this can be used to set breakpoints in
9565 parts of your program which do not have debugging information or
9568 Here @var{address} may be any expression valid in the current working
9569 language (@pxref{Languages, working language}) that specifies a code
9570 address. In addition, as a convenience, @value{GDBN} extends the
9571 semantics of expressions used in locations to cover several situations
9572 that frequently occur during debugging. Here are the various forms
9576 @item @var{expression}
9577 Any expression valid in the current working language.
9579 @item @var{funcaddr}
9580 An address of a function or procedure derived from its name. In C,
9581 C@t{++}, Objective-C, Fortran, minimal, and assembly, this is
9582 simply the function's name @var{function} (and actually a special case
9583 of a valid expression). In Pascal and Modula-2, this is
9584 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
9585 (although the Pascal form also works).
9587 This form specifies the address of the function's first instruction,
9588 before the stack frame and arguments have been set up.
9590 @item '@var{filename}':@var{funcaddr}
9591 Like @var{funcaddr} above, but also specifies the name of the source
9592 file explicitly. This is useful if the name of the function does not
9593 specify the function unambiguously, e.g., if there are several
9594 functions with identical names in different source files.
9598 @section Editing Source Files
9599 @cindex editing source files
9602 @kindex e @r{(@code{edit})}
9603 To edit the lines in a source file, use the @code{edit} command.
9604 The editing program of your choice
9605 is invoked with the current line set to
9606 the active line in the program.
9607 Alternatively, there are several ways to specify what part of the file you
9608 want to print if you want to see other parts of the program:
9611 @item edit @var{locspec}
9612 Edit the source file of the code location that results from resolving
9613 @code{locspec}. Editing starts at the source file and source line
9614 @code{locspec} resolves to.
9615 @xref{Location Specifications}, for all the possible forms of the
9616 @var{locspec} argument.
9618 If @code{locspec} resolves to more than one source line in your
9619 program, then the command prints the list of resolved source lines and
9620 does not proceed with the editing.
9622 Here are the forms of the @code{edit} command most commonly used:
9625 @item edit @var{number}
9626 Edit the current source file with @var{number} as the active line number.
9628 @item edit @var{function}
9629 Edit the file containing @var{function} at the beginning of its definition.
9634 @subsection Choosing your Editor
9635 You can customize @value{GDBN} to use any editor you want
9637 The only restriction is that your editor (say @code{ex}), recognizes the
9638 following command-line syntax:
9640 ex +@var{number} file
9642 The optional numeric value +@var{number} specifies the number of the line in
9643 the file where to start editing.}.
9644 By default, it is @file{@value{EDITOR}}, but you can change this
9645 by setting the environment variable @env{EDITOR} before using
9646 @value{GDBN}. For example, to configure @value{GDBN} to use the
9647 @code{vi} editor, you could use these commands with the @code{sh} shell:
9653 or in the @code{csh} shell,
9655 setenv EDITOR /usr/bin/vi
9660 @section Searching Source Files
9661 @cindex searching source files
9663 There are two commands for searching through the current source file for a
9668 @kindex forward-search
9669 @kindex fo @r{(@code{forward-search})}
9670 @item forward-search @var{regexp}
9671 @itemx search @var{regexp}
9672 The command @samp{forward-search @var{regexp}} checks each line,
9673 starting with the one following the last line listed, for a match for
9674 @var{regexp}. It lists the line that is found. You can use the
9675 synonym @samp{search @var{regexp}} or abbreviate the command name as
9678 @kindex reverse-search
9679 @item reverse-search @var{regexp}
9680 The command @samp{reverse-search @var{regexp}} checks each line, starting
9681 with the one before the last line listed and going backward, for a match
9682 for @var{regexp}. It lists the line that is found. You can abbreviate
9683 this command as @code{rev}.
9687 @section Specifying Source Directories
9690 @cindex directories for source files
9691 Executable programs sometimes do not record the directories of the source
9692 files from which they were compiled, just the names. Even when they do,
9693 the directories could be moved between the compilation and your debugging
9694 session. @value{GDBN} has a list of directories to search for source files;
9695 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
9696 it tries all the directories in the list, in the order they are present
9697 in the list, until it finds a file with the desired name.
9699 For example, suppose an executable references the file
9700 @file{/usr/src/foo-1.0/lib/foo.c}, does not record a compilation
9701 directory, and the @dfn{source path} is @file{/mnt/cross}.
9702 @value{GDBN} would look for the source file in the following
9707 @item @file{/usr/src/foo-1.0/lib/foo.c}
9708 @item @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c}
9709 @item @file{/mnt/cross/foo.c}
9713 If the source file is not present at any of the above locations then
9714 an error is printed. @value{GDBN} does not look up the parts of the
9715 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
9716 Likewise, the subdirectories of the source path are not searched: if
9717 the source path is @file{/mnt/cross}, and the binary refers to
9718 @file{foo.c}, @value{GDBN} would not find it under
9719 @file{/mnt/cross/usr/src/foo-1.0/lib}.
9721 Plain file names, relative file names with leading directories, file
9722 names containing dots, etc.@: are all treated as described above,
9723 except that non-absolute file names are not looked up literally. If
9724 the @dfn{source path} is @file{/mnt/cross}, the source file is
9725 recorded as @file{../lib/foo.c}, and no compilation directory is
9726 recorded, then @value{GDBN} will search in the following locations:
9730 @item @file{/mnt/cross/../lib/foo.c}
9731 @item @file{/mnt/cross/foo.c}
9737 @vindex $cdir@r{, convenience variable}
9738 @vindex $cwd@r{, convenience variable}
9739 @cindex compilation directory
9740 @cindex current directory
9741 @cindex working directory
9742 @cindex directory, current
9743 @cindex directory, compilation
9744 The @dfn{source path} will always include two special entries
9745 @samp{$cdir} and @samp{$cwd}, these refer to the compilation directory
9746 (if one is recorded) and the current working directory respectively.
9748 @samp{$cdir} causes @value{GDBN} to search within the compilation
9749 directory, if one is recorded in the debug information. If no
9750 compilation directory is recorded in the debug information then
9751 @samp{$cdir} is ignored.
9753 @samp{$cwd} is not the same as @samp{.}---the former tracks the
9754 current working directory as it changes during your @value{GDBN}
9755 session, while the latter is immediately expanded to the current
9756 directory at the time you add an entry to the source path.
9758 If a compilation directory is recorded in the debug information, and
9759 @value{GDBN} has not found the source file after the first search
9760 using @dfn{source path}, then @value{GDBN} will combine the
9761 compilation directory and the filename, and then search for the source
9762 file again using the @dfn{source path}.
9764 For example, if the executable records the source file as
9765 @file{/usr/src/foo-1.0/lib/foo.c}, the compilation directory is
9766 recorded as @file{/project/build}, and the @dfn{source path} is
9767 @file{/mnt/cross:$cdir:$cwd} while the current working directory of
9768 the @value{GDBN} session is @file{/home/user}, then @value{GDBN} will
9769 search for the source file in the following locations:
9773 @item @file{/usr/src/foo-1.0/lib/foo.c}
9774 @item @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c}
9775 @item @file{/project/build/usr/src/foo-1.0/lib/foo.c}
9776 @item @file{/home/user/usr/src/foo-1.0/lib/foo.c}
9777 @item @file{/mnt/cross/project/build/usr/src/foo-1.0/lib/foo.c}
9778 @item @file{/project/build/project/build/usr/src/foo-1.0/lib/foo.c}
9779 @item @file{/home/user/project/build/usr/src/foo-1.0/lib/foo.c}
9780 @item @file{/mnt/cross/foo.c}
9781 @item @file{/project/build/foo.c}
9782 @item @file{/home/user/foo.c}
9786 If the file name in the previous example had been recorded in the
9787 executable as a relative path rather than an absolute path, then the
9788 first look up would not have occurred, but all of the remaining steps
9791 When searching for source files on MS-DOS and MS-Windows, where
9792 absolute paths start with a drive letter (e.g.@:
9793 @file{C:/project/foo.c}), @value{GDBN} will remove the drive letter
9794 from the file name before appending it to a search directory from
9795 @dfn{source path}; for instance if the executable references the
9796 source file @file{C:/project/foo.c} and @dfn{source path} is set to
9797 @file{D:/mnt/cross}, then @value{GDBN} will search in the following
9798 locations for the source file:
9802 @item @file{C:/project/foo.c}
9803 @item @file{D:/mnt/cross/project/foo.c}
9804 @item @file{D:/mnt/cross/foo.c}
9808 Note that the executable search path is @emph{not} used to locate the
9811 Whenever you reset or rearrange the source path, @value{GDBN} clears out
9812 any information it has cached about where source files are found and where
9813 each line is in the file.
9817 When you start @value{GDBN}, its source path includes only @samp{$cdir}
9818 and @samp{$cwd}, in that order.
9819 To add other directories, use the @code{directory} command.
9821 The search path is used to find both program source files and @value{GDBN}
9822 script files (read using the @samp{-command} option and @samp{source} command).
9824 In addition to the source path, @value{GDBN} provides a set of commands
9825 that manage a list of source path substitution rules. A @dfn{substitution
9826 rule} specifies how to rewrite source directories stored in the program's
9827 debug information in case the sources were moved to a different
9828 directory between compilation and debugging. A rule is made of
9829 two strings, the first specifying what needs to be rewritten in
9830 the path, and the second specifying how it should be rewritten.
9831 In @ref{set substitute-path}, we name these two parts @var{from} and
9832 @var{to} respectively. @value{GDBN} does a simple string replacement
9833 of @var{from} with @var{to} at the start of the directory part of the
9834 source file name, and uses that result instead of the original file
9835 name to look up the sources.
9837 Using the previous example, suppose the @file{foo-1.0} tree has been
9838 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
9839 @value{GDBN} to replace @file{/usr/src} in all source path names with
9840 @file{/mnt/cross}. The first lookup will then be
9841 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
9842 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
9843 substitution rule, use the @code{set substitute-path} command
9844 (@pxref{set substitute-path}).
9846 To avoid unexpected substitution results, a rule is applied only if the
9847 @var{from} part of the directory name ends at a directory separator.
9848 For instance, a rule substituting @file{/usr/source} into
9849 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
9850 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
9851 is applied only at the beginning of the directory name, this rule will
9852 not be applied to @file{/root/usr/source/baz.c} either.
9854 In many cases, you can achieve the same result using the @code{directory}
9855 command. However, @code{set substitute-path} can be more efficient in
9856 the case where the sources are organized in a complex tree with multiple
9857 subdirectories. With the @code{directory} command, you need to add each
9858 subdirectory of your project. If you moved the entire tree while
9859 preserving its internal organization, then @code{set substitute-path}
9860 allows you to direct the debugger to all the sources with one single
9863 @code{set substitute-path} is also more than just a shortcut command.
9864 The source path is only used if the file at the original location no
9865 longer exists. On the other hand, @code{set substitute-path} modifies
9866 the debugger behavior to look at the rewritten location instead. So, if
9867 for any reason a source file that is not relevant to your executable is
9868 located at the original location, a substitution rule is the only
9869 method available to point @value{GDBN} at the new location.
9871 @cindex @samp{--with-relocated-sources}
9872 @cindex default source path substitution
9873 You can configure a default source path substitution rule by
9874 configuring @value{GDBN} with the
9875 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
9876 should be the name of a directory under @value{GDBN}'s configured
9877 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
9878 directory names in debug information under @var{dir} will be adjusted
9879 automatically if the installed @value{GDBN} is moved to a new
9880 location. This is useful if @value{GDBN}, libraries or executables
9881 with debug information and corresponding source code are being moved
9885 @item directory @var{dirname} @dots{}
9886 @item dir @var{dirname} @dots{}
9887 Add directory @var{dirname} to the front of the source path. Several
9888 directory names may be given to this command, separated by @samp{:}
9889 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
9890 part of absolute file names) or
9891 whitespace. You may specify a directory that is already in the source
9892 path; this moves it forward, so @value{GDBN} searches it sooner.
9894 The special strings @samp{$cdir} (to refer to the compilation
9895 directory, if one is recorded), and @samp{$cwd} (to refer to the
9896 current working directory) can also be included in the list of
9897 directories @var{dirname}. Though these will already be in the source
9898 path they will be moved forward in the list so @value{GDBN} searches
9902 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
9904 @c RET-repeat for @code{directory} is explicitly disabled, but since
9905 @c repeating it would be a no-op we do not say that. (thanks to RMS)
9907 @item set directories @var{path-list}
9908 @kindex set directories
9909 Set the source path to @var{path-list}.
9910 @samp{$cdir:$cwd} are added if missing.
9912 @item show directories
9913 @kindex show directories
9914 Print the source path: show which directories it contains.
9916 @anchor{set substitute-path}
9917 @item set substitute-path @var{from} @var{to}
9918 @kindex set substitute-path
9919 Define a source path substitution rule, and add it at the end of the
9920 current list of existing substitution rules. If a rule with the same
9921 @var{from} was already defined, then the old rule is also deleted.
9923 For example, if the file @file{/foo/bar/baz.c} was moved to
9924 @file{/mnt/cross/baz.c}, then the command
9927 (@value{GDBP}) set substitute-path /foo/bar /mnt/cross
9931 will tell @value{GDBN} to replace @samp{/foo/bar} with
9932 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
9933 @file{baz.c} even though it was moved.
9935 In the case when more than one substitution rule have been defined,
9936 the rules are evaluated one by one in the order where they have been
9937 defined. The first one matching, if any, is selected to perform
9940 For instance, if we had entered the following commands:
9943 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
9944 (@value{GDBP}) set substitute-path /usr/src /mnt/src
9948 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
9949 @file{/mnt/include/defs.h} by using the first rule. However, it would
9950 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
9951 @file{/mnt/src/lib/foo.c}.
9954 @item unset substitute-path [path]
9955 @kindex unset substitute-path
9956 If a path is specified, search the current list of substitution rules
9957 for a rule that would rewrite that path. Delete that rule if found.
9958 A warning is emitted by the debugger if no rule could be found.
9960 If no path is specified, then all substitution rules are deleted.
9962 @item show substitute-path [path]
9963 @kindex show substitute-path
9964 If a path is specified, then print the source path substitution rule
9965 which would rewrite that path, if any.
9967 If no path is specified, then print all existing source path substitution
9972 If your source path is cluttered with directories that are no longer of
9973 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
9974 versions of source. You can correct the situation as follows:
9978 Use @code{directory} with no argument to reset the source path to its default value.
9981 Use @code{directory} with suitable arguments to reinstall the
9982 directories you want in the source path. You can add all the
9983 directories in one command.
9987 @section Source and Machine Code
9988 @cindex source line and its code address
9990 You can use the command @code{info line} to map source lines to program
9991 addresses (and vice versa), and the command @code{disassemble} to display
9992 a range of addresses as machine instructions. You can use the command
9993 @code{set disassemble-next-line} to set whether to disassemble next
9994 source line when execution stops. When run under @sc{gnu} Emacs
9995 mode, the @code{info line} command causes the arrow to point to the
9996 line specified. Also, @code{info line} prints addresses in symbolic form as
10002 @itemx info line @var{locspec}
10003 Print the starting and ending addresses of the compiled code for the
10004 source lines of the code locations that result from resolving
10005 @var{locspec}. @xref{Location Specifications}, for the various forms
10007 With no @var{locspec}, information about the current source line is
10011 For example, we can use @code{info line} to discover the location of
10012 the object code for the first line of function
10013 @code{m4_changequote}:
10016 (@value{GDBP}) info line m4_changequote
10017 Line 895 of "builtin.c" starts at pc 0x634c <m4_changequote> and \
10018 ends at 0x6350 <m4_changequote+4>.
10022 @cindex code address and its source line
10023 We can also inquire, using @code{*@var{addr}} as the form for
10024 @var{locspec}, what source line covers a particular address
10027 (@value{GDBP}) info line *0x63ff
10028 Line 926 of "builtin.c" starts at pc 0x63e4 <m4_changequote+152> and \
10029 ends at 0x6404 <m4_changequote+184>.
10032 @cindex @code{$_} and @code{info line}
10033 @cindex @code{x} command, default address
10034 @kindex x@r{(examine), and} info line
10035 After @code{info line}, the default address for the @code{x} command
10036 is changed to the starting address of the line, so that @samp{x/i} is
10037 sufficient to begin examining the machine code (@pxref{Memory,
10038 ,Examining Memory}). Also, this address is saved as the value of the
10039 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
10042 @cindex info line, repeated calls
10043 After @code{info line}, using @code{info line} again without
10044 specifying a location will display information about the next source
10047 @anchor{disassemble}
10049 @kindex disassemble
10050 @cindex assembly instructions
10051 @cindex instructions, assembly
10052 @cindex machine instructions
10053 @cindex listing machine instructions
10055 @itemx disassemble /m
10056 @itemx disassemble /s
10057 @itemx disassemble /r
10058 @itemx disassemble /b
10059 This specialized command dumps a range of memory as machine
10060 instructions. It can also print mixed source+disassembly by specifying
10061 the @code{/m} or @code{/s} modifier and print the raw instructions in
10062 hex as well as in symbolic form by specifying the @code{/r} or @code{/b}
10065 Only one of @code{/m} and @code{/s} can be used, attempting to use
10066 both flag will give an error.
10068 Only one of @code{/r} and @code{/b} can be used, attempting to use
10069 both flag will give an error.
10071 The default memory range is the function surrounding the program
10072 counter of the selected frame. A single argument to this command is a
10073 program counter value; @value{GDBN} dumps the function surrounding
10074 this value. When two arguments are given, they should be separated by
10075 a comma, possibly surrounded by whitespace. The arguments specify a
10076 range of addresses to dump, in one of two forms:
10079 @item @var{start},@var{end}
10080 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
10081 @item @var{start},+@var{length}
10082 the addresses from @var{start} (inclusive) to
10083 @code{@var{start}+@var{length}} (exclusive).
10087 When 2 arguments are specified, the name of the function is also
10088 printed (since there could be several functions in the given range).
10090 The argument(s) can be any expression yielding a numeric value, such as
10091 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
10093 If the range of memory being disassembled contains current program counter,
10094 the instruction at that location is shown with a @code{=>} marker.
10097 The following example shows the disassembly of a range of addresses of
10098 HP PA-RISC 2.0 code:
10101 (@value{GDBP}) disas 0x32c4, 0x32e4
10102 Dump of assembler code from 0x32c4 to 0x32e4:
10103 0x32c4 <main+204>: addil 0,dp
10104 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
10105 0x32cc <main+212>: ldil 0x3000,r31
10106 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
10107 0x32d4 <main+220>: ldo 0(r31),rp
10108 0x32d8 <main+224>: addil -0x800,dp
10109 0x32dc <main+228>: ldo 0x588(r1),r26
10110 0x32e0 <main+232>: ldil 0x3000,r31
10111 End of assembler dump.
10114 The following two examples are for RISC-V, and demonstrates the
10115 difference between the @code{/r} and @code{/b} modifiers. First with
10116 @code{/b}, the bytes of the instruction are printed, in hex, in memory
10120 (@value{GDBP}) disassemble /b 0x00010150,0x0001015c
10121 Dump of assembler code from 0x10150 to 0x1015c:
10122 0x00010150 <call_me+4>: 22 dc sw s0,56(sp)
10123 0x00010152 <call_me+6>: 80 00 addi s0,sp,64
10124 0x00010154 <call_me+8>: 23 26 a4 fe sw a0,-20(s0)
10125 0x00010158 <call_me+12>: 23 24 b4 fe sw a1,-24(s0)
10126 End of assembler dump.
10129 In contrast, with @code{/r} the bytes of the instruction are displayed
10130 in the instruction order, for RISC-V this means that the bytes have been
10131 swapped to little-endian order:
10134 (@value{GDBP}) disassemble /r 0x00010150,0x0001015c
10135 Dump of assembler code from 0x10150 to 0x1015c:
10136 0x00010150 <call_me+4>: dc22 sw s0,56(sp)
10137 0x00010152 <call_me+6>: 0080 addi s0,sp,64
10138 0x00010154 <call_me+8>: fea42623 sw a0,-20(s0)
10139 0x00010158 <call_me+12>: feb42423 sw a1,-24(s0)
10140 End of assembler dump.
10143 Here is an example showing mixed source+assembly for Intel x86
10144 with @code{/m} or @code{/s}, when the program is stopped just after
10145 function prologue in a non-optimized function with no inline code.
10148 (@value{GDBP}) disas /m main
10149 Dump of assembler code for function main:
10151 0x08048330 <+0>: push %ebp
10152 0x08048331 <+1>: mov %esp,%ebp
10153 0x08048333 <+3>: sub $0x8,%esp
10154 0x08048336 <+6>: and $0xfffffff0,%esp
10155 0x08048339 <+9>: sub $0x10,%esp
10157 6 printf ("Hello.\n");
10158 => 0x0804833c <+12>: movl $0x8048440,(%esp)
10159 0x08048343 <+19>: call 0x8048284 <puts@@plt>
10163 0x08048348 <+24>: mov $0x0,%eax
10164 0x0804834d <+29>: leave
10165 0x0804834e <+30>: ret
10167 End of assembler dump.
10170 The @code{/m} option is deprecated as its output is not useful when
10171 there is either inlined code or re-ordered code.
10172 The @code{/s} option is the preferred choice.
10173 Here is an example for AMD x86-64 showing the difference between
10174 @code{/m} output and @code{/s} output.
10175 This example has one inline function defined in a header file,
10176 and the code is compiled with @samp{-O2} optimization.
10177 Note how the @code{/m} output is missing the disassembly of
10178 several instructions that are present in the @code{/s} output.
10208 (@value{GDBP}) disas /m main
10209 Dump of assembler code for function main:
10213 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
10214 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
10218 0x000000000040041d <+29>: xor %eax,%eax
10219 0x000000000040041f <+31>: retq
10220 0x0000000000400420 <+32>: add %eax,%eax
10221 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
10223 End of assembler dump.
10224 (@value{GDBP}) disas /s main
10225 Dump of assembler code for function main:
10229 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
10233 0x0000000000400406 <+6>: test %eax,%eax
10234 0x0000000000400408 <+8>: js 0x400420 <main+32>
10239 0x000000000040040a <+10>: lea 0xa(%rax),%edx
10240 0x000000000040040d <+13>: test %eax,%eax
10241 0x000000000040040f <+15>: mov $0x1,%eax
10242 0x0000000000400414 <+20>: cmovne %edx,%eax
10246 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
10250 0x000000000040041d <+29>: xor %eax,%eax
10251 0x000000000040041f <+31>: retq
10255 0x0000000000400420 <+32>: add %eax,%eax
10256 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
10257 End of assembler dump.
10260 Here is another example showing raw instructions in hex for AMD x86-64,
10263 (@value{GDBP}) disas /r 0x400281,+10
10264 Dump of assembler code from 0x400281 to 0x40028b:
10265 0x0000000000400281: 38 36 cmp %dh,(%rsi)
10266 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
10267 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
10268 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
10269 End of assembler dump.
10272 Note that the @samp{disassemble} command's address arguments are
10273 specified using expressions in your programming language
10274 (@pxref{Expressions, ,Expressions}), not location specs
10275 (@pxref{Location Specifications}). So, for example, if you want to
10276 disassemble function @code{bar} in file @file{foo.c}, you must type
10277 @samp{disassemble 'foo.c'::bar} and not @samp{disassemble foo.c:bar}.
10279 Some architectures have more than one commonly-used set of instruction
10280 mnemonics or other syntax.
10282 For programs that were dynamically linked and use shared libraries,
10283 instructions that call functions or branch to locations in the shared
10284 libraries might show a seemingly bogus location---it's actually a
10285 location of the relocation table. On some architectures, @value{GDBN}
10286 might be able to resolve these to actual function names.
10289 @kindex set disassembler-options
10290 @cindex disassembler options
10291 @item set disassembler-options @var{option1}[,@var{option2}@dots{}]
10292 This command controls the passing of target specific information to
10293 the disassembler. For a list of valid options, please refer to the
10294 @code{-M}/@code{--disassembler-options} section of the @samp{objdump}
10295 manual and/or the output of @kbd{objdump --help}
10296 (@pxref{objdump,,objdump,binutils,The GNU Binary Utilities}).
10297 The default value is the empty string.
10299 If it is necessary to specify more than one disassembler option, then
10300 multiple options can be placed together into a comma separated list.
10301 Currently this command is only supported on targets ARC, ARM, MIPS,
10304 @kindex show disassembler-options
10305 @item show disassembler-options
10306 Show the current setting of the disassembler options.
10310 @kindex set disassembly-flavor
10311 @cindex Intel disassembly flavor
10312 @cindex AT&T disassembly flavor
10313 @item set disassembly-flavor @var{instruction-set}
10314 Select the instruction set to use when disassembling the
10315 program via the @code{disassemble} or @code{x/i} commands.
10317 Currently this command is only defined for the Intel x86 family. You
10318 can set @var{instruction-set} to either @code{intel} or @code{att}.
10319 The default is @code{att}, the AT&T flavor used by default by Unix
10320 assemblers for x86-based targets.
10322 @kindex show disassembly-flavor
10323 @item show disassembly-flavor
10324 Show the current setting of the disassembly flavor.
10328 @kindex set disassemble-next-line
10329 @kindex show disassemble-next-line
10330 @item set disassemble-next-line
10331 @itemx show disassemble-next-line
10332 Control whether or not @value{GDBN} will disassemble the next source
10333 line or instruction when execution stops. If ON, @value{GDBN} will
10334 display disassembly of the next source line when execution of the
10335 program being debugged stops. This is @emph{in addition} to
10336 displaying the source line itself, which @value{GDBN} always does if
10337 possible. If the next source line cannot be displayed for some reason
10338 (e.g., if @value{GDBN} cannot find the source file, or there's no line
10339 info in the debug info), @value{GDBN} will display disassembly of the
10340 next @emph{instruction} instead of showing the next source line. If
10341 AUTO, @value{GDBN} will display disassembly of next instruction only
10342 if the source line cannot be displayed. This setting causes
10343 @value{GDBN} to display some feedback when you step through a function
10344 with no line info or whose source file is unavailable. The default is
10345 OFF, which means never display the disassembly of the next line or
10349 @node Disable Reading Source
10350 @section Disable Reading Source Code
10351 @cindex source code, disable access
10353 In some cases it can be desirable to prevent @value{GDBN} from
10354 accessing source code files. One case where this might be desirable
10355 is if the source code files are located over a slow network
10358 The following command can be used to control whether @value{GDBN}
10359 should access source code files or not:
10362 @kindex set source open
10363 @kindex show source open
10364 @item set source open @r{[}on@r{|}off@r{]}
10365 @itemx show source open
10366 When this option is @code{on}, which is the default, @value{GDBN} will
10367 access source code files when needed, for example to print source
10368 lines when @value{GDBN} stops, or in response to the @code{list}
10371 When this option is @code{off}, @value{GDBN} will not access source
10376 @chapter Examining Data
10378 @cindex printing data
10379 @cindex examining data
10382 The usual way to examine data in your program is with the @code{print}
10383 command (abbreviated @code{p}), or its synonym @code{inspect}. It
10384 evaluates and prints the value of an expression of the language your
10385 program is written in (@pxref{Languages, ,Using @value{GDBN} with
10386 Different Languages}). It may also print the expression using a
10387 Python-based pretty-printer (@pxref{Pretty Printing}).
10390 @item print [[@var{options}] --] @var{expr}
10391 @itemx print [[@var{options}] --] /@var{f} @var{expr}
10392 @var{expr} is an expression (in the source language). By default the
10393 value of @var{expr} is printed in a format appropriate to its data type;
10394 you can choose a different format by specifying @samp{/@var{f}}, where
10395 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
10398 @anchor{print options}
10399 The @code{print} command supports a number of options that allow
10400 overriding relevant global print settings as set by @code{set print}
10404 @item -address [@code{on}|@code{off}]
10405 Set printing of addresses.
10406 Related setting: @ref{set print address}.
10408 @item -array [@code{on}|@code{off}]
10409 Pretty formatting of arrays.
10410 Related setting: @ref{set print array}.
10412 @item -array-indexes [@code{on}|@code{off}]
10413 Set printing of array indexes.
10414 Related setting: @ref{set print array-indexes}.
10416 @item -characters @var{number-of-characters}|@code{elements}|@code{unlimited}
10417 Set limit on string characters to print. The value @code{elements}
10418 causes the limit on array elements to print to be used. The value
10419 @code{unlimited} causes there to be no limit. Related setting:
10420 @ref{set print characters}.
10422 @item -elements @var{number-of-elements}|@code{unlimited}
10423 Set limit on array elements and optionally string characters to print.
10424 See @ref{set print characters}, and the @code{-characters} option above
10425 for when this option applies to strings. The value @code{unlimited}
10426 causes there to be no limit. @xref{set print elements}, for a related
10429 @item -max-depth @var{depth}|@code{unlimited}
10430 Set the threshold after which nested structures are replaced with
10431 ellipsis. Related setting: @ref{set print max-depth}.
10433 @item -nibbles [@code{on}|@code{off}]
10434 Set whether to print binary values in groups of four bits, known
10435 as ``nibbles''. @xref{set print nibbles}.
10437 @item -memory-tag-violations [@code{on}|@code{off}]
10438 Set printing of additional information about memory tag violations.
10439 @xref{set print memory-tag-violations}.
10441 @item -null-stop [@code{on}|@code{off}]
10442 Set printing of char arrays to stop at first null char. Related
10443 setting: @ref{set print null-stop}.
10445 @item -object [@code{on}|@code{off}]
10446 Set printing C@t{++} virtual function tables. Related setting:
10447 @ref{set print object}.
10449 @item -pretty [@code{on}|@code{off}]
10450 Set pretty formatting of structures. Related setting: @ref{set print
10453 @item -raw-values [@code{on}|@code{off}]
10454 Set whether to print values in raw form, bypassing any
10455 pretty-printers for that value. Related setting: @ref{set print
10458 @item -repeats @var{number-of-repeats}|@code{unlimited}
10459 Set threshold for repeated print elements. @code{unlimited} causes
10460 all elements to be individually printed. Related setting: @ref{set
10463 @item -static-members [@code{on}|@code{off}]
10464 Set printing C@t{++} static members. Related setting: @ref{set print
10467 @item -symbol [@code{on}|@code{off}]
10468 Set printing of symbol names when printing pointers. Related setting:
10469 @ref{set print symbol}.
10471 @item -union [@code{on}|@code{off}]
10472 Set printing of unions interior to structures. Related setting:
10473 @ref{set print union}.
10475 @item -vtbl [@code{on}|@code{off}]
10476 Set printing of C++ virtual function tables. Related setting:
10477 @ref{set print vtbl}.
10480 Because the @code{print} command accepts arbitrary expressions which
10481 may look like options (including abbreviations), if you specify any
10482 command option, then you must use a double dash (@code{--}) to mark
10483 the end of option processing.
10485 For example, this prints the value of the @code{-p} expression:
10488 (@value{GDBP}) print -p
10491 While this repeats the last value in the value history (see below)
10492 with the @code{-pretty} option in effect:
10495 (@value{GDBP}) print -p --
10498 Here is an example including both on option and an expression:
10502 (@value{GDBP}) print -pretty -- *myptr
10514 @item print [@var{options}]
10515 @itemx print [@var{options}] /@var{f}
10516 @cindex reprint the last value
10517 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
10518 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
10519 conveniently inspect the same value in an alternative format.
10522 If the architecture supports memory tagging, the @code{print} command will
10523 display pointer/memory tag mismatches if what is being printed is a pointer
10524 or reference type. @xref{Memory Tagging}.
10526 A more low-level way of examining data is with the @code{x} command.
10527 It examines data in memory at a specified address and prints it in a
10528 specified format. @xref{Memory, ,Examining Memory}.
10530 If you are interested in information about types, or about how the
10531 fields of a struct or a class are declared, use the @code{ptype @var{expr}}
10532 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
10535 @cindex exploring hierarchical data structures
10537 Another way of examining values of expressions and type information is
10538 through the Python extension command @code{explore} (available only if
10539 the @value{GDBN} build is configured with @code{--with-python}). It
10540 offers an interactive way to start at the highest level (or, the most
10541 abstract level) of the data type of an expression (or, the data type
10542 itself) and explore all the way down to leaf scalar values/fields
10543 embedded in the higher level data types.
10546 @item explore @var{arg}
10547 @var{arg} is either an expression (in the source language), or a type
10548 visible in the current context of the program being debugged.
10551 The working of the @code{explore} command can be illustrated with an
10552 example. If a data type @code{struct ComplexStruct} is defined in your
10556 struct SimpleStruct
10562 struct ComplexStruct
10564 struct SimpleStruct *ss_p;
10570 followed by variable declarations as
10573 struct SimpleStruct ss = @{ 10, 1.11 @};
10574 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
10578 then, the value of the variable @code{cs} can be explored using the
10579 @code{explore} command as follows.
10582 (@value{GDBP}) explore cs
10583 The value of `cs' is a struct/class of type `struct ComplexStruct' with
10584 the following fields:
10586 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
10587 arr = <Enter 1 to explore this field of type `int [10]'>
10589 Enter the field number of choice:
10593 Since the fields of @code{cs} are not scalar values, you are being
10594 prompted to chose the field you want to explore. Let's say you choose
10595 the field @code{ss_p} by entering @code{0}. Then, since this field is a
10596 pointer, you will be asked if it is pointing to a single value. From
10597 the declaration of @code{cs} above, it is indeed pointing to a single
10598 value, hence you enter @code{y}. If you enter @code{n}, then you will
10599 be asked if it were pointing to an array of values, in which case this
10600 field will be explored as if it were an array.
10603 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
10604 Continue exploring it as a pointer to a single value [y/n]: y
10605 The value of `*(cs.ss_p)' is a struct/class of type `struct
10606 SimpleStruct' with the following fields:
10608 i = 10 .. (Value of type `int')
10609 d = 1.1100000000000001 .. (Value of type `double')
10611 Press enter to return to parent value:
10615 If the field @code{arr} of @code{cs} was chosen for exploration by
10616 entering @code{1} earlier, then since it is as array, you will be
10617 prompted to enter the index of the element in the array that you want
10621 `cs.arr' is an array of `int'.
10622 Enter the index of the element you want to explore in `cs.arr': 5
10624 `(cs.arr)[5]' is a scalar value of type `int'.
10628 Press enter to return to parent value:
10631 In general, at any stage of exploration, you can go deeper towards the
10632 leaf values by responding to the prompts appropriately, or hit the
10633 return key to return to the enclosing data structure (the @i{higher}
10634 level data structure).
10636 Similar to exploring values, you can use the @code{explore} command to
10637 explore types. Instead of specifying a value (which is typically a
10638 variable name or an expression valid in the current context of the
10639 program being debugged), you specify a type name. If you consider the
10640 same example as above, your can explore the type
10641 @code{struct ComplexStruct} by passing the argument
10642 @code{struct ComplexStruct} to the @code{explore} command.
10645 (@value{GDBP}) explore struct ComplexStruct
10649 By responding to the prompts appropriately in the subsequent interactive
10650 session, you can explore the type @code{struct ComplexStruct} in a
10651 manner similar to how the value @code{cs} was explored in the above
10654 The @code{explore} command also has two sub-commands,
10655 @code{explore value} and @code{explore type}. The former sub-command is
10656 a way to explicitly specify that value exploration of the argument is
10657 being invoked, while the latter is a way to explicitly specify that type
10658 exploration of the argument is being invoked.
10661 @item explore value @var{expr}
10662 @cindex explore value
10663 This sub-command of @code{explore} explores the value of the
10664 expression @var{expr} (if @var{expr} is an expression valid in the
10665 current context of the program being debugged). The behavior of this
10666 command is identical to that of the behavior of the @code{explore}
10667 command being passed the argument @var{expr}.
10669 @item explore type @var{arg}
10670 @cindex explore type
10671 This sub-command of @code{explore} explores the type of @var{arg} (if
10672 @var{arg} is a type visible in the current context of program being
10673 debugged), or the type of the value/expression @var{arg} (if @var{arg}
10674 is an expression valid in the current context of the program being
10675 debugged). If @var{arg} is a type, then the behavior of this command is
10676 identical to that of the @code{explore} command being passed the
10677 argument @var{arg}. If @var{arg} is an expression, then the behavior of
10678 this command will be identical to that of the @code{explore} command
10679 being passed the type of @var{arg} as the argument.
10683 * Expressions:: Expressions
10684 * Ambiguous Expressions:: Ambiguous Expressions
10685 * Variables:: Program variables
10686 * Arrays:: Artificial arrays
10687 * Output Formats:: Output formats
10688 * Memory:: Examining memory
10689 * Memory Tagging:: Memory Tagging
10690 * Auto Display:: Automatic display
10691 * Print Settings:: Print settings
10692 * Pretty Printing:: Python pretty printing
10693 * Value History:: Value history
10694 * Convenience Vars:: Convenience variables
10695 * Convenience Funs:: Convenience functions
10696 * Registers:: Registers
10697 * Floating Point Hardware:: Floating point hardware
10698 * Vector Unit:: Vector Unit
10699 * OS Information:: Auxiliary data provided by operating system
10700 * Memory Region Attributes:: Memory region attributes
10701 * Dump/Restore Files:: Copy between memory and a file
10702 * Core File Generation:: Cause a program dump its core
10703 * Character Sets:: Debugging programs that use a different
10704 character set than GDB does
10705 * Caching Target Data:: Data caching for targets
10706 * Searching Memory:: Searching memory for a sequence of bytes
10707 * Value Sizes:: Managing memory allocated for values
10711 @section Expressions
10713 @cindex expressions
10714 @code{print} and many other @value{GDBN} commands accept an expression and
10715 compute its value. Any kind of constant, variable or operator defined
10716 by the programming language you are using is valid in an expression in
10717 @value{GDBN}. This includes conditional expressions, function calls,
10718 casts, and string constants. It also includes preprocessor macros, if
10719 you compiled your program to include this information; see
10722 @cindex arrays in expressions
10723 @value{GDBN} supports array constants in expressions input by
10724 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
10725 you can use the command @code{print @{1, 2, 3@}} to create an array
10726 of three integers. If you pass an array to a function or assign it
10727 to a program variable, @value{GDBN} copies the array to memory that
10728 is @code{malloc}ed in the target program.
10730 Because C is so widespread, most of the expressions shown in examples in
10731 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
10732 Languages}, for information on how to use expressions in other
10735 In this section, we discuss operators that you can use in @value{GDBN}
10736 expressions regardless of your programming language.
10738 @cindex casts, in expressions
10739 Casts are supported in all languages, not just in C, because it is so
10740 useful to cast a number into a pointer in order to examine a structure
10741 at that address in memory.
10742 @c FIXME: casts supported---Mod2 true?
10744 @value{GDBN} supports these operators, in addition to those common
10745 to programming languages:
10749 @samp{@@} is a binary operator for treating parts of memory as arrays.
10750 @xref{Arrays, ,Artificial Arrays}, for more information.
10753 @samp{::} allows you to specify a variable in terms of the file or
10754 function where it is defined. @xref{Variables, ,Program Variables}.
10756 @cindex @{@var{type}@}
10757 @cindex type casting memory
10758 @cindex memory, viewing as typed object
10759 @cindex casts, to view memory
10760 @item @{@var{type}@} @var{addr}
10761 Refers to an object of type @var{type} stored at address @var{addr} in
10762 memory. The address @var{addr} may be any expression whose value is
10763 an integer or pointer (but parentheses are required around binary
10764 operators, just as in a cast). This construct is allowed regardless
10765 of what kind of data is normally supposed to reside at @var{addr}.
10768 @node Ambiguous Expressions
10769 @section Ambiguous Expressions
10770 @cindex ambiguous expressions
10772 Expressions can sometimes contain some ambiguous elements. For instance,
10773 some programming languages (notably Ada, C@t{++} and Objective-C) permit
10774 a single function name to be defined several times, for application in
10775 different contexts. This is called @dfn{overloading}. Another example
10776 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
10777 templates and is typically instantiated several times, resulting in
10778 the same function name being defined in different contexts.
10780 In some cases and depending on the language, it is possible to adjust
10781 the expression to remove the ambiguity. For instance in C@t{++}, you
10782 can specify the signature of the function you want to break on, as in
10783 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
10784 qualified name of your function often makes the expression unambiguous
10787 When an ambiguity that needs to be resolved is detected, the debugger
10788 has the capability to display a menu of numbered choices for each
10789 possibility, and then waits for the selection with the prompt @samp{>}.
10790 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
10791 aborts the current command. If the command in which the expression was
10792 used allows more than one choice to be selected, the next option in the
10793 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
10796 For example, the following session excerpt shows an attempt to set a
10797 breakpoint at the overloaded symbol @code{String::after}.
10798 We choose three particular definitions of that function name:
10800 @c FIXME! This is likely to change to show arg type lists, at least
10803 (@value{GDBP}) b String::after
10806 [2] file:String.cc; line number:867
10807 [3] file:String.cc; line number:860
10808 [4] file:String.cc; line number:875
10809 [5] file:String.cc; line number:853
10810 [6] file:String.cc; line number:846
10811 [7] file:String.cc; line number:735
10813 Breakpoint 1 at 0xb26c: file String.cc, line 867.
10814 Breakpoint 2 at 0xb344: file String.cc, line 875.
10815 Breakpoint 3 at 0xafcc: file String.cc, line 846.
10816 Multiple breakpoints were set.
10817 Use the "delete" command to delete unwanted
10824 @kindex set multiple-symbols
10825 @item set multiple-symbols @var{mode}
10826 @cindex multiple-symbols menu
10828 This option allows you to adjust the debugger behavior when an expression
10831 By default, @var{mode} is set to @code{all}. If the command with which
10832 the expression is used allows more than one choice, then @value{GDBN}
10833 automatically selects all possible choices. For instance, inserting
10834 a breakpoint on a function using an ambiguous name results in a breakpoint
10835 inserted on each possible match. However, if a unique choice must be made,
10836 then @value{GDBN} uses the menu to help you disambiguate the expression.
10837 For instance, printing the address of an overloaded function will result
10838 in the use of the menu.
10840 When @var{mode} is set to @code{ask}, the debugger always uses the menu
10841 when an ambiguity is detected.
10843 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
10844 an error due to the ambiguity and the command is aborted.
10846 @kindex show multiple-symbols
10847 @item show multiple-symbols
10848 Show the current value of the @code{multiple-symbols} setting.
10852 @section Program Variables
10854 The most common kind of expression to use is the name of a variable
10857 Variables in expressions are understood in the selected stack frame
10858 (@pxref{Selection, ,Selecting a Frame}); they must be either:
10862 global (or file-static)
10869 visible according to the scope rules of the
10870 programming language from the point of execution in that frame
10873 @noindent This means that in the function
10888 you can examine and use the variable @code{a} whenever your program is
10889 executing within the function @code{foo}, but you can only use or
10890 examine the variable @code{b} while your program is executing inside
10891 the block where @code{b} is declared.
10893 @cindex variable name conflict
10894 There is an exception: you can refer to a variable or function whose
10895 scope is a single source file even if the current execution point is not
10896 in this file. But it is possible to have more than one such variable or
10897 function with the same name (in different source files). If that
10898 happens, referring to that name has unpredictable effects. If you wish,
10899 you can specify a static variable in a particular function or file by
10900 using the colon-colon (@code{::}) notation:
10902 @cindex colon-colon, context for variables/functions
10904 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
10905 @cindex @code{::}, context for variables/functions
10908 @var{file}::@var{variable}
10909 @var{function}::@var{variable}
10913 Here @var{file} or @var{function} is the name of the context for the
10914 static @var{variable}. In the case of file names, you can use quotes to
10915 make sure @value{GDBN} parses the file name as a single word---for example,
10916 to print a global value of @code{x} defined in @file{f2.c}:
10919 (@value{GDBP}) p 'f2.c'::x
10922 The @code{::} notation is normally used for referring to
10923 static variables, since you typically disambiguate uses of local variables
10924 in functions by selecting the appropriate frame and using the
10925 simple name of the variable. However, you may also use this notation
10926 to refer to local variables in frames enclosing the selected frame:
10935 process (a); /* Stop here */
10946 For example, if there is a breakpoint at the commented line,
10947 here is what you might see
10948 when the program stops after executing the call @code{bar(0)}:
10953 (@value{GDBP}) p bar::a
10955 (@value{GDBP}) up 2
10956 #2 0x080483d0 in foo (a=5) at foobar.c:12
10959 (@value{GDBP}) p bar::a
10963 @cindex C@t{++} scope resolution
10964 These uses of @samp{::} are very rarely in conflict with the very
10965 similar use of the same notation in C@t{++}. When they are in
10966 conflict, the C@t{++} meaning takes precedence; however, this can be
10967 overridden by quoting the file or function name with single quotes.
10969 For example, suppose the program is stopped in a method of a class
10970 that has a field named @code{includefile}, and there is also an
10971 include file named @file{includefile} that defines a variable,
10972 @code{some_global}.
10975 (@value{GDBP}) p includefile
10977 (@value{GDBP}) p includefile::some_global
10978 A syntax error in expression, near `'.
10979 (@value{GDBP}) p 'includefile'::some_global
10983 @cindex wrong values
10984 @cindex variable values, wrong
10985 @cindex function entry/exit, wrong values of variables
10986 @cindex optimized code, wrong values of variables
10988 @emph{Warning:} Occasionally, a local variable may appear to have the
10989 wrong value at certain points in a function---just after entry to a new
10990 scope, and just before exit.
10992 You may see this problem when you are stepping by machine instructions.
10993 This is because, on most machines, it takes more than one instruction to
10994 set up a stack frame (including local variable definitions); if you are
10995 stepping by machine instructions, variables may appear to have the wrong
10996 values until the stack frame is completely built. On exit, it usually
10997 also takes more than one machine instruction to destroy a stack frame;
10998 after you begin stepping through that group of instructions, local
10999 variable definitions may be gone.
11001 This may also happen when the compiler does significant optimizations.
11002 To be sure of always seeing accurate values, turn off all optimization
11005 @cindex ``No symbol "foo" in current context''
11006 Another possible effect of compiler optimizations is to optimize
11007 unused variables out of existence, or assign variables to registers (as
11008 opposed to memory addresses). Depending on the support for such cases
11009 offered by the debug info format used by the compiler, @value{GDBN}
11010 might not be able to display values for such local variables. If that
11011 happens, @value{GDBN} will print a message like this:
11014 No symbol "foo" in current context.
11017 To solve such problems, either recompile without optimizations, or use a
11018 different debug info format, if the compiler supports several such
11019 formats. @xref{Compilation}, for more information on choosing compiler
11020 options. @xref{C, ,C and C@t{++}}, for more information about debug
11021 info formats that are best suited to C@t{++} programs.
11023 If you ask to print an object whose contents are unknown to
11024 @value{GDBN}, e.g., because its data type is not completely specified
11025 by the debug information, @value{GDBN} will say @samp{<incomplete
11026 type>}. @xref{Symbols, incomplete type}, for more about this.
11028 @cindex no debug info variables
11029 If you try to examine or use the value of a (global) variable for
11030 which @value{GDBN} has no type information, e.g., because the program
11031 includes no debug information, @value{GDBN} displays an error message.
11032 @xref{Symbols, unknown type}, for more about unknown types. If you
11033 cast the variable to its declared type, @value{GDBN} gets the
11034 variable's value using the cast-to type as the variable's type. For
11035 example, in a C program:
11038 (@value{GDBP}) p var
11039 'var' has unknown type; cast it to its declared type
11040 (@value{GDBP}) p (float) var
11044 If you append @kbd{@@entry} string to a function parameter name you get its
11045 value at the time the function got called. If the value is not available an
11046 error message is printed. Entry values are available only with some compilers.
11047 Entry values are normally also printed at the function parameter list according
11048 to @ref{set print entry-values}.
11051 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
11053 (@value{GDBP}) next
11055 (@value{GDBP}) print i
11057 (@value{GDBP}) print i@@entry
11061 Strings are identified as arrays of @code{char} values without specified
11062 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
11063 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
11064 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
11065 defines literal string type @code{"char"} as @code{char} without a sign.
11070 signed char var1[] = "A";
11073 You get during debugging
11075 (@value{GDBP}) print var0
11077 (@value{GDBP}) print var1
11078 $2 = @{65 'A', 0 '\0'@}
11082 @section Artificial Arrays
11084 @cindex artificial array
11086 @kindex @@@r{, referencing memory as an array}
11087 It is often useful to print out several successive objects of the
11088 same type in memory; a section of an array, or an array of
11089 dynamically determined size for which only a pointer exists in the
11092 You can do this by referring to a contiguous span of memory as an
11093 @dfn{artificial array}, using the binary operator @samp{@@}. The left
11094 operand of @samp{@@} should be the first element of the desired array
11095 and be an individual object. The right operand should be the desired length
11096 of the array. The result is an array value whose elements are all of
11097 the type of the left argument. The first element is actually the left
11098 argument; the second element comes from bytes of memory immediately
11099 following those that hold the first element, and so on. Here is an
11100 example. If a program says
11103 int *array = (int *) malloc (len * sizeof (int));
11107 you can print the contents of @code{array} with
11113 The left operand of @samp{@@} must reside in memory. Array values made
11114 with @samp{@@} in this way behave just like other arrays in terms of
11115 subscripting, and are coerced to pointers when used in expressions.
11116 Artificial arrays most often appear in expressions via the value history
11117 (@pxref{Value History, ,Value History}), after printing one out.
11119 Another way to create an artificial array is to use a cast.
11120 This re-interprets a value as if it were an array.
11121 The value need not be in memory:
11123 (@value{GDBP}) p/x (short[2])0x12345678
11124 $1 = @{0x1234, 0x5678@}
11127 As a convenience, if you leave the array length out (as in
11128 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
11129 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
11131 (@value{GDBP}) p/x (short[])0x12345678
11132 $2 = @{0x1234, 0x5678@}
11135 Sometimes the artificial array mechanism is not quite enough; in
11136 moderately complex data structures, the elements of interest may not
11137 actually be adjacent---for example, if you are interested in the values
11138 of pointers in an array. One useful work-around in this situation is
11139 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
11140 Variables}) as a counter in an expression that prints the first
11141 interesting value, and then repeat that expression via @key{RET}. For
11142 instance, suppose you have an array @code{dtab} of pointers to
11143 structures, and you are interested in the values of a field @code{fv}
11144 in each structure. Here is an example of what you might type:
11154 @node Output Formats
11155 @section Output Formats
11157 @cindex formatted output
11158 @cindex output formats
11159 By default, @value{GDBN} prints a value according to its data type. Sometimes
11160 this is not what you want. For example, you might want to print a number
11161 in hex, or a pointer in decimal. Or you might want to view data in memory
11162 at a certain address as a character string or as an instruction. To do
11163 these things, specify an @dfn{output format} when you print a value.
11165 The simplest use of output formats is to say how to print a value
11166 already computed. This is done by starting the arguments of the
11167 @code{print} command with a slash and a format letter. The format
11168 letters supported are:
11172 Print the binary representation of the value in hexadecimal.
11175 Print the binary representation of the value in decimal.
11178 Print the binary representation of the value as an decimal, as if it
11182 Print the binary representation of the value in octal.
11185 Print the binary representation of the value in binary. The letter
11186 @samp{t} stands for ``two''. @footnote{@samp{b} cannot be used
11187 because these format letters are also used with the @code{x} command,
11188 where @samp{b} stands for ``byte''; see @ref{Memory,,Examining
11192 @cindex unknown address, locating
11193 @cindex locate address
11194 Print as an address, both absolute in hexadecimal and as an offset from
11195 the nearest preceding symbol. You can use this format used to discover
11196 where (in what function) an unknown address is located:
11199 (@value{GDBP}) p/a 0x54320
11200 $3 = 0x54320 <_initialize_vx+396>
11204 The command @code{info symbol 0x54320} yields similar results.
11205 @xref{Symbols, info symbol}.
11208 Cast the value to an integer (unlike other formats, this does not just
11209 reinterpret the underlying bits) and print it as a character constant.
11210 This prints both the numerical value and its character representation.
11211 The character representation is replaced with the octal escape
11212 @samp{\nnn} for characters outside the 7-bit @sc{ascii} range.
11214 Without this format, @value{GDBN} displays @code{char},
11215 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
11216 constants. Single-byte members of vectors are displayed as integer
11220 Regard the bits of the value as a floating point number and print
11221 using typical floating point syntax.
11224 @cindex printing strings
11225 @cindex printing byte arrays
11226 Regard as a string, if possible. With this format, pointers to single-byte
11227 data are displayed as null-terminated strings and arrays of single-byte data
11228 are displayed as fixed-length strings. Other values are displayed in their
11231 Without this format, @value{GDBN} displays pointers to and arrays of
11232 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
11233 strings. Single-byte members of a vector are displayed as an integer
11237 Like @samp{x} formatting, the value is treated as an integer and
11238 printed as hexadecimal, but leading zeros are printed to pad the value
11239 to the size of the integer type.
11242 @cindex raw printing
11243 Print using the @samp{raw} formatting. By default, @value{GDBN} will
11244 use a Python-based pretty-printer, if one is available (@pxref{Pretty
11245 Printing}). This typically results in a higher-level display of the
11246 value's contents. The @samp{r} format bypasses any Python
11247 pretty-printer which might exist.
11250 For example, to print the program counter in hex (@pxref{Registers}), type
11257 Note that no space is required before the slash; this is because command
11258 names in @value{GDBN} cannot contain a slash.
11260 To reprint the last value in the value history with a different format,
11261 you can use the @code{print} command with just a format and no
11262 expression. For example, @samp{p/x} reprints the last value in hex.
11265 @section Examining Memory
11267 You can use the command @code{x} (for ``examine'') to examine memory in
11268 any of several formats, independently of your program's data types.
11270 @cindex examining memory
11272 @kindex x @r{(examine memory)}
11273 @item x/@var{nfu} @var{addr}
11274 @itemx x @var{addr}
11276 Use the @code{x} command to examine memory.
11279 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
11280 much memory to display and how to format it; @var{addr} is an
11281 expression giving the address where you want to start displaying memory.
11282 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
11283 Several commands set convenient defaults for @var{addr}.
11286 @item @var{n}, the repeat count
11287 The repeat count is a decimal integer; the default is 1. It specifies
11288 how much memory (counting by units @var{u}) to display. If a negative
11289 number is specified, memory is examined backward from @var{addr}.
11290 @c This really is **decimal**; unaffected by 'set radix' as of GDB
11293 @item @var{f}, the display format
11294 The display format is one of the formats used by @code{print}
11295 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
11296 @samp{f}, @samp{s}), @samp{i} (for machine instructions) and
11297 @samp{m} (for displaying memory tags).
11298 The default is @samp{x} (hexadecimal) initially. The default changes
11299 each time you use either @code{x} or @code{print}.
11301 @item @var{u}, the unit size
11302 The unit size is any of
11308 Halfwords (two bytes).
11310 Words (four bytes). This is the initial default.
11312 Giant words (eight bytes).
11315 Each time you specify a unit size with @code{x}, that size becomes the
11316 default unit the next time you use @code{x}. For the @samp{i} format,
11317 the unit size is ignored and is normally not written. For the @samp{s} format,
11318 the unit size defaults to @samp{b}, unless it is explicitly given.
11319 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
11320 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
11321 Note that the results depend on the programming language of the
11322 current compilation unit. If the language is C, the @samp{s}
11323 modifier will use the UTF-16 encoding while @samp{w} will use
11324 UTF-32. The encoding is set by the programming language and cannot
11327 @item @var{addr}, starting display address
11328 @var{addr} is the address where you want @value{GDBN} to begin displaying
11329 memory. The expression need not have a pointer value (though it may);
11330 it is always interpreted as an integer address of a byte of memory.
11331 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
11332 @var{addr} is usually just after the last address examined---but several
11333 other commands also set the default address: @code{info breakpoints} (to
11334 the address of the last breakpoint listed), @code{info line} (to the
11335 starting address of a line), and @code{print} (if you use it to display
11336 a value from memory).
11339 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
11340 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
11341 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
11342 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
11343 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
11345 You can also specify a negative repeat count to examine memory backward
11346 from the given address. For example, @samp{x/-3uh 0x54320} prints three
11347 halfwords (@code{h}) at @code{0x5431a}, @code{0x5431c}, and @code{0x5431e}.
11349 Since the letters indicating unit sizes are all distinct from the
11350 letters specifying output formats, you do not have to remember whether
11351 unit size or format comes first; either order works. The output
11352 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
11353 (However, the count @var{n} must come first; @samp{wx4} does not work.)
11355 Even though the unit size @var{u} is ignored for the formats @samp{s}
11356 and @samp{i}, you might still want to use a count @var{n}; for example,
11357 @samp{3i} specifies that you want to see three machine instructions,
11358 including any operands. For convenience, especially when used with
11359 the @code{display} command, the @samp{i} format also prints branch delay
11360 slot instructions, if any, beyond the count specified, which immediately
11361 follow the last instruction that is within the count. The command
11362 @code{disassemble} gives an alternative way of inspecting machine
11363 instructions; see @ref{Machine Code,,Source and Machine Code}.
11365 If a negative repeat count is specified for the formats @samp{s} or @samp{i},
11366 the command displays null-terminated strings or instructions before the given
11367 address as many as the absolute value of the given number. For the @samp{i}
11368 format, we use line number information in the debug info to accurately locate
11369 instruction boundaries while disassembling backward. If line info is not
11370 available, the command stops examining memory with an error message.
11372 All the defaults for the arguments to @code{x} are designed to make it
11373 easy to continue scanning memory with minimal specifications each time
11374 you use @code{x}. For example, after you have inspected three machine
11375 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
11376 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
11377 the repeat count @var{n} is used again; the other arguments default as
11378 for successive uses of @code{x}.
11380 When examining machine instructions, the instruction at current program
11381 counter is shown with a @code{=>} marker. For example:
11384 (@value{GDBP}) x/5i $pc-6
11385 0x804837f <main+11>: mov %esp,%ebp
11386 0x8048381 <main+13>: push %ecx
11387 0x8048382 <main+14>: sub $0x4,%esp
11388 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
11389 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
11392 If the architecture supports memory tagging, the tags can be displayed by
11393 using @samp{m}. @xref{Memory Tagging}.
11395 The information will be displayed once per granule size
11396 (the amount of bytes a particular memory tag covers). For example, AArch64
11397 has a granule size of 16 bytes, so it will display a tag every 16 bytes.
11399 Due to the way @value{GDBN} prints information with the @code{x} command (not
11400 aligned to a particular boundary), the tag information will refer to the
11401 initial address displayed on a particular line. If a memory tag boundary
11402 is crossed in the middle of a line displayed by the @code{x} command, it
11403 will be displayed on the next line.
11405 The @samp{m} format doesn't affect any other specified formats that were
11406 passed to the @code{x} command.
11408 @cindex @code{$_}, @code{$__}, and value history
11409 The addresses and contents printed by the @code{x} command are not saved
11410 in the value history because there is often too much of them and they
11411 would get in the way. Instead, @value{GDBN} makes these values available for
11412 subsequent use in expressions as values of the convenience variables
11413 @code{$_} and @code{$__}. After an @code{x} command, the last address
11414 examined is available for use in expressions in the convenience variable
11415 @code{$_}. The contents of that address, as examined, are available in
11416 the convenience variable @code{$__}.
11418 If the @code{x} command has a repeat count, the address and contents saved
11419 are from the last memory unit printed; this is not the same as the last
11420 address printed if several units were printed on the last line of output.
11422 @anchor{addressable memory unit}
11423 @cindex addressable memory unit
11424 Most targets have an addressable memory unit size of 8 bits. This means
11425 that to each memory address are associated 8 bits of data. Some
11426 targets, however, have other addressable memory unit sizes.
11427 Within @value{GDBN} and this document, the term
11428 @dfn{addressable memory unit} (or @dfn{memory unit} for short) is used
11429 when explicitly referring to a chunk of data of that size. The word
11430 @dfn{byte} is used to refer to a chunk of data of 8 bits, regardless of
11431 the addressable memory unit size of the target. For most systems,
11432 addressable memory unit is a synonym of byte.
11434 @cindex remote memory comparison
11435 @cindex target memory comparison
11436 @cindex verify remote memory image
11437 @cindex verify target memory image
11438 When you are debugging a program running on a remote target machine
11439 (@pxref{Remote Debugging}), you may wish to verify the program's image
11440 in the remote machine's memory against the executable file you
11441 downloaded to the target. Or, on any target, you may want to check
11442 whether the program has corrupted its own read-only sections. The
11443 @code{compare-sections} command is provided for such situations.
11446 @kindex compare-sections
11447 @item compare-sections @r{[}@var{section-name}@r{|}@code{-r}@r{]}
11448 Compare the data of a loadable section @var{section-name} in the
11449 executable file of the program being debugged with the same section in
11450 the target machine's memory, and report any mismatches. With no
11451 arguments, compares all loadable sections. With an argument of
11452 @code{-r}, compares all loadable read-only sections.
11454 Note: for remote targets, this command can be accelerated if the
11455 target supports computing the CRC checksum of a block of memory
11456 (@pxref{qCRC packet}).
11459 @node Memory Tagging
11460 @section Memory Tagging
11462 Memory tagging is a memory protection technology that uses a pair of tags to
11463 validate memory accesses through pointers. The tags are integer values
11464 usually comprised of a few bits, depending on the architecture.
11466 There are two types of tags that are used in this setup: logical and
11467 allocation. A logical tag is stored in the pointers themselves, usually at the
11468 higher bits of the pointers. An allocation tag is the tag associated
11469 with particular ranges of memory in the physical address space, against which
11470 the logical tags from pointers are compared.
11472 The pointer tag (logical tag) must match the memory tag (allocation tag)
11473 for the memory access to be valid. If the logical tag does not match the
11474 allocation tag, that will raise a memory violation.
11476 Allocation tags cover multiple contiguous bytes of physical memory. This
11477 range of bytes is called a memory tag granule and is architecture-specific.
11478 For example, AArch64 has a tag granule of 16 bytes, meaning each allocation
11479 tag spans 16 bytes of memory.
11481 If the underlying architecture supports memory tagging, like AArch64 MTE
11482 or SPARC ADI do, @value{GDBN} can make use of it to validate pointers
11483 against memory allocation tags.
11485 The @code{print} (@pxref{Data}) and @code{x} (@pxref{Memory}) commands will
11486 display tag information when appropriate, and a command prefix of
11487 @code{memory-tag} gives access to the various memory tagging commands.
11489 The @code{memory-tag} commands are the following:
11492 @kindex memory-tag print-logical-tag
11493 @item memory-tag print-logical-tag @var{pointer_expression}
11494 Print the logical tag stored in @var{pointer_expression}.
11495 @kindex memory-tag with-logical-tag
11496 @item memory-tag with-logical-tag @var{pointer_expression} @var{tag_bytes}
11497 Print the pointer given by @var{pointer_expression}, augmented with a logical
11498 tag of @var{tag_bytes}.
11499 @kindex memory-tag print-allocation-tag
11500 @item memory-tag print-allocation-tag @var{address_expression}
11501 Print the allocation tag associated with the memory address given by
11502 @var{address_expression}.
11503 @kindex memory-tag setatag
11504 @item memory-tag setatag @var{starting_address} @var{length} @var{tag_bytes}
11505 Set the allocation tag(s) for memory range @r{[}@var{starting_address},
11506 @var{starting_address} + @var{length}@r{)} to @var{tag_bytes}.
11507 @kindex memory-tag check
11508 @item memory-tag check @var{pointer_expression}
11509 Check if the logical tag in the pointer given by @var{pointer_expression}
11510 matches the allocation tag for the memory referenced by the pointer.
11512 This essentially emulates the hardware validation that is done when tagged
11513 memory is accessed through a pointer, but does not cause a memory fault as
11514 it would during hardware validation.
11516 It can be used to inspect potential memory tagging violations in the running
11517 process, before any faults get triggered.
11521 @section Automatic Display
11522 @cindex automatic display
11523 @cindex display of expressions
11525 If you find that you want to print the value of an expression frequently
11526 (to see how it changes), you might want to add it to the @dfn{automatic
11527 display list} so that @value{GDBN} prints its value each time your program stops.
11528 Each expression added to the list is given a number to identify it;
11529 to remove an expression from the list, you specify that number.
11530 The automatic display looks like this:
11534 3: bar[5] = (struct hack *) 0x3804
11538 This display shows item numbers, expressions and their current values. As with
11539 displays you request manually using @code{x} or @code{print}, you can
11540 specify the output format you prefer; in fact, @code{display} decides
11541 whether to use @code{print} or @code{x} depending your format
11542 specification---it uses @code{x} if you specify either the @samp{i}
11543 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
11547 @item display @var{expr}
11548 Add the expression @var{expr} to the list of expressions to display
11549 each time your program stops. @xref{Expressions, ,Expressions}.
11551 @code{display} does not repeat if you press @key{RET} again after using it.
11553 @item display/@var{fmt} @var{expr}
11554 For @var{fmt} specifying only a display format and not a size or
11555 count, add the expression @var{expr} to the auto-display list but
11556 arrange to display it each time in the specified format @var{fmt}.
11557 @xref{Output Formats,,Output Formats}.
11559 @item display/@var{fmt} @var{addr}
11560 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
11561 number of units, add the expression @var{addr} as a memory address to
11562 be examined each time your program stops. Examining means in effect
11563 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
11566 For example, @samp{display/i $pc} can be helpful, to see the machine
11567 instruction about to be executed each time execution stops (@samp{$pc}
11568 is a common name for the program counter; @pxref{Registers, ,Registers}).
11571 @kindex delete display
11573 @item undisplay @var{dnums}@dots{}
11574 @itemx delete display @var{dnums}@dots{}
11575 Remove items from the list of expressions to display. Specify the
11576 numbers of the displays that you want affected with the command
11577 argument @var{dnums}. It can be a single display number, one of the
11578 numbers shown in the first field of the @samp{info display} display;
11579 or it could be a range of display numbers, as in @code{2-4}.
11581 @code{undisplay} does not repeat if you press @key{RET} after using it.
11582 (Otherwise you would just get the error @samp{No display number @dots{}}.)
11584 @kindex disable display
11585 @item disable display @var{dnums}@dots{}
11586 Disable the display of item numbers @var{dnums}. A disabled display
11587 item is not printed automatically, but is not forgotten. It may be
11588 enabled again later. Specify the numbers of the displays that you
11589 want affected with the command argument @var{dnums}. It can be a
11590 single display number, one of the numbers shown in the first field of
11591 the @samp{info display} display; or it could be a range of display
11592 numbers, as in @code{2-4}.
11594 @kindex enable display
11595 @item enable display @var{dnums}@dots{}
11596 Enable display of item numbers @var{dnums}. It becomes effective once
11597 again in auto display of its expression, until you specify otherwise.
11598 Specify the numbers of the displays that you want affected with the
11599 command argument @var{dnums}. It can be a single display number, one
11600 of the numbers shown in the first field of the @samp{info display}
11601 display; or it could be a range of display numbers, as in @code{2-4}.
11604 Display the current values of the expressions on the list, just as is
11605 done when your program stops.
11607 @kindex info display
11609 Print the list of expressions previously set up to display
11610 automatically, each one with its item number, but without showing the
11611 values. This includes disabled expressions, which are marked as such.
11612 It also includes expressions which would not be displayed right now
11613 because they refer to automatic variables not currently available.
11616 @cindex display disabled out of scope
11617 If a display expression refers to local variables, then it does not make
11618 sense outside the lexical context for which it was set up. Such an
11619 expression is disabled when execution enters a context where one of its
11620 variables is not defined. For example, if you give the command
11621 @code{display last_char} while inside a function with an argument
11622 @code{last_char}, @value{GDBN} displays this argument while your program
11623 continues to stop inside that function. When it stops elsewhere---where
11624 there is no variable @code{last_char}---the display is disabled
11625 automatically. The next time your program stops where @code{last_char}
11626 is meaningful, you can enable the display expression once again.
11628 @node Print Settings
11629 @section Print Settings
11631 @cindex format options
11632 @cindex print settings
11633 @value{GDBN} provides the following ways to control how arrays, structures,
11634 and symbols are printed.
11637 These settings are useful for debugging programs in any language:
11641 @anchor{set print address}
11642 @item set print address
11643 @itemx set print address on
11644 @cindex print/don't print memory addresses
11645 @value{GDBN} prints memory addresses showing the location of stack
11646 traces, structure values, pointer values, breakpoints, and so forth,
11647 even when it also displays the contents of those addresses. The default
11648 is @code{on}. For example, this is what a stack frame display looks like with
11649 @code{set print address on}:
11654 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
11656 530 if (lquote != def_lquote)
11660 @item set print address off
11661 Do not print addresses when displaying their contents. For example,
11662 this is the same stack frame displayed with @code{set print address off}:
11666 (@value{GDBP}) set print addr off
11668 #0 set_quotes (lq="<<", rq=">>") at input.c:530
11669 530 if (lquote != def_lquote)
11673 You can use @samp{set print address off} to eliminate all machine
11674 dependent displays from the @value{GDBN} interface. For example, with
11675 @code{print address off}, you should get the same text for backtraces on
11676 all machines---whether or not they involve pointer arguments.
11679 @item show print address
11680 Show whether or not addresses are to be printed.
11683 When @value{GDBN} prints a symbolic address, it normally prints the
11684 closest earlier symbol plus an offset. If that symbol does not uniquely
11685 identify the address (for example, it is a name whose scope is a single
11686 source file), you may need to clarify. One way to do this is with
11687 @code{info line}, for example @samp{info line *0x4537}. Alternately,
11688 you can set @value{GDBN} to print the source file and line number when
11689 it prints a symbolic address:
11692 @item set print symbol-filename on
11693 @cindex source file and line of a symbol
11694 @cindex symbol, source file and line
11695 Tell @value{GDBN} to print the source file name and line number of a
11696 symbol in the symbolic form of an address.
11698 @item set print symbol-filename off
11699 Do not print source file name and line number of a symbol. This is the
11702 @item show print symbol-filename
11703 Show whether or not @value{GDBN} will print the source file name and
11704 line number of a symbol in the symbolic form of an address.
11707 Another situation where it is helpful to show symbol filenames and line
11708 numbers is when disassembling code; @value{GDBN} shows you the line
11709 number and source file that corresponds to each instruction.
11711 Also, you may wish to see the symbolic form only if the address being
11712 printed is reasonably close to the closest earlier symbol:
11715 @item set print max-symbolic-offset @var{max-offset}
11716 @itemx set print max-symbolic-offset unlimited
11717 @cindex maximum value for offset of closest symbol
11718 Tell @value{GDBN} to only display the symbolic form of an address if the
11719 offset between the closest earlier symbol and the address is less than
11720 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
11721 to always print the symbolic form of an address if any symbol precedes
11722 it. Zero is equivalent to @code{unlimited}.
11724 @item show print max-symbolic-offset
11725 Ask how large the maximum offset is that @value{GDBN} prints in a
11729 @cindex wild pointer, interpreting
11730 @cindex pointer, finding referent
11731 If you have a pointer and you are not sure where it points, try
11732 @samp{set print symbol-filename on}. Then you can determine the name
11733 and source file location of the variable where it points, using
11734 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
11735 For example, here @value{GDBN} shows that a variable @code{ptt} points
11736 at another variable @code{t}, defined in @file{hi2.c}:
11739 (@value{GDBP}) set print symbol-filename on
11740 (@value{GDBP}) p/a ptt
11741 $4 = 0xe008 <t in hi2.c>
11745 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
11746 does not show the symbol name and filename of the referent, even with
11747 the appropriate @code{set print} options turned on.
11750 You can also enable @samp{/a}-like formatting all the time using
11751 @samp{set print symbol on}:
11753 @anchor{set print symbol}
11755 @item set print symbol on
11756 Tell @value{GDBN} to print the symbol corresponding to an address, if
11759 @item set print symbol off
11760 Tell @value{GDBN} not to print the symbol corresponding to an
11761 address. In this mode, @value{GDBN} will still print the symbol
11762 corresponding to pointers to functions. This is the default.
11764 @item show print symbol
11765 Show whether @value{GDBN} will display the symbol corresponding to an
11769 Other settings control how different kinds of objects are printed:
11772 @anchor{set print array}
11773 @item set print array
11774 @itemx set print array on
11775 @cindex pretty print arrays
11776 Pretty print arrays. This format is more convenient to read,
11777 but uses more space. The default is off.
11779 @item set print array off
11780 Return to compressed format for arrays.
11782 @item show print array
11783 Show whether compressed or pretty format is selected for displaying
11786 @cindex print array indexes
11787 @anchor{set print array-indexes}
11788 @item set print array-indexes
11789 @itemx set print array-indexes on
11790 Print the index of each element when displaying arrays. May be more
11791 convenient to locate a given element in the array or quickly find the
11792 index of a given element in that printed array. The default is off.
11794 @item set print array-indexes off
11795 Stop printing element indexes when displaying arrays.
11797 @item show print array-indexes
11798 Show whether the index of each element is printed when displaying
11801 @anchor{set print nibbles}
11802 @item set print nibbles
11803 @itemx set print nibbles on
11804 @cindex print binary values in groups of four bits
11805 Print binary values in groups of four bits, known as @dfn{nibbles},
11806 when using the print command of @value{GDBN} with the option @samp{/t}.
11807 For example, this is what it looks like with @code{set print nibbles on}:
11811 (@value{GDBP}) print val_flags
11813 (@value{GDBP}) print/t val_flags
11814 $2 = 0100 1100 1110
11818 @item set print nibbles off
11819 Don't printing binary values in groups. This is the default.
11821 @item show print nibbles
11822 Show whether to print binary values in groups of four bits.
11824 @anchor{set print characters}
11825 @item set print characters @var{number-of-characters}
11826 @itemx set print characters elements
11827 @itemx set print characters unlimited
11828 @cindex number of string characters to print
11829 @cindex limit on number of printed string characters
11830 Set a limit on how many characters of a string @value{GDBN} will print.
11831 If @value{GDBN} is printing a large string, it stops printing after it
11832 has printed the number of characters set by the @code{set print
11833 characters} command. This equally applies to multi-byte and wide
11834 character strings, that is for strings whose character type is
11835 @code{wchar_t}, @code{char16_t}, or @code{char32_t} it is the number of
11836 actual characters rather than underlying bytes the encoding uses that
11837 this setting controls.
11838 Setting @var{number-of-characters} to @code{elements} means that the
11839 limit on the number of characters to print follows one for array
11840 elements; see @ref{set print elements}.
11841 Setting @var{number-of-characters} to @code{unlimited} means that the
11842 number of characters to print is unlimited.
11843 When @value{GDBN} starts, this limit is set to @code{elements}.
11845 @item show print characters
11846 Display the number of characters of a large string that @value{GDBN}
11849 @anchor{set print elements}
11850 @item set print elements @var{number-of-elements}
11851 @itemx set print elements unlimited
11852 @cindex number of array elements to print
11853 @cindex limit on number of printed array elements
11854 Set a limit on how many elements of an array @value{GDBN} will print.
11855 If @value{GDBN} is printing a large array, it stops printing after it has
11856 printed the number of elements set by the @code{set print elements} command.
11857 By default this limit also applies to the display of strings; see
11858 @ref{set print characters}.
11859 When @value{GDBN} starts, this limit is set to 200.
11860 Setting @var{number-of-elements} to @code{unlimited} or zero means
11861 that the number of elements to print is unlimited.
11863 When printing very large arrays, whose size is greater than
11864 @code{max-value-size} (@pxref{set max-value-size,,max-value-size}),
11865 if the @code{print elements} is set such that the size of the elements
11866 being printed is less than or equal to @code{max-value-size}, then
11867 @value{GDBN} will print the array (up to the @code{print elements} limit),
11868 and only @code{max-value-size} worth of data will be added into the value
11869 history (@pxref{Value History, ,Value History}).
11871 @item show print elements
11872 Display the number of elements of a large array that @value{GDBN} will print.
11874 @anchor{set print frame-arguments}
11875 @item set print frame-arguments @var{value}
11876 @kindex set print frame-arguments
11877 @cindex printing frame argument values
11878 @cindex print all frame argument values
11879 @cindex print frame argument values for scalars only
11880 @cindex do not print frame arguments
11881 This command allows to control how the values of arguments are printed
11882 when the debugger prints a frame (@pxref{Frames}). The possible
11887 The values of all arguments are printed.
11890 Print the value of an argument only if it is a scalar. The value of more
11891 complex arguments such as arrays, structures, unions, etc, is replaced
11892 by @code{@dots{}}. This is the default. Here is an example where
11893 only scalar arguments are shown:
11896 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
11901 None of the argument values are printed. Instead, the value of each argument
11902 is replaced by @code{@dots{}}. In this case, the example above now becomes:
11905 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
11910 Only the presence of arguments is indicated by @code{@dots{}}.
11911 The @code{@dots{}} are not printed for function without any arguments.
11912 None of the argument names and values are printed.
11913 In this case, the example above now becomes:
11916 #1 0x08048361 in call_me (@dots{}) at frame-args.c:23
11921 By default, only scalar arguments are printed. This command can be used
11922 to configure the debugger to print the value of all arguments, regardless
11923 of their type. However, it is often advantageous to not print the value
11924 of more complex parameters. For instance, it reduces the amount of
11925 information printed in each frame, making the backtrace more readable.
11926 Also, it improves performance when displaying Ada frames, because
11927 the computation of large arguments can sometimes be CPU-intensive,
11928 especially in large applications. Setting @code{print frame-arguments}
11929 to @code{scalars} (the default), @code{none} or @code{presence} avoids
11930 this computation, thus speeding up the display of each Ada frame.
11932 @item show print frame-arguments
11933 Show how the value of arguments should be displayed when printing a frame.
11935 @anchor{set print raw-frame-arguments}
11936 @item set print raw-frame-arguments on
11937 Print frame arguments in raw, non pretty-printed, form.
11939 @item set print raw-frame-arguments off
11940 Print frame arguments in pretty-printed form, if there is a pretty-printer
11941 for the value (@pxref{Pretty Printing}),
11942 otherwise print the value in raw form.
11943 This is the default.
11945 @item show print raw-frame-arguments
11946 Show whether to print frame arguments in raw form.
11948 @anchor{set print entry-values}
11949 @item set print entry-values @var{value}
11950 @kindex set print entry-values
11951 Set printing of frame argument values at function entry. In some cases
11952 @value{GDBN} can determine the value of function argument which was passed by
11953 the function caller, even if the value was modified inside the called function
11954 and therefore is different. With optimized code, the current value could be
11955 unavailable, but the entry value may still be known.
11957 The default value is @code{default} (see below for its description). Older
11958 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
11959 this feature will behave in the @code{default} setting the same way as with the
11962 This functionality is currently supported only by DWARF 2 debugging format and
11963 the compiler has to produce @samp{DW_TAG_call_site} tags. With
11964 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
11967 The @var{value} parameter can be one of the following:
11971 Print only actual parameter values, never print values from function entry
11975 #0 different (val=6)
11976 #0 lost (val=<optimized out>)
11978 #0 invalid (val=<optimized out>)
11982 Print only parameter values from function entry point. The actual parameter
11983 values are never printed.
11985 #0 equal (val@@entry=5)
11986 #0 different (val@@entry=5)
11987 #0 lost (val@@entry=5)
11988 #0 born (val@@entry=<optimized out>)
11989 #0 invalid (val@@entry=<optimized out>)
11993 Print only parameter values from function entry point. If value from function
11994 entry point is not known while the actual value is known, print the actual
11995 value for such parameter.
11997 #0 equal (val@@entry=5)
11998 #0 different (val@@entry=5)
11999 #0 lost (val@@entry=5)
12001 #0 invalid (val@@entry=<optimized out>)
12005 Print actual parameter values. If actual parameter value is not known while
12006 value from function entry point is known, print the entry point value for such
12010 #0 different (val=6)
12011 #0 lost (val@@entry=5)
12013 #0 invalid (val=<optimized out>)
12017 Always print both the actual parameter value and its value from function entry
12018 point, even if values of one or both are not available due to compiler
12021 #0 equal (val=5, val@@entry=5)
12022 #0 different (val=6, val@@entry=5)
12023 #0 lost (val=<optimized out>, val@@entry=5)
12024 #0 born (val=10, val@@entry=<optimized out>)
12025 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
12029 Print the actual parameter value if it is known and also its value from
12030 function entry point if it is known. If neither is known, print for the actual
12031 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
12032 values are known and identical, print the shortened
12033 @code{param=param@@entry=VALUE} notation.
12035 #0 equal (val=val@@entry=5)
12036 #0 different (val=6, val@@entry=5)
12037 #0 lost (val@@entry=5)
12039 #0 invalid (val=<optimized out>)
12043 Always print the actual parameter value. Print also its value from function
12044 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
12045 if both values are known and identical, print the shortened
12046 @code{param=param@@entry=VALUE} notation.
12048 #0 equal (val=val@@entry=5)
12049 #0 different (val=6, val@@entry=5)
12050 #0 lost (val=<optimized out>, val@@entry=5)
12052 #0 invalid (val=<optimized out>)
12056 For analysis messages on possible failures of frame argument values at function
12057 entry resolution see @ref{set debug entry-values}.
12059 @item show print entry-values
12060 Show the method being used for printing of frame argument values at function
12063 @anchor{set print frame-info}
12064 @item set print frame-info @var{value}
12065 @kindex set print frame-info
12066 @cindex printing frame information
12067 @cindex frame information, printing
12068 This command allows to control the information printed when
12069 the debugger prints a frame. See @ref{Frames}, @ref{Backtrace},
12070 for a general explanation about frames and frame information.
12071 Note that some other settings (such as @code{set print frame-arguments}
12072 and @code{set print address}) are also influencing if and how some frame
12073 information is displayed. In particular, the frame program counter is never
12074 printed if @code{set print address} is off.
12076 The possible values for @code{set print frame-info} are:
12078 @item short-location
12079 Print the frame level, the program counter (if not at the
12080 beginning of the location source line), the function, the function
12083 Same as @code{short-location} but also print the source file and source line
12085 @item location-and-address
12086 Same as @code{location} but print the program counter even if located at the
12087 beginning of the location source line.
12089 Print the program counter (if not at the beginning of the location
12090 source line), the line number and the source line.
12091 @item source-and-location
12092 Print what @code{location} and @code{source-line} are printing.
12094 The information printed for a frame is decided automatically
12095 by the @value{GDBN} command that prints a frame.
12096 For example, @code{frame} prints the information printed by
12097 @code{source-and-location} while @code{stepi} will switch between
12098 @code{source-line} and @code{source-and-location} depending on the program
12100 The default value is @code{auto}.
12103 @anchor{set print repeats}
12104 @item set print repeats @var{number-of-repeats}
12105 @itemx set print repeats unlimited
12106 @cindex repeated array elements
12107 Set the threshold for suppressing display of repeated array
12108 elements. When the number of consecutive identical elements of an
12109 array exceeds the threshold, @value{GDBN} prints the string
12110 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
12111 identical repetitions, instead of displaying the identical elements
12112 themselves. Setting the threshold to @code{unlimited} or zero will
12113 cause all elements to be individually printed. The default threshold
12116 @item show print repeats
12117 Display the current threshold for printing repeated identical
12120 @anchor{set print max-depth}
12121 @item set print max-depth @var{depth}
12122 @item set print max-depth unlimited
12123 @cindex printing nested structures
12124 Set the threshold after which nested structures are replaced with
12125 ellipsis, this can make visualising deeply nested structures easier.
12127 For example, given this C code
12130 typedef struct s1 @{ int a; @} s1;
12131 typedef struct s2 @{ s1 b; @} s2;
12132 typedef struct s3 @{ s2 c; @} s3;
12133 typedef struct s4 @{ s3 d; @} s4;
12135 s4 var = @{ @{ @{ @{ 3 @} @} @} @};
12138 The following table shows how different values of @var{depth} will
12139 effect how @code{var} is printed by @value{GDBN}:
12141 @multitable @columnfractions .3 .7
12142 @headitem @var{depth} setting @tab Result of @samp{p var}
12144 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
12146 @tab @code{$1 = @{...@}}
12148 @tab @code{$1 = @{d = @{...@}@}}
12150 @tab @code{$1 = @{d = @{c = @{...@}@}@}}
12152 @tab @code{$1 = @{d = @{c = @{b = @{...@}@}@}@}}
12154 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
12157 To see the contents of structures that have been hidden the user can
12158 either increase the print max-depth, or they can print the elements of
12159 the structure that are visible, for example
12162 (@value{GDBP}) set print max-depth 2
12163 (@value{GDBP}) p var
12164 $1 = @{d = @{c = @{...@}@}@}
12165 (@value{GDBP}) p var.d
12166 $2 = @{c = @{b = @{...@}@}@}
12167 (@value{GDBP}) p var.d.c
12168 $3 = @{b = @{a = 3@}@}
12171 The pattern used to replace nested structures varies based on
12172 language, for most languages @code{@{...@}} is used, but Fortran uses
12175 @item show print max-depth
12176 Display the current threshold after which nested structures are
12177 replaces with ellipsis.
12179 @anchor{set print memory-tag-violations}
12180 @cindex printing memory tag violation information
12181 @item set print memory-tag-violations
12182 @itemx set print memory-tag-violations on
12183 Cause @value{GDBN} to display additional information about memory tag violations
12184 when printing pointers and addresses.
12186 @item set print memory-tag-violations off
12187 Stop printing memory tag violation information.
12189 @item show print memory-tag-violations
12190 Show whether memory tag violation information is displayed when printing
12191 pointers and addresses.
12193 @anchor{set print null-stop}
12194 @item set print null-stop
12195 @cindex @sc{null} elements in arrays
12196 Cause @value{GDBN} to stop printing the characters of an array when the first
12197 @sc{null} is encountered. This is useful when large arrays actually
12198 contain only short strings.
12199 The default is off.
12201 @item show print null-stop
12202 Show whether @value{GDBN} stops printing an array on the first
12203 @sc{null} character.
12205 @anchor{set print pretty}
12206 @item set print pretty on
12207 @cindex print structures in indented form
12208 @cindex indentation in structure display
12209 Cause @value{GDBN} to print structures in an indented format with one member
12210 per line, like this:
12225 @item set print pretty off
12226 Cause @value{GDBN} to print structures in a compact format, like this:
12230 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
12231 meat = 0x54 "Pork"@}
12236 This is the default format.
12238 @item show print pretty
12239 Show which format @value{GDBN} is using to print structures.
12241 @anchor{set print raw-values}
12242 @item set print raw-values on
12243 Print values in raw form, without applying the pretty
12244 printers for the value.
12246 @item set print raw-values off
12247 Print values in pretty-printed form, if there is a pretty-printer
12248 for the value (@pxref{Pretty Printing}),
12249 otherwise print the value in raw form.
12251 The default setting is ``off''.
12253 @item show print raw-values
12254 Show whether to print values in raw form.
12256 @item set print sevenbit-strings on
12257 @cindex eight-bit characters in strings
12258 @cindex octal escapes in strings
12259 Print using only seven-bit characters; if this option is set,
12260 @value{GDBN} displays any eight-bit characters (in strings or
12261 character values) using the notation @code{\}@var{nnn}. This setting is
12262 best if you are working in English (@sc{ascii}) and you use the
12263 high-order bit of characters as a marker or ``meta'' bit.
12265 @item set print sevenbit-strings off
12266 Print full eight-bit characters. This allows the use of more
12267 international character sets, and is the default.
12269 @item show print sevenbit-strings
12270 Show whether or not @value{GDBN} is printing only seven-bit characters.
12272 @anchor{set print union}
12273 @item set print union on
12274 @cindex unions in structures, printing
12275 Tell @value{GDBN} to print unions which are contained in structures
12276 and other unions. This is the default setting.
12278 @item set print union off
12279 Tell @value{GDBN} not to print unions which are contained in
12280 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
12283 @item show print union
12284 Ask @value{GDBN} whether or not it will print unions which are contained in
12285 structures and other unions.
12287 For example, given the declarations
12290 typedef enum @{Tree, Bug@} Species;
12291 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
12292 typedef enum @{Caterpillar, Cocoon, Butterfly@}
12303 struct thing foo = @{Tree, @{Acorn@}@};
12307 with @code{set print union on} in effect @samp{p foo} would print
12310 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
12314 and with @code{set print union off} in effect it would print
12317 $1 = @{it = Tree, form = @{...@}@}
12321 @code{set print union} affects programs written in C-like languages
12327 These settings are of interest when debugging C@t{++} programs:
12330 @cindex demangling C@t{++} names
12331 @item set print demangle
12332 @itemx set print demangle on
12333 Print C@t{++} names in their source form rather than in the encoded
12334 (``mangled'') form passed to the assembler and linker for type-safe
12335 linkage. The default is on.
12337 @item show print demangle
12338 Show whether C@t{++} names are printed in mangled or demangled form.
12340 @item set print asm-demangle
12341 @itemx set print asm-demangle on
12342 Print C@t{++} names in their source form rather than their mangled form, even
12343 in assembler code printouts such as instruction disassemblies.
12344 The default is off.
12346 @item show print asm-demangle
12347 Show whether C@t{++} names in assembly listings are printed in mangled
12350 @cindex C@t{++} symbol decoding style
12351 @cindex symbol decoding style, C@t{++}
12352 @kindex set demangle-style
12353 @item set demangle-style @var{style}
12354 Choose among several encoding schemes used by different compilers to represent
12355 C@t{++} names. If you omit @var{style}, you will see a list of possible
12356 formats. The default value is @var{auto}, which lets @value{GDBN} choose a
12357 decoding style by inspecting your program.
12359 @item show demangle-style
12360 Display the encoding style currently in use for decoding C@t{++} symbols.
12362 @anchor{set print object}
12363 @item set print object
12364 @itemx set print object on
12365 @cindex derived type of an object, printing
12366 @cindex display derived types
12367 When displaying a pointer to an object, identify the @emph{actual}
12368 (derived) type of the object rather than the @emph{declared} type, using
12369 the virtual function table. Note that the virtual function table is
12370 required---this feature can only work for objects that have run-time
12371 type identification; a single virtual method in the object's declared
12372 type is sufficient. Note that this setting is also taken into account when
12373 working with variable objects via MI (@pxref{GDB/MI}).
12375 @item set print object off
12376 Display only the declared type of objects, without reference to the
12377 virtual function table. This is the default setting.
12379 @item show print object
12380 Show whether actual, or declared, object types are displayed.
12382 @anchor{set print static-members}
12383 @item set print static-members
12384 @itemx set print static-members on
12385 @cindex static members of C@t{++} objects
12386 Print static members when displaying a C@t{++} object. The default is on.
12388 @item set print static-members off
12389 Do not print static members when displaying a C@t{++} object.
12391 @item show print static-members
12392 Show whether C@t{++} static members are printed or not.
12394 @item set print pascal_static-members
12395 @itemx set print pascal_static-members on
12396 @cindex static members of Pascal objects
12397 @cindex Pascal objects, static members display
12398 Print static members when displaying a Pascal object. The default is on.
12400 @item set print pascal_static-members off
12401 Do not print static members when displaying a Pascal object.
12403 @item show print pascal_static-members
12404 Show whether Pascal static members are printed or not.
12406 @c These don't work with HP ANSI C++ yet.
12407 @anchor{set print vtbl}
12408 @item set print vtbl
12409 @itemx set print vtbl on
12410 @cindex pretty print C@t{++} virtual function tables
12411 @cindex virtual functions (C@t{++}) display
12412 @cindex VTBL display
12413 Pretty print C@t{++} virtual function tables. The default is off.
12414 (The @code{vtbl} commands do not work on programs compiled with the HP
12415 ANSI C@t{++} compiler (@code{aCC}).)
12417 @item set print vtbl off
12418 Do not pretty print C@t{++} virtual function tables.
12420 @item show print vtbl
12421 Show whether C@t{++} virtual function tables are pretty printed, or not.
12424 @node Pretty Printing
12425 @section Pretty Printing
12427 @value{GDBN} provides a mechanism to allow pretty-printing of values using
12428 Python code. It greatly simplifies the display of complex objects. This
12429 mechanism works for both MI and the CLI.
12432 * Pretty-Printer Introduction:: Introduction to pretty-printers
12433 * Pretty-Printer Example:: An example pretty-printer
12434 * Pretty-Printer Commands:: Pretty-printer commands
12437 @node Pretty-Printer Introduction
12438 @subsection Pretty-Printer Introduction
12440 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
12441 registered for the value. If there is then @value{GDBN} invokes the
12442 pretty-printer to print the value. Otherwise the value is printed normally.
12444 Pretty-printers are normally named. This makes them easy to manage.
12445 The @samp{info pretty-printer} command will list all the installed
12446 pretty-printers with their names.
12447 If a pretty-printer can handle multiple data types, then its
12448 @dfn{subprinters} are the printers for the individual data types.
12449 Each such subprinter has its own name.
12450 The format of the name is @var{printer-name};@var{subprinter-name}.
12452 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
12453 Typically they are automatically loaded and registered when the corresponding
12454 debug information is loaded, thus making them available without having to
12455 do anything special.
12457 There are three places where a pretty-printer can be registered.
12461 Pretty-printers registered globally are available when debugging
12465 Pretty-printers registered with a program space are available only
12466 when debugging that program.
12467 @xref{Progspaces In Python}, for more details on program spaces in Python.
12470 Pretty-printers registered with an objfile are loaded and unloaded
12471 with the corresponding objfile (e.g., shared library).
12472 @xref{Objfiles In Python}, for more details on objfiles in Python.
12475 @xref{Selecting Pretty-Printers}, for further information on how
12476 pretty-printers are selected,
12478 @xref{Writing a Pretty-Printer}, for implementing pretty printers
12481 @node Pretty-Printer Example
12482 @subsection Pretty-Printer Example
12484 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
12487 (@value{GDBP}) print s
12489 static npos = 4294967295,
12491 <std::allocator<char>> = @{
12492 <__gnu_cxx::new_allocator<char>> = @{
12493 <No data fields>@}, <No data fields>
12495 members of std::basic_string<char, std::char_traits<char>,
12496 std::allocator<char> >::_Alloc_hider:
12497 _M_p = 0x804a014 "abcd"
12502 With a pretty-printer for @code{std::string} only the contents are printed:
12505 (@value{GDBP}) print s
12509 @node Pretty-Printer Commands
12510 @subsection Pretty-Printer Commands
12511 @cindex pretty-printer commands
12514 @kindex info pretty-printer
12515 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
12516 Print the list of installed pretty-printers.
12517 This includes disabled pretty-printers, which are marked as such.
12519 @var{object-regexp} is a regular expression matching the objects
12520 whose pretty-printers to list.
12521 Objects can be @code{global}, the program space's file
12522 (@pxref{Progspaces In Python}),
12523 and the object files within that program space (@pxref{Objfiles In Python}).
12524 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
12525 looks up a printer from these three objects.
12527 @var{name-regexp} is a regular expression matching the name of the printers
12530 @kindex disable pretty-printer
12531 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
12532 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
12533 A disabled pretty-printer is not forgotten, it may be enabled again later.
12535 @kindex enable pretty-printer
12536 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
12537 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
12542 Suppose we have three pretty-printers installed: one from library1.so
12543 named @code{foo} that prints objects of type @code{foo}, and
12544 another from library2.so named @code{bar} that prints two types of objects,
12545 @code{bar1} and @code{bar2}.
12549 (@value{GDBP}) info pretty-printer
12558 (@value{GDBP}) info pretty-printer library2
12565 (@value{GDBP}) disable pretty-printer library1
12567 2 of 3 printers enabled
12568 (@value{GDBP}) info pretty-printer
12577 (@value{GDBP}) disable pretty-printer library2 bar;bar1
12579 1 of 3 printers enabled
12580 (@value{GDBP}) info pretty-printer library2
12587 (@value{GDBP}) disable pretty-printer library2 bar
12589 0 of 3 printers enabled
12590 (@value{GDBP}) info pretty-printer
12600 Note that for @code{bar} the entire printer can be disabled,
12601 as can each individual subprinter.
12603 Printing values and frame arguments is done by default using
12604 the enabled pretty printers.
12606 The print option @code{-raw-values} and @value{GDBN} setting
12607 @code{set print raw-values} (@pxref{set print raw-values}) can be
12608 used to print values without applying the enabled pretty printers.
12610 Similarly, the backtrace option @code{-raw-frame-arguments} and
12611 @value{GDBN} setting @code{set print raw-frame-arguments}
12612 (@pxref{set print raw-frame-arguments}) can be used to ignore the
12613 enabled pretty printers when printing frame argument values.
12615 @node Value History
12616 @section Value History
12618 @cindex value history
12619 @cindex history of values printed by @value{GDBN}
12620 Values printed by the @code{print} command are saved in the @value{GDBN}
12621 @dfn{value history}. This allows you to refer to them in other expressions.
12622 Values are kept until the symbol table is re-read or discarded
12623 (for example with the @code{file} or @code{symbol-file} commands).
12624 When the symbol table changes, the value history is discarded,
12625 since the values may contain pointers back to the types defined in the
12630 @cindex history number
12631 The values printed are given @dfn{history numbers} by which you can
12632 refer to them. These are successive integers starting with one.
12633 @code{print} shows you the history number assigned to a value by
12634 printing @samp{$@var{num} = } before the value; here @var{num} is the
12637 To refer to any previous value, use @samp{$} followed by the value's
12638 history number. The way @code{print} labels its output is designed to
12639 remind you of this. Just @code{$} refers to the most recent value in
12640 the history, and @code{$$} refers to the value before that.
12641 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
12642 is the value just prior to @code{$$}, @code{$$1} is equivalent to
12643 @code{$$}, and @code{$$0} is equivalent to @code{$}.
12645 For example, suppose you have just printed a pointer to a structure and
12646 want to see the contents of the structure. It suffices to type
12652 If you have a chain of structures where the component @code{next} points
12653 to the next one, you can print the contents of the next one with this:
12660 You can print successive links in the chain by repeating this
12661 command---which you can do by just typing @key{RET}.
12663 Note that the history records values, not expressions. If the value of
12664 @code{x} is 4 and you type these commands:
12672 then the value recorded in the value history by the @code{print} command
12673 remains 4 even though the value of @code{x} has changed.
12676 @kindex show values
12678 Print the last ten values in the value history, with their item numbers.
12679 This is like @samp{p@ $$9} repeated ten times, except that @code{show
12680 values} does not change the history.
12682 @item show values @var{n}
12683 Print ten history values centered on history item number @var{n}.
12685 @item show values +
12686 Print ten history values just after the values last printed. If no more
12687 values are available, @code{show values +} produces no display.
12690 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
12691 same effect as @samp{show values +}.
12693 @node Convenience Vars
12694 @section Convenience Variables
12696 @cindex convenience variables
12697 @cindex user-defined variables
12698 @value{GDBN} provides @dfn{convenience variables} that you can use within
12699 @value{GDBN} to hold on to a value and refer to it later. These variables
12700 exist entirely within @value{GDBN}; they are not part of your program, and
12701 setting a convenience variable has no direct effect on further execution
12702 of your program. That is why you can use them freely.
12704 Convenience variables are prefixed with @samp{$}. Any name preceded by
12705 @samp{$} can be used for a convenience variable, unless it is one of
12706 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
12707 (Value history references, in contrast, are @emph{numbers} preceded
12708 by @samp{$}. @xref{Value History, ,Value History}.)
12710 You can save a value in a convenience variable with an assignment
12711 expression, just as you would set a variable in your program.
12715 set $foo = *object_ptr
12719 would save in @code{$foo} the value contained in the object pointed to by
12722 Using a convenience variable for the first time creates it, but its
12723 value is @code{void} until you assign a new value. You can alter the
12724 value with another assignment at any time.
12726 Convenience variables have no fixed types. You can assign a convenience
12727 variable any type of value, including structures and arrays, even if
12728 that variable already has a value of a different type. The convenience
12729 variable, when used as an expression, has the type of its current value.
12732 @kindex show convenience
12733 @cindex show all user variables and functions
12734 @item show convenience
12735 Print a list of convenience variables used so far, and their values,
12736 as well as a list of the convenience functions.
12737 Abbreviated @code{show conv}.
12739 @kindex init-if-undefined
12740 @cindex convenience variables, initializing
12741 @item init-if-undefined $@var{variable} = @var{expression}
12742 Set a convenience variable if it has not already been set. This is useful
12743 for user-defined commands that keep some state. It is similar, in concept,
12744 to using local static variables with initializers in C (except that
12745 convenience variables are global). It can also be used to allow users to
12746 override default values used in a command script.
12748 If the variable is already defined then the expression is not evaluated so
12749 any side-effects do not occur.
12752 One of the ways to use a convenience variable is as a counter to be
12753 incremented or a pointer to be advanced. For example, to print
12754 a field from successive elements of an array of structures:
12758 print bar[$i++]->contents
12762 Repeat that command by typing @key{RET}.
12764 Some convenience variables are created automatically by @value{GDBN} and given
12765 values likely to be useful.
12768 @vindex $_@r{, convenience variable}
12770 The variable @code{$_} is automatically set by the @code{x} command to
12771 the last address examined (@pxref{Memory, ,Examining Memory}). Other
12772 commands which provide a default address for @code{x} to examine also
12773 set @code{$_} to that address; these commands include @code{info line}
12774 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
12775 except when set by the @code{x} command, in which case it is a pointer
12776 to the type of @code{$__}.
12778 @vindex $__@r{, convenience variable}
12780 The variable @code{$__} is automatically set by the @code{x} command
12781 to the value found in the last address examined. Its type is chosen
12782 to match the format in which the data was printed.
12785 @vindex $_exitcode@r{, convenience variable}
12786 When the program being debugged terminates normally, @value{GDBN}
12787 automatically sets this variable to the exit code of the program, and
12788 resets @code{$_exitsignal} to @code{void}.
12791 @vindex $_exitsignal@r{, convenience variable}
12792 When the program being debugged dies due to an uncaught signal,
12793 @value{GDBN} automatically sets this variable to that signal's number,
12794 and resets @code{$_exitcode} to @code{void}.
12796 To distinguish between whether the program being debugged has exited
12797 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
12798 @code{$_exitsignal} is not @code{void}), the convenience function
12799 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
12800 Functions}). For example, considering the following source code:
12803 #include <signal.h>
12806 main (int argc, char *argv[])
12813 A valid way of telling whether the program being debugged has exited
12814 or signalled would be:
12817 (@value{GDBP}) define has_exited_or_signalled
12818 Type commands for definition of ``has_exited_or_signalled''.
12819 End with a line saying just ``end''.
12820 >if $_isvoid ($_exitsignal)
12821 >echo The program has exited\n
12823 >echo The program has signalled\n
12829 Program terminated with signal SIGALRM, Alarm clock.
12830 The program no longer exists.
12831 (@value{GDBP}) has_exited_or_signalled
12832 The program has signalled
12835 As can be seen, @value{GDBN} correctly informs that the program being
12836 debugged has signalled, since it calls @code{raise} and raises a
12837 @code{SIGALRM} signal. If the program being debugged had not called
12838 @code{raise}, then @value{GDBN} would report a normal exit:
12841 (@value{GDBP}) has_exited_or_signalled
12842 The program has exited
12846 The variable @code{$_exception} is set to the exception object being
12847 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
12849 @item $_ada_exception
12850 The variable @code{$_ada_exception} is set to the address of the
12851 exception being caught or thrown at an Ada exception-related
12852 catchpoint. @xref{Set Catchpoints}.
12855 @itemx $_probe_arg0@dots{}$_probe_arg11
12856 Arguments to a static probe. @xref{Static Probe Points}.
12859 @vindex $_sdata@r{, inspect, convenience variable}
12860 The variable @code{$_sdata} contains extra collected static tracepoint
12861 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
12862 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
12863 if extra static tracepoint data has not been collected.
12866 @vindex $_siginfo@r{, convenience variable}
12867 The variable @code{$_siginfo} contains extra signal information
12868 (@pxref{extra signal information}). Note that @code{$_siginfo}
12869 could be empty, if the application has not yet received any signals.
12870 For example, it will be empty before you execute the @code{run} command.
12873 @vindex $_tlb@r{, convenience variable}
12874 The variable @code{$_tlb} is automatically set when debugging
12875 applications running on MS-Windows in native mode or connected to
12876 gdbserver that supports the @code{qGetTIBAddr} request.
12877 @xref{General Query Packets}.
12878 This variable contains the address of the thread information block.
12881 The number of the current inferior. @xref{Inferiors Connections and
12882 Programs, ,Debugging Multiple Inferiors Connections and Programs}.
12885 The thread number of the current thread. @xref{thread numbers}.
12888 The global number of the current thread. @xref{global thread numbers}.
12890 @item $_inferior_thread_count
12891 The number of live threads in the current inferior. @xref{Threads}.
12895 @vindex $_gdb_major@r{, convenience variable}
12896 @vindex $_gdb_minor@r{, convenience variable}
12897 The major and minor version numbers of the running @value{GDBN}.
12898 Development snapshots and pretest versions have their minor version
12899 incremented by one; thus, @value{GDBN} pretest 9.11.90 will produce
12900 the value 12 for @code{$_gdb_minor}. These variables allow you to
12901 write scripts that work with different versions of @value{GDBN}
12902 without errors caused by features unavailable in some of those
12905 @item $_shell_exitcode
12906 @itemx $_shell_exitsignal
12907 @vindex $_shell_exitcode@r{, convenience variable}
12908 @vindex $_shell_exitsignal@r{, convenience variable}
12909 @cindex shell command, exit code
12910 @cindex shell command, exit signal
12911 @cindex exit status of shell commands
12912 @value{GDBN} commands such as @code{shell} and @code{|} are launching
12913 shell commands. When a launched command terminates, @value{GDBN}
12914 automatically maintains the variables @code{$_shell_exitcode}
12915 and @code{$_shell_exitsignal} according to the exit status of the last
12916 launched command. These variables are set and used similarly to
12917 the variables @code{$_exitcode} and @code{$_exitsignal}.
12921 @node Convenience Funs
12922 @section Convenience Functions
12924 @cindex convenience functions
12925 @value{GDBN} also supplies some @dfn{convenience functions}. These
12926 have a syntax similar to convenience variables. A convenience
12927 function can be used in an expression just like an ordinary function;
12928 however, a convenience function is implemented internally to
12931 These functions do not require @value{GDBN} to be configured with
12932 @code{Python} support, which means that they are always available.
12936 @findex $_isvoid@r{, convenience function}
12937 @item $_isvoid (@var{expr})
12938 Return one if the expression @var{expr} is @code{void}. Otherwise it
12941 A @code{void} expression is an expression where the type of the result
12942 is @code{void}. For example, you can examine a convenience variable
12943 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
12947 (@value{GDBP}) print $_exitcode
12949 (@value{GDBP}) print $_isvoid ($_exitcode)
12952 Starting program: ./a.out
12953 [Inferior 1 (process 29572) exited normally]
12954 (@value{GDBP}) print $_exitcode
12956 (@value{GDBP}) print $_isvoid ($_exitcode)
12960 In the example above, we used @code{$_isvoid} to check whether
12961 @code{$_exitcode} is @code{void} before and after the execution of the
12962 program being debugged. Before the execution there is no exit code to
12963 be examined, therefore @code{$_exitcode} is @code{void}. After the
12964 execution the program being debugged returned zero, therefore
12965 @code{$_exitcode} is zero, which means that it is not @code{void}
12968 The @code{void} expression can also be a call of a function from the
12969 program being debugged. For example, given the following function:
12978 The result of calling it inside @value{GDBN} is @code{void}:
12981 (@value{GDBP}) print foo ()
12983 (@value{GDBP}) print $_isvoid (foo ())
12985 (@value{GDBP}) set $v = foo ()
12986 (@value{GDBP}) print $v
12988 (@value{GDBP}) print $_isvoid ($v)
12992 @findex $_gdb_setting_str@r{, convenience function}
12993 @item $_gdb_setting_str (@var{setting})
12994 Return the value of the @value{GDBN} @var{setting} as a string.
12995 @var{setting} is any setting that can be used in a @code{set} or
12996 @code{show} command (@pxref{Controlling GDB}).
12999 (@value{GDBP}) show print frame-arguments
13000 Printing of non-scalar frame arguments is "scalars".
13001 (@value{GDBP}) p $_gdb_setting_str("print frame-arguments")
13003 (@value{GDBP}) p $_gdb_setting_str("height")
13008 @findex $_gdb_setting@r{, convenience function}
13009 @item $_gdb_setting (@var{setting})
13010 Return the value of the @value{GDBN} @var{setting}.
13011 The type of the returned value depends on the setting.
13013 The value type for boolean and auto boolean settings is @code{int}.
13014 The boolean values @code{off} and @code{on} are converted to
13015 the integer values @code{0} and @code{1}. The value @code{auto} is
13016 converted to the value @code{-1}.
13018 The value type for integer settings is either @code{unsigned int}
13019 or @code{int}, depending on the setting.
13021 Some integer settings accept an @code{unlimited} value.
13022 Depending on the setting, the @code{set} command also accepts
13023 the value @code{0} or the value @code{@minus{}1} as a synonym for
13025 For example, @code{set height unlimited} is equivalent to
13026 @code{set height 0}.
13028 Some other settings that accept the @code{unlimited} value
13029 use the value @code{0} to literally mean zero.
13030 For example, @code{set history size 0} indicates to not
13031 record any @value{GDBN} commands in the command history.
13032 For such settings, @code{@minus{}1} is the synonym
13033 for @code{unlimited}.
13035 See the documentation of the corresponding @code{set} command for
13036 the numerical value equivalent to @code{unlimited}.
13038 The @code{$_gdb_setting} function converts the unlimited value
13039 to a @code{0} or a @code{@minus{}1} value according to what the
13040 @code{set} command uses.
13044 (@value{GDBP}) p $_gdb_setting_str("height")
13046 (@value{GDBP}) p $_gdb_setting("height")
13048 (@value{GDBP}) set height unlimited
13049 (@value{GDBP}) p $_gdb_setting_str("height")
13051 (@value{GDBP}) p $_gdb_setting("height")
13055 (@value{GDBP}) p $_gdb_setting_str("history size")
13057 (@value{GDBP}) p $_gdb_setting("history size")
13059 (@value{GDBP}) p $_gdb_setting_str("disassemble-next-line")
13061 (@value{GDBP}) p $_gdb_setting("disassemble-next-line")
13067 Other setting types (enum, filename, optional filename, string, string noescape)
13068 are returned as string values.
13071 @findex $_gdb_maint_setting_str@r{, convenience function}
13072 @item $_gdb_maint_setting_str (@var{setting})
13073 Like the @code{$_gdb_setting_str} function, but works with
13074 @code{maintenance set} variables.
13076 @findex $_gdb_maint_setting@r{, convenience function}
13077 @item $_gdb_maint_setting (@var{setting})
13078 Like the @code{$_gdb_setting} function, but works with
13079 @code{maintenance set} variables.
13081 @anchor{$_shell convenience function}
13082 @findex $_shell@r{, convenience function}
13083 @item $_shell (@var{command-string})
13085 Invoke a shell to execute @var{command-string}. @var{command-string}
13086 must be a string. The shell runs on the host machine, the machine
13087 @value{GDBN} is running on. Returns the command's exit status. On
13088 Unix systems, a command which exits with a zero exit status has
13089 succeeded, and non-zero exit status indicates failure. When a command
13090 terminates on a fatal signal whose number is @var{N}, @value{GDBN}
13091 uses the value 128+@var{N} as the exit status, as is standard in Unix
13092 shells. Note that @var{N} is a host signal number, not a target
13093 signal number. If you're native debugging, they will be the same, but
13094 if cross debugging, the host vs target signal numbers may be
13095 completely unrelated. Please consult your host operating system's
13096 documentation for the mapping between host signal numbers and signal
13097 names. The shell to run is determined in the same way as for the
13098 @code{shell} command. @xref{Shell Commands, ,Shell Commands}.
13101 (@value{GDBP}) print $_shell("true")
13103 (@value{GDBP}) print $_shell("false")
13105 (@value{GDBP}) p $_shell("echo hello")
13108 (@value{GDBP}) p $_shell("foobar")
13109 bash: line 1: foobar: command not found
13113 This may also be useful in breakpoint conditions. For example:
13116 (@value{GDBP}) break function if $_shell("some command") == 0
13119 In this scenario, you'll want to make sure that the shell command you
13120 run in the breakpoint condition takes the least amount of time
13121 possible. For example, avoid running a command that may block
13122 indefinitely, or that sleeps for a while before exiting. Prefer a
13123 command or script which analyzes some state and exits immediately.
13124 This is important because the debugged program stops for the
13125 breakpoint every time, and then @value{GDBN} evaluates the breakpoint
13126 condition. If the condition is false, the program is re-resumed
13127 transparently, without informing you of the stop. A quick shell
13128 command thus avoids significantly slowing down the debugged program
13131 Note: unlike the @code{shell} command, the @code{$_shell} convenience
13132 function does not affect the @code{$_shell_exitcode} and
13133 @code{$_shell_exitsignal} convenience variables.
13137 The following functions require @value{GDBN} to be configured with
13138 @code{Python} support.
13142 @findex $_memeq@r{, convenience function}
13143 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
13144 Returns one if the @var{length} bytes at the addresses given by
13145 @var{buf1} and @var{buf2} are equal.
13146 Otherwise it returns zero.
13148 @findex $_regex@r{, convenience function}
13149 @item $_regex(@var{str}, @var{regex})
13150 Returns one if the string @var{str} matches the regular expression
13151 @var{regex}. Otherwise it returns zero.
13152 The syntax of the regular expression is that specified by @code{Python}'s
13153 regular expression support.
13155 @findex $_streq@r{, convenience function}
13156 @item $_streq(@var{str1}, @var{str2})
13157 Returns one if the strings @var{str1} and @var{str2} are equal.
13158 Otherwise it returns zero.
13160 @findex $_strlen@r{, convenience function}
13161 @item $_strlen(@var{str})
13162 Returns the length of string @var{str}.
13164 @findex $_caller_is@r{, convenience function}
13165 @item $_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
13166 Returns one if the calling function's name is equal to @var{name}.
13167 Otherwise it returns zero.
13169 If the optional argument @var{number_of_frames} is provided,
13170 it is the number of frames up in the stack to look.
13176 (@value{GDBP}) backtrace
13178 at testsuite/gdb.python/py-caller-is.c:21
13179 #1 0x00000000004005a0 in middle_func ()
13180 at testsuite/gdb.python/py-caller-is.c:27
13181 #2 0x00000000004005ab in top_func ()
13182 at testsuite/gdb.python/py-caller-is.c:33
13183 #3 0x00000000004005b6 in main ()
13184 at testsuite/gdb.python/py-caller-is.c:39
13185 (@value{GDBP}) print $_caller_is ("middle_func")
13187 (@value{GDBP}) print $_caller_is ("top_func", 2)
13191 @findex $_caller_matches@r{, convenience function}
13192 @item $_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
13193 Returns one if the calling function's name matches the regular expression
13194 @var{regexp}. Otherwise it returns zero.
13196 If the optional argument @var{number_of_frames} is provided,
13197 it is the number of frames up in the stack to look.
13200 @findex $_any_caller_is@r{, convenience function}
13201 @item $_any_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
13202 Returns one if any calling function's name is equal to @var{name}.
13203 Otherwise it returns zero.
13205 If the optional argument @var{number_of_frames} is provided,
13206 it is the number of frames up in the stack to look.
13209 This function differs from @code{$_caller_is} in that this function
13210 checks all stack frames from the immediate caller to the frame specified
13211 by @var{number_of_frames}, whereas @code{$_caller_is} only checks the
13212 frame specified by @var{number_of_frames}.
13214 @findex $_any_caller_matches@r{, convenience function}
13215 @item $_any_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
13216 Returns one if any calling function's name matches the regular expression
13217 @var{regexp}. Otherwise it returns zero.
13219 If the optional argument @var{number_of_frames} is provided,
13220 it is the number of frames up in the stack to look.
13223 This function differs from @code{$_caller_matches} in that this function
13224 checks all stack frames from the immediate caller to the frame specified
13225 by @var{number_of_frames}, whereas @code{$_caller_matches} only checks the
13226 frame specified by @var{number_of_frames}.
13228 @findex $_as_string@r{, convenience function}
13229 @item $_as_string(@var{value})
13230 This convenience function is considered deprecated, and could be
13231 removed from future versions of @value{GDBN}. Use the @samp{%V} format
13232 specifier instead (@pxref{%V Format Specifier}).
13234 Return the string representation of @var{value}.
13236 This function is useful to obtain the textual label (enumerator) of an
13237 enumeration value. For example, assuming the variable @var{node} is of
13238 an enumerated type:
13241 (@value{GDBP}) printf "Visiting node of type %s\n", $_as_string(node)
13242 Visiting node of type NODE_INTEGER
13245 @findex $_cimag@r{, convenience function}
13246 @findex $_creal@r{, convenience function}
13247 @item $_cimag(@var{value})
13248 @itemx $_creal(@var{value})
13249 Return the imaginary (@code{$_cimag}) or real (@code{$_creal}) part of
13250 the complex number @var{value}.
13252 The type of the imaginary or real part depends on the type of the
13253 complex number, e.g., using @code{$_cimag} on a @code{float complex}
13254 will return an imaginary part of type @code{float}.
13258 @value{GDBN} provides the ability to list and get help on
13259 convenience functions.
13262 @item help function
13263 @kindex help function
13264 @cindex show all convenience functions
13265 Print a list of all convenience functions.
13272 You can refer to machine register contents, in expressions, as variables
13273 with names starting with @samp{$}. The names of registers are different
13274 for each machine; use @code{info registers} to see the names used on
13278 @kindex info registers
13279 @item info registers
13280 Print the names and values of all registers except floating-point
13281 and vector registers (in the selected stack frame).
13283 @kindex info all-registers
13284 @cindex floating point registers
13285 @item info all-registers
13286 Print the names and values of all registers, including floating-point
13287 and vector registers (in the selected stack frame).
13289 @anchor{info_registers_reggroup}
13290 @item info registers @var{reggroup} @dots{}
13291 Print the name and value of the registers in each of the specified
13292 @var{reggroup}s. The @var{reggroup} can be any of those returned by
13293 @code{maint print reggroups} (@pxref{Maintenance Commands}).
13295 @item info registers @var{regname} @dots{}
13296 Print the @dfn{relativized} value of each specified register @var{regname}.
13297 As discussed in detail below, register values are normally relative to
13298 the selected stack frame. The @var{regname} may be any register name valid on
13299 the machine you are using, with or without the initial @samp{$}.
13302 @anchor{standard registers}
13303 @cindex stack pointer register
13304 @cindex program counter register
13305 @cindex process status register
13306 @cindex frame pointer register
13307 @cindex standard registers
13308 @value{GDBN} has four ``standard'' register names that are available (in
13309 expressions) on most machines---whenever they do not conflict with an
13310 architecture's canonical mnemonics for registers. The register names
13311 @code{$pc} and @code{$sp} are used for the program counter register and
13312 the stack pointer. @code{$fp} is used for a register that contains a
13313 pointer to the current stack frame, and @code{$ps} is used for a
13314 register that contains the processor status. For example,
13315 you could print the program counter in hex with
13322 or print the instruction to be executed next with
13329 or add four to the stack pointer@footnote{This is a way of removing
13330 one word from the stack, on machines where stacks grow downward in
13331 memory (most machines, nowadays). This assumes that the innermost
13332 stack frame is selected; setting @code{$sp} is not allowed when other
13333 stack frames are selected. To pop entire frames off the stack,
13334 regardless of machine architecture, use @code{return};
13335 see @ref{Returning, ,Returning from a Function}.} with
13341 Whenever possible, these four standard register names are available on
13342 your machine even though the machine has different canonical mnemonics,
13343 so long as there is no conflict. The @code{info registers} command
13344 shows the canonical names. For example, on the SPARC, @code{info
13345 registers} displays the processor status register as @code{$psr} but you
13346 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
13347 is an alias for the @sc{eflags} register.
13349 @value{GDBN} always considers the contents of an ordinary register as an
13350 integer when the register is examined in this way. Some machines have
13351 special registers which can hold nothing but floating point; these
13352 registers are considered to have floating point values. There is no way
13353 to refer to the contents of an ordinary register as floating point value
13354 (although you can @emph{print} it as a floating point value with
13355 @samp{print/f $@var{regname}}).
13357 Some registers have distinct ``raw'' and ``virtual'' data formats. This
13358 means that the data format in which the register contents are saved by
13359 the operating system is not the same one that your program normally
13360 sees. For example, the registers of the 68881 floating point
13361 coprocessor are always saved in ``extended'' (raw) format, but all C
13362 programs expect to work with ``double'' (virtual) format. In such
13363 cases, @value{GDBN} normally works with the virtual format only (the format
13364 that makes sense for your program), but the @code{info registers} command
13365 prints the data in both formats.
13367 @cindex SSE registers (x86)
13368 @cindex MMX registers (x86)
13369 Some machines have special registers whose contents can be interpreted
13370 in several different ways. For example, modern x86-based machines
13371 have SSE and MMX registers that can hold several values packed
13372 together in several different formats. @value{GDBN} refers to such
13373 registers in @code{struct} notation:
13376 (@value{GDBP}) print $xmm1
13378 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
13379 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
13380 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
13381 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
13382 v4_int32 = @{0, 20657912, 11, 13@},
13383 v2_int64 = @{88725056443645952, 55834574859@},
13384 uint128 = 0x0000000d0000000b013b36f800000000
13389 To set values of such registers, you need to tell @value{GDBN} which
13390 view of the register you wish to change, as if you were assigning
13391 value to a @code{struct} member:
13394 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
13397 Normally, register values are relative to the selected stack frame
13398 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
13399 value that the register would contain if all stack frames farther in
13400 were exited and their saved registers restored. In order to see the
13401 true contents of hardware registers, you must select the innermost
13402 frame (with @samp{frame 0}).
13404 @cindex caller-saved registers
13405 @cindex call-clobbered registers
13406 @cindex volatile registers
13407 @cindex <not saved> values
13408 Usually ABIs reserve some registers as not needed to be saved by the
13409 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
13410 registers). It may therefore not be possible for @value{GDBN} to know
13411 the value a register had before the call (in other words, in the outer
13412 frame), if the register value has since been changed by the callee.
13413 @value{GDBN} tries to deduce where the inner frame saved
13414 (``callee-saved'') registers, from the debug info, unwind info, or the
13415 machine code generated by your compiler. If some register is not
13416 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
13417 its own knowledge of the ABI, or because the debug/unwind info
13418 explicitly says the register's value is undefined), @value{GDBN}
13419 displays @w{@samp{<not saved>}} as the register's value. With targets
13420 that @value{GDBN} has no knowledge of the register saving convention,
13421 if a register was not saved by the callee, then its value and location
13422 in the outer frame are assumed to be the same of the inner frame.
13423 This is usually harmless, because if the register is call-clobbered,
13424 the caller either does not care what is in the register after the
13425 call, or has code to restore the value that it does care about. Note,
13426 however, that if you change such a register in the outer frame, you
13427 may also be affecting the inner frame. Also, the more ``outer'' the
13428 frame is you're looking at, the more likely a call-clobbered
13429 register's value is to be wrong, in the sense that it doesn't actually
13430 represent the value the register had just before the call.
13432 @node Floating Point Hardware
13433 @section Floating Point Hardware
13434 @cindex floating point
13436 Depending on the configuration, @value{GDBN} may be able to give
13437 you more information about the status of the floating point hardware.
13442 Display hardware-dependent information about the floating
13443 point unit. The exact contents and layout vary depending on the
13444 floating point chip. Currently, @samp{info float} is supported on
13445 the ARM and x86 machines.
13449 @section Vector Unit
13450 @cindex vector unit
13452 Depending on the configuration, @value{GDBN} may be able to give you
13453 more information about the status of the vector unit.
13456 @kindex info vector
13458 Display information about the vector unit. The exact contents and
13459 layout vary depending on the hardware.
13462 @node OS Information
13463 @section Operating System Auxiliary Information
13464 @cindex OS information
13466 @value{GDBN} provides interfaces to useful OS facilities that can help
13467 you debug your program.
13469 @cindex auxiliary vector
13470 @cindex vector, auxiliary
13471 Some operating systems supply an @dfn{auxiliary vector} to programs at
13472 startup. This is akin to the arguments and environment that you
13473 specify for a program, but contains a system-dependent variety of
13474 binary values that tell system libraries important details about the
13475 hardware, operating system, and process. Each value's purpose is
13476 identified by an integer tag; the meanings are well-known but system-specific.
13477 Depending on the configuration and operating system facilities,
13478 @value{GDBN} may be able to show you this information. For remote
13479 targets, this functionality may further depend on the remote stub's
13480 support of the @samp{qXfer:auxv:read} packet, see
13481 @ref{qXfer auxiliary vector read}.
13486 Display the auxiliary vector of the inferior, which can be either a
13487 live process or a core dump file. @value{GDBN} prints each tag value
13488 numerically, and also shows names and text descriptions for recognized
13489 tags. Some values in the vector are numbers, some bit masks, and some
13490 pointers to strings or other data. @value{GDBN} displays each value in the
13491 most appropriate form for a recognized tag, and in hexadecimal for
13492 an unrecognized tag.
13495 On some targets, @value{GDBN} can access operating system-specific
13496 information and show it to you. The types of information available
13497 will differ depending on the type of operating system running on the
13498 target. The mechanism used to fetch the data is described in
13499 @ref{Operating System Information}. For remote targets, this
13500 functionality depends on the remote stub's support of the
13501 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
13505 @item info os @var{infotype}
13507 Display OS information of the requested type.
13509 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
13511 @anchor{linux info os infotypes}
13513 @kindex info os cpus
13515 Display the list of all CPUs/cores. For each CPU/core, @value{GDBN} prints
13516 the available fields from /proc/cpuinfo. For each supported architecture
13517 different fields are available. Two common entries are processor which gives
13518 CPU number and bogomips; a system constant that is calculated during
13519 kernel initialization.
13521 @kindex info os files
13523 Display the list of open file descriptors on the target. For each
13524 file descriptor, @value{GDBN} prints the identifier of the process
13525 owning the descriptor, the command of the owning process, the value
13526 of the descriptor, and the target of the descriptor.
13528 @kindex info os modules
13530 Display the list of all loaded kernel modules on the target. For each
13531 module, @value{GDBN} prints the module name, the size of the module in
13532 bytes, the number of times the module is used, the dependencies of the
13533 module, the status of the module, and the address of the loaded module
13536 @kindex info os msg
13538 Display the list of all System V message queues on the target. For each
13539 message queue, @value{GDBN} prints the message queue key, the message
13540 queue identifier, the access permissions, the current number of bytes
13541 on the queue, the current number of messages on the queue, the processes
13542 that last sent and received a message on the queue, the user and group
13543 of the owner and creator of the message queue, the times at which a
13544 message was last sent and received on the queue, and the time at which
13545 the message queue was last changed.
13547 @kindex info os processes
13549 Display the list of processes on the target. For each process,
13550 @value{GDBN} prints the process identifier, the name of the user, the
13551 command corresponding to the process, and the list of processor cores
13552 that the process is currently running on. (To understand what these
13553 properties mean, for this and the following info types, please consult
13554 the general @sc{gnu}/Linux documentation.)
13556 @kindex info os procgroups
13558 Display the list of process groups on the target. For each process,
13559 @value{GDBN} prints the identifier of the process group that it belongs
13560 to, the command corresponding to the process group leader, the process
13561 identifier, and the command line of the process. The list is sorted
13562 first by the process group identifier, then by the process identifier,
13563 so that processes belonging to the same process group are grouped together
13564 and the process group leader is listed first.
13566 @kindex info os semaphores
13568 Display the list of all System V semaphore sets on the target. For each
13569 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
13570 set identifier, the access permissions, the number of semaphores in the
13571 set, the user and group of the owner and creator of the semaphore set,
13572 and the times at which the semaphore set was operated upon and changed.
13574 @kindex info os shm
13576 Display the list of all System V shared-memory regions on the target.
13577 For each shared-memory region, @value{GDBN} prints the region key,
13578 the shared-memory identifier, the access permissions, the size of the
13579 region, the process that created the region, the process that last
13580 attached to or detached from the region, the current number of live
13581 attaches to the region, and the times at which the region was last
13582 attached to, detach from, and changed.
13584 @kindex info os sockets
13586 Display the list of Internet-domain sockets on the target. For each
13587 socket, @value{GDBN} prints the address and port of the local and
13588 remote endpoints, the current state of the connection, the creator of
13589 the socket, the IP address family of the socket, and the type of the
13592 @kindex info os threads
13594 Display the list of threads running on the target. For each thread,
13595 @value{GDBN} prints the identifier of the process that the thread
13596 belongs to, the command of the process, the thread identifier, and the
13597 processor core that it is currently running on. The main thread of a
13598 process is not listed.
13602 If @var{infotype} is omitted, then list the possible values for
13603 @var{infotype} and the kind of OS information available for each
13604 @var{infotype}. If the target does not return a list of possible
13605 types, this command will report an error.
13608 @node Memory Region Attributes
13609 @section Memory Region Attributes
13610 @cindex memory region attributes
13612 @dfn{Memory region attributes} allow you to describe special handling
13613 required by regions of your target's memory. @value{GDBN} uses
13614 attributes to determine whether to allow certain types of memory
13615 accesses; whether to use specific width accesses; and whether to cache
13616 target memory. By default the description of memory regions is
13617 fetched from the target (if the current target supports this), but the
13618 user can override the fetched regions.
13620 Defined memory regions can be individually enabled and disabled. When a
13621 memory region is disabled, @value{GDBN} uses the default attributes when
13622 accessing memory in that region. Similarly, if no memory regions have
13623 been defined, @value{GDBN} uses the default attributes when accessing
13626 When a memory region is defined, it is given a number to identify it;
13627 to enable, disable, or remove a memory region, you specify that number.
13631 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
13632 Define a memory region bounded by @var{lower} and @var{upper} with
13633 attributes @var{attributes}@dots{}, and add it to the list of regions
13634 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
13635 case: it is treated as the target's maximum memory address.
13636 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
13639 Discard any user changes to the memory regions and use target-supplied
13640 regions, if available, or no regions if the target does not support.
13643 @item delete mem @var{nums}@dots{}
13644 Remove memory regions @var{nums}@dots{} from the list of regions
13645 monitored by @value{GDBN}.
13647 @kindex disable mem
13648 @item disable mem @var{nums}@dots{}
13649 Disable monitoring of memory regions @var{nums}@dots{}.
13650 A disabled memory region is not forgotten.
13651 It may be enabled again later.
13654 @item enable mem @var{nums}@dots{}
13655 Enable monitoring of memory regions @var{nums}@dots{}.
13659 Print a table of all defined memory regions, with the following columns
13663 @item Memory Region Number
13664 @item Enabled or Disabled.
13665 Enabled memory regions are marked with @samp{y}.
13666 Disabled memory regions are marked with @samp{n}.
13669 The address defining the inclusive lower bound of the memory region.
13672 The address defining the exclusive upper bound of the memory region.
13675 The list of attributes set for this memory region.
13680 @subsection Attributes
13682 @subsubsection Memory Access Mode
13683 The access mode attributes set whether @value{GDBN} may make read or
13684 write accesses to a memory region.
13686 While these attributes prevent @value{GDBN} from performing invalid
13687 memory accesses, they do nothing to prevent the target system, I/O DMA,
13688 etc.@: from accessing memory.
13692 Memory is read only.
13694 Memory is write only.
13696 Memory is read/write. This is the default.
13699 @subsubsection Memory Access Size
13700 The access size attribute tells @value{GDBN} to use specific sized
13701 accesses in the memory region. Often memory mapped device registers
13702 require specific sized accesses. If no access size attribute is
13703 specified, @value{GDBN} may use accesses of any size.
13707 Use 8 bit memory accesses.
13709 Use 16 bit memory accesses.
13711 Use 32 bit memory accesses.
13713 Use 64 bit memory accesses.
13716 @c @subsubsection Hardware/Software Breakpoints
13717 @c The hardware/software breakpoint attributes set whether @value{GDBN}
13718 @c will use hardware or software breakpoints for the internal breakpoints
13719 @c used by the step, next, finish, until, etc. commands.
13723 @c Always use hardware breakpoints
13724 @c @item swbreak (default)
13727 @subsubsection Data Cache
13728 The data cache attributes set whether @value{GDBN} will cache target
13729 memory. While this generally improves performance by reducing debug
13730 protocol overhead, it can lead to incorrect results because @value{GDBN}
13731 does not know about volatile variables or memory mapped device
13736 Enable @value{GDBN} to cache target memory.
13738 Disable @value{GDBN} from caching target memory. This is the default.
13741 @subsection Memory Access Checking
13742 @value{GDBN} can be instructed to refuse accesses to memory that is
13743 not explicitly described. This can be useful if accessing such
13744 regions has undesired effects for a specific target, or to provide
13745 better error checking. The following commands control this behaviour.
13748 @kindex set mem inaccessible-by-default
13749 @item set mem inaccessible-by-default [on|off]
13750 If @code{on} is specified, make @value{GDBN} treat memory not
13751 explicitly described by the memory ranges as non-existent and refuse accesses
13752 to such memory. The checks are only performed if there's at least one
13753 memory range defined. If @code{off} is specified, make @value{GDBN}
13754 treat the memory not explicitly described by the memory ranges as RAM.
13755 The default value is @code{on}.
13756 @kindex show mem inaccessible-by-default
13757 @item show mem inaccessible-by-default
13758 Show the current handling of accesses to unknown memory.
13762 @c @subsubsection Memory Write Verification
13763 @c The memory write verification attributes set whether @value{GDBN}
13764 @c will re-reads data after each write to verify the write was successful.
13768 @c @item noverify (default)
13771 @node Dump/Restore Files
13772 @section Copy Between Memory and a File
13773 @cindex dump/restore files
13774 @cindex append data to a file
13775 @cindex dump data to a file
13776 @cindex restore data from a file
13778 You can use the commands @code{dump}, @code{append}, and
13779 @code{restore} to copy data between target memory and a file. The
13780 @code{dump} and @code{append} commands write data to a file, and the
13781 @code{restore} command reads data from a file back into the inferior's
13782 memory. Files may be in binary, Motorola S-record, Intel hex,
13783 Tektronix Hex, or Verilog Hex format; however, @value{GDBN} can only
13784 append to binary files, and cannot read from Verilog Hex files.
13789 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
13790 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
13791 Dump the contents of memory from @var{start_addr} to @var{end_addr},
13792 or the value of @var{expr}, to @var{filename} in the given format.
13794 The @var{format} parameter may be any one of:
13801 Motorola S-record format.
13803 Tektronix Hex format.
13805 Verilog Hex format.
13808 @value{GDBN} uses the same definitions of these formats as the
13809 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
13810 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
13814 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
13815 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
13816 Append the contents of memory from @var{start_addr} to @var{end_addr},
13817 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
13818 (@value{GDBN} can only append data to files in raw binary form.)
13821 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
13822 Restore the contents of file @var{filename} into memory. The
13823 @code{restore} command can automatically recognize any known @sc{bfd}
13824 file format, except for raw binary. To restore a raw binary file you
13825 must specify the optional keyword @code{binary} after the filename.
13827 If @var{bias} is non-zero, its value will be added to the addresses
13828 contained in the file. Binary files always start at address zero, so
13829 they will be restored at address @var{bias}. Other bfd files have
13830 a built-in location; they will be restored at offset @var{bias}
13831 from that location.
13833 If @var{start} and/or @var{end} are non-zero, then only data between
13834 file offset @var{start} and file offset @var{end} will be restored.
13835 These offsets are relative to the addresses in the file, before
13836 the @var{bias} argument is applied.
13840 @node Core File Generation
13841 @section How to Produce a Core File from Your Program
13842 @cindex dump core from inferior
13844 A @dfn{core file} or @dfn{core dump} is a file that records the memory
13845 image of a running process and its process status (register values
13846 etc.). Its primary use is post-mortem debugging of a program that
13847 crashed while it ran outside a debugger. A program that crashes
13848 automatically produces a core file, unless this feature is disabled by
13849 the user. @xref{Files}, for information on invoking @value{GDBN} in
13850 the post-mortem debugging mode.
13852 Occasionally, you may wish to produce a core file of the program you
13853 are debugging in order to preserve a snapshot of its state.
13854 @value{GDBN} has a special command for that.
13858 @kindex generate-core-file
13859 @item generate-core-file [@var{file}]
13860 @itemx gcore [@var{file}]
13861 Produce a core dump of the inferior process. The optional argument
13862 @var{file} specifies the file name where to put the core dump. If not
13863 specified, the file name defaults to @file{core.@var{pid}}, where
13864 @var{pid} is the inferior process ID.
13866 Note that this command is implemented only for some systems (as of
13867 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
13869 On @sc{gnu}/Linux, this command can take into account the value of the
13870 file @file{/proc/@var{pid}/coredump_filter} when generating the core
13871 dump (@pxref{set use-coredump-filter}), and by default honors the
13872 @code{VM_DONTDUMP} flag for mappings where it is present in the file
13873 @file{/proc/@var{pid}/smaps} (@pxref{set dump-excluded-mappings}).
13875 @kindex set use-coredump-filter
13876 @anchor{set use-coredump-filter}
13877 @item set use-coredump-filter on
13878 @itemx set use-coredump-filter off
13879 Enable or disable the use of the file
13880 @file{/proc/@var{pid}/coredump_filter} when generating core dump
13881 files. This file is used by the Linux kernel to decide what types of
13882 memory mappings will be dumped or ignored when generating a core dump
13883 file. @var{pid} is the process ID of a currently running process.
13885 To make use of this feature, you have to write in the
13886 @file{/proc/@var{pid}/coredump_filter} file a value, in hexadecimal,
13887 which is a bit mask representing the memory mapping types. If a bit
13888 is set in the bit mask, then the memory mappings of the corresponding
13889 types will be dumped; otherwise, they will be ignored. This
13890 configuration is inherited by child processes. For more information
13891 about the bits that can be set in the
13892 @file{/proc/@var{pid}/coredump_filter} file, please refer to the
13893 manpage of @code{core(5)}.
13895 By default, this option is @code{on}. If this option is turned
13896 @code{off}, @value{GDBN} does not read the @file{coredump_filter} file
13897 and instead uses the same default value as the Linux kernel in order
13898 to decide which pages will be dumped in the core dump file. This
13899 value is currently @code{0x33}, which means that bits @code{0}
13900 (anonymous private mappings), @code{1} (anonymous shared mappings),
13901 @code{4} (ELF headers) and @code{5} (private huge pages) are active.
13902 This will cause these memory mappings to be dumped automatically.
13904 @kindex set dump-excluded-mappings
13905 @anchor{set dump-excluded-mappings}
13906 @item set dump-excluded-mappings on
13907 @itemx set dump-excluded-mappings off
13908 If @code{on} is specified, @value{GDBN} will dump memory mappings
13909 marked with the @code{VM_DONTDUMP} flag. This flag is represented in
13910 the file @file{/proc/@var{pid}/smaps} with the acronym @code{dd}.
13912 The default value is @code{off}.
13915 @node Character Sets
13916 @section Character Sets
13917 @cindex character sets
13919 @cindex translating between character sets
13920 @cindex host character set
13921 @cindex target character set
13923 If the program you are debugging uses a different character set to
13924 represent characters and strings than the one @value{GDBN} uses itself,
13925 @value{GDBN} can automatically translate between the character sets for
13926 you. The character set @value{GDBN} uses we call the @dfn{host
13927 character set}; the one the inferior program uses we call the
13928 @dfn{target character set}.
13930 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
13931 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
13932 remote protocol (@pxref{Remote Debugging}) to debug a program
13933 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
13934 then the host character set is Latin-1, and the target character set is
13935 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
13936 target-charset EBCDIC-US}, then @value{GDBN} translates between
13937 @sc{ebcdic} and Latin 1 as you print character or string values, or use
13938 character and string literals in expressions.
13940 @value{GDBN} has no way to automatically recognize which character set
13941 the inferior program uses; you must tell it, using the @code{set
13942 target-charset} command, described below.
13944 Here are the commands for controlling @value{GDBN}'s character set
13948 @item set target-charset @var{charset}
13949 @kindex set target-charset
13950 Set the current target character set to @var{charset}. To display the
13951 list of supported target character sets, type
13952 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
13954 @item set host-charset @var{charset}
13955 @kindex set host-charset
13956 Set the current host character set to @var{charset}.
13958 By default, @value{GDBN} uses a host character set appropriate to the
13959 system it is running on; you can override that default using the
13960 @code{set host-charset} command. On some systems, @value{GDBN} cannot
13961 automatically determine the appropriate host character set. In this
13962 case, @value{GDBN} uses @samp{UTF-8}.
13964 @value{GDBN} can only use certain character sets as its host character
13965 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
13966 @value{GDBN} will list the host character sets it supports.
13968 @item set charset @var{charset}
13969 @kindex set charset
13970 Set the current host and target character sets to @var{charset}. As
13971 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
13972 @value{GDBN} will list the names of the character sets that can be used
13973 for both host and target.
13976 @kindex show charset
13977 Show the names of the current host and target character sets.
13979 @item show host-charset
13980 @kindex show host-charset
13981 Show the name of the current host character set.
13983 @item show target-charset
13984 @kindex show target-charset
13985 Show the name of the current target character set.
13987 @item set target-wide-charset @var{charset}
13988 @kindex set target-wide-charset
13989 Set the current target's wide character set to @var{charset}. This is
13990 the character set used by the target's @code{wchar_t} type. To
13991 display the list of supported wide character sets, type
13992 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
13994 @item show target-wide-charset
13995 @kindex show target-wide-charset
13996 Show the name of the current target's wide character set.
13999 Here is an example of @value{GDBN}'s character set support in action.
14000 Assume that the following source code has been placed in the file
14001 @file{charset-test.c}:
14007 = @{72, 101, 108, 108, 111, 44, 32, 119,
14008 111, 114, 108, 100, 33, 10, 0@};
14009 char ibm1047_hello[]
14010 = @{200, 133, 147, 147, 150, 107, 64, 166,
14011 150, 153, 147, 132, 90, 37, 0@};
14015 printf ("Hello, world!\n");
14019 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
14020 containing the string @samp{Hello, world!} followed by a newline,
14021 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
14023 We compile the program, and invoke the debugger on it:
14026 $ gcc -g charset-test.c -o charset-test
14027 $ gdb -nw charset-test
14028 GNU gdb 2001-12-19-cvs
14029 Copyright 2001 Free Software Foundation, Inc.
14034 We can use the @code{show charset} command to see what character sets
14035 @value{GDBN} is currently using to interpret and display characters and
14039 (@value{GDBP}) show charset
14040 The current host and target character set is `ISO-8859-1'.
14044 For the sake of printing this manual, let's use @sc{ascii} as our
14045 initial character set:
14047 (@value{GDBP}) set charset ASCII
14048 (@value{GDBP}) show charset
14049 The current host and target character set is `ASCII'.
14053 Let's assume that @sc{ascii} is indeed the correct character set for our
14054 host system --- in other words, let's assume that if @value{GDBN} prints
14055 characters using the @sc{ascii} character set, our terminal will display
14056 them properly. Since our current target character set is also
14057 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
14060 (@value{GDBP}) print ascii_hello
14061 $1 = 0x401698 "Hello, world!\n"
14062 (@value{GDBP}) print ascii_hello[0]
14067 @value{GDBN} uses the target character set for character and string
14068 literals you use in expressions:
14071 (@value{GDBP}) print '+'
14076 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
14079 @value{GDBN} relies on the user to tell it which character set the
14080 target program uses. If we print @code{ibm1047_hello} while our target
14081 character set is still @sc{ascii}, we get jibberish:
14084 (@value{GDBP}) print ibm1047_hello
14085 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
14086 (@value{GDBP}) print ibm1047_hello[0]
14091 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
14092 @value{GDBN} tells us the character sets it supports:
14095 (@value{GDBP}) set target-charset
14096 ASCII EBCDIC-US IBM1047 ISO-8859-1
14097 (@value{GDBP}) set target-charset
14100 We can select @sc{ibm1047} as our target character set, and examine the
14101 program's strings again. Now the @sc{ascii} string is wrong, but
14102 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
14103 target character set, @sc{ibm1047}, to the host character set,
14104 @sc{ascii}, and they display correctly:
14107 (@value{GDBP}) set target-charset IBM1047
14108 (@value{GDBP}) show charset
14109 The current host character set is `ASCII'.
14110 The current target character set is `IBM1047'.
14111 (@value{GDBP}) print ascii_hello
14112 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
14113 (@value{GDBP}) print ascii_hello[0]
14115 (@value{GDBP}) print ibm1047_hello
14116 $8 = 0x4016a8 "Hello, world!\n"
14117 (@value{GDBP}) print ibm1047_hello[0]
14122 As above, @value{GDBN} uses the target character set for character and
14123 string literals you use in expressions:
14126 (@value{GDBP}) print '+'
14131 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
14134 @node Caching Target Data
14135 @section Caching Data of Targets
14136 @cindex caching data of targets
14138 @value{GDBN} caches data exchanged between the debugger and a target.
14139 Each cache is associated with the address space of the inferior.
14140 @xref{Inferiors Connections and Programs}, about inferior and address space.
14141 Such caching generally improves performance in remote debugging
14142 (@pxref{Remote Debugging}), because it reduces the overhead of the
14143 remote protocol by bundling memory reads and writes into large chunks.
14144 Unfortunately, simply caching everything would lead to incorrect results,
14145 since @value{GDBN} does not necessarily know anything about volatile
14146 values, memory-mapped I/O addresses, etc. Furthermore, in non-stop mode
14147 (@pxref{Non-Stop Mode}) memory can be changed @emph{while} a gdb command
14149 Therefore, by default, @value{GDBN} only caches data
14150 known to be on the stack@footnote{In non-stop mode, it is moderately
14151 rare for a running thread to modify the stack of a stopped thread
14152 in a way that would interfere with a backtrace, and caching of
14153 stack reads provides a significant speed up of remote backtraces.} or
14154 in the code segment.
14155 Other regions of memory can be explicitly marked as
14156 cacheable; @pxref{Memory Region Attributes}.
14159 @kindex set remotecache
14160 @item set remotecache on
14161 @itemx set remotecache off
14162 This option no longer does anything; it exists for compatibility
14165 @kindex show remotecache
14166 @item show remotecache
14167 Show the current state of the obsolete remotecache flag.
14169 @kindex set stack-cache
14170 @item set stack-cache on
14171 @itemx set stack-cache off
14172 Enable or disable caching of stack accesses. When @code{on}, use
14173 caching. By default, this option is @code{on}.
14175 @kindex show stack-cache
14176 @item show stack-cache
14177 Show the current state of data caching for memory accesses.
14179 @kindex set code-cache
14180 @item set code-cache on
14181 @itemx set code-cache off
14182 Enable or disable caching of code segment accesses. When @code{on},
14183 use caching. By default, this option is @code{on}. This improves
14184 performance of disassembly in remote debugging.
14186 @kindex show code-cache
14187 @item show code-cache
14188 Show the current state of target memory cache for code segment
14191 @kindex info dcache
14192 @item info dcache @r{[}line@r{]}
14193 Print the information about the performance of data cache of the
14194 current inferior's address space. The information displayed
14195 includes the dcache width and depth, and for each cache line, its
14196 number, address, and how many times it was referenced. This
14197 command is useful for debugging the data cache operation.
14199 If a line number is specified, the contents of that line will be
14202 @item set dcache size @var{size}
14203 @cindex dcache size
14204 @kindex set dcache size
14205 Set maximum number of entries in dcache (dcache depth above).
14207 @item set dcache line-size @var{line-size}
14208 @cindex dcache line-size
14209 @kindex set dcache line-size
14210 Set number of bytes each dcache entry caches (dcache width above).
14211 Must be a power of 2.
14213 @item show dcache size
14214 @kindex show dcache size
14215 Show maximum number of dcache entries. @xref{Caching Target Data, info dcache}.
14217 @item show dcache line-size
14218 @kindex show dcache line-size
14219 Show default size of dcache lines.
14221 @item maint flush dcache
14222 @cindex dcache, flushing
14223 @kindex maint flush dcache
14224 Flush the contents (if any) of the dcache. This maintainer command is
14225 useful when debugging the dcache implementation.
14229 @node Searching Memory
14230 @section Search Memory
14231 @cindex searching memory
14233 Memory can be searched for a particular sequence of bytes with the
14234 @code{find} command.
14238 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
14239 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
14240 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
14241 etc. The search begins at address @var{start_addr} and continues for either
14242 @var{len} bytes or through to @var{end_addr} inclusive.
14245 @var{s} and @var{n} are optional parameters.
14246 They may be specified in either order, apart or together.
14249 @item @var{s}, search query size
14250 The size of each search query value.
14256 halfwords (two bytes)
14260 giant words (eight bytes)
14263 All values are interpreted in the current language.
14264 This means, for example, that if the current source language is C/C@t{++}
14265 then searching for the string ``hello'' includes the trailing '\0'.
14266 The null terminator can be removed from searching by using casts,
14267 e.g.: @samp{@{char[5]@}"hello"}.
14269 If the value size is not specified, it is taken from the
14270 value's type in the current language.
14271 This is useful when one wants to specify the search
14272 pattern as a mixture of types.
14273 Note that this means, for example, that in the case of C-like languages
14274 a search for an untyped 0x42 will search for @samp{(int) 0x42}
14275 which is typically four bytes.
14277 @item @var{n}, maximum number of finds
14278 The maximum number of matches to print. The default is to print all finds.
14281 You can use strings as search values. Quote them with double-quotes
14283 The string value is copied into the search pattern byte by byte,
14284 regardless of the endianness of the target and the size specification.
14286 The address of each match found is printed as well as a count of the
14287 number of matches found.
14289 The address of the last value found is stored in convenience variable
14291 A count of the number of matches is stored in @samp{$numfound}.
14293 For example, if stopped at the @code{printf} in this function:
14299 static char hello[] = "hello-hello";
14300 static struct @{ char c; short s; int i; @}
14301 __attribute__ ((packed)) mixed
14302 = @{ 'c', 0x1234, 0x87654321 @};
14303 printf ("%s\n", hello);
14308 you get during debugging:
14311 (@value{GDBP}) find &hello[0], +sizeof(hello), "hello"
14312 0x804956d <hello.1620+6>
14314 (@value{GDBP}) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
14315 0x8049567 <hello.1620>
14316 0x804956d <hello.1620+6>
14318 (@value{GDBP}) find &hello[0], +sizeof(hello), @{char[5]@}"hello"
14319 0x8049567 <hello.1620>
14320 0x804956d <hello.1620+6>
14322 (@value{GDBP}) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
14323 0x8049567 <hello.1620>
14325 (@value{GDBP}) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
14326 0x8049560 <mixed.1625>
14328 (@value{GDBP}) print $numfound
14330 (@value{GDBP}) print $_
14331 $2 = (void *) 0x8049560
14335 @section Value Sizes
14337 Whenever @value{GDBN} prints a value memory will be allocated within
14338 @value{GDBN} to hold the contents of the value. It is possible in
14339 some languages with dynamic typing systems, that an invalid program
14340 may indicate a value that is incorrectly large, this in turn may cause
14341 @value{GDBN} to try and allocate an overly large amount of memory.
14344 @anchor{set max-value-size}
14345 @kindex set max-value-size
14346 @item set max-value-size @var{bytes}
14347 @itemx set max-value-size unlimited
14348 Set the maximum size of memory that @value{GDBN} will allocate for the
14349 contents of a value to @var{bytes}, trying to display a value that
14350 requires more memory than that will result in an error.
14352 Setting this variable does not effect values that have already been
14353 allocated within @value{GDBN}, only future allocations.
14355 There's a minimum size that @code{max-value-size} can be set to in
14356 order that @value{GDBN} can still operate correctly, this minimum is
14357 currently 16 bytes.
14359 The limit applies to the results of some subexpressions as well as to
14360 complete expressions. For example, an expression denoting a simple
14361 integer component, such as @code{x.y.z}, may fail if the size of
14362 @var{x.y} is dynamic and exceeds @var{bytes}. On the other hand,
14363 @value{GDBN} is sometimes clever; the expression @code{A[i]}, where
14364 @var{A} is an array variable with non-constant size, will generally
14365 succeed regardless of the bounds on @var{A}, as long as the component
14366 size is less than @var{bytes}.
14368 The default value of @code{max-value-size} is currently 64k.
14370 @kindex show max-value-size
14371 @item show max-value-size
14372 Show the maximum size of memory, in bytes, that @value{GDBN} will
14373 allocate for the contents of a value.
14376 @node Optimized Code
14377 @chapter Debugging Optimized Code
14378 @cindex optimized code, debugging
14379 @cindex debugging optimized code
14381 Almost all compilers support optimization. With optimization
14382 disabled, the compiler generates assembly code that corresponds
14383 directly to your source code, in a simplistic way. As the compiler
14384 applies more powerful optimizations, the generated assembly code
14385 diverges from your original source code. With help from debugging
14386 information generated by the compiler, @value{GDBN} can map from
14387 the running program back to constructs from your original source.
14389 @value{GDBN} is more accurate with optimization disabled. If you
14390 can recompile without optimization, it is easier to follow the
14391 progress of your program during debugging. But, there are many cases
14392 where you may need to debug an optimized version.
14394 When you debug a program compiled with @samp{-g -O}, remember that the
14395 optimizer has rearranged your code; the debugger shows you what is
14396 really there. Do not be too surprised when the execution path does not
14397 exactly match your source file! An extreme example: if you define a
14398 variable, but never use it, @value{GDBN} never sees that
14399 variable---because the compiler optimizes it out of existence.
14401 Some things do not work as well with @samp{-g -O} as with just
14402 @samp{-g}, particularly on machines with instruction scheduling. If in
14403 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
14404 please report it to us as a bug (including a test case!).
14405 @xref{Variables}, for more information about debugging optimized code.
14408 * Inline Functions:: How @value{GDBN} presents inlining
14409 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
14412 @node Inline Functions
14413 @section Inline Functions
14414 @cindex inline functions, debugging
14416 @dfn{Inlining} is an optimization that inserts a copy of the function
14417 body directly at each call site, instead of jumping to a shared
14418 routine. @value{GDBN} displays inlined functions just like
14419 non-inlined functions. They appear in backtraces. You can view their
14420 arguments and local variables, step into them with @code{step}, skip
14421 them with @code{next}, and escape from them with @code{finish}.
14422 You can check whether a function was inlined by using the
14423 @code{info frame} command.
14425 For @value{GDBN} to support inlined functions, the compiler must
14426 record information about inlining in the debug information ---
14427 @value{NGCC} using the @sc{dwarf 2} format does this, and several
14428 other compilers do also. @value{GDBN} only supports inlined functions
14429 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
14430 do not emit two required attributes (@samp{DW_AT_call_file} and
14431 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
14432 function calls with earlier versions of @value{NGCC}. It instead
14433 displays the arguments and local variables of inlined functions as
14434 local variables in the caller.
14436 The body of an inlined function is directly included at its call site;
14437 unlike a non-inlined function, there are no instructions devoted to
14438 the call. @value{GDBN} still pretends that the call site and the
14439 start of the inlined function are different instructions. Stepping to
14440 the call site shows the call site, and then stepping again shows
14441 the first line of the inlined function, even though no additional
14442 instructions are executed.
14444 This makes source-level debugging much clearer; you can see both the
14445 context of the call and then the effect of the call. Only stepping by
14446 a single instruction using @code{stepi} or @code{nexti} does not do
14447 this; single instruction steps always show the inlined body.
14449 There are some ways that @value{GDBN} does not pretend that inlined
14450 function calls are the same as normal calls:
14454 Setting breakpoints at the call site of an inlined function may not
14455 work, because the call site does not contain any code. @value{GDBN}
14456 may incorrectly move the breakpoint to the next line of the enclosing
14457 function, after the call. This limitation will be removed in a future
14458 version of @value{GDBN}; until then, set a breakpoint on an earlier line
14459 or inside the inlined function instead.
14462 @value{GDBN} cannot locate the return value of inlined calls after
14463 using the @code{finish} command. This is a limitation of compiler-generated
14464 debugging information; after @code{finish}, you can step to the next line
14465 and print a variable where your program stored the return value.
14469 @node Tail Call Frames
14470 @section Tail Call Frames
14471 @cindex tail call frames, debugging
14473 Function @code{B} can call function @code{C} in its very last statement. In
14474 unoptimized compilation the call of @code{C} is immediately followed by return
14475 instruction at the end of @code{B} code. Optimizing compiler may replace the
14476 call and return in function @code{B} into one jump to function @code{C}
14477 instead. Such use of a jump instruction is called @dfn{tail call}.
14479 During execution of function @code{C}, there will be no indication in the
14480 function call stack frames that it was tail-called from @code{B}. If function
14481 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
14482 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
14483 some cases @value{GDBN} can determine that @code{C} was tail-called from
14484 @code{B}, and it will then create fictitious call frame for that, with the
14485 return address set up as if @code{B} called @code{C} normally.
14487 This functionality is currently supported only by DWARF 2 debugging format and
14488 the compiler has to produce @samp{DW_TAG_call_site} tags. With
14489 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
14492 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
14493 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
14496 (@value{GDBP}) x/i $pc - 2
14497 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
14498 (@value{GDBP}) info frame
14499 Stack level 1, frame at 0x7fffffffda30:
14500 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
14501 tail call frame, caller of frame at 0x7fffffffda30
14502 source language c++.
14503 Arglist at unknown address.
14504 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
14507 The detection of all the possible code path executions can find them ambiguous.
14508 There is no execution history stored (possible @ref{Reverse Execution} is never
14509 used for this purpose) and the last known caller could have reached the known
14510 callee by multiple different jump sequences. In such case @value{GDBN} still
14511 tries to show at least all the unambiguous top tail callers and all the
14512 unambiguous bottom tail calees, if any.
14515 @anchor{set debug entry-values}
14516 @item set debug entry-values
14517 @kindex set debug entry-values
14518 When set to on, enables printing of analysis messages for both frame argument
14519 values at function entry and tail calls. It will show all the possible valid
14520 tail calls code paths it has considered. It will also print the intersection
14521 of them with the final unambiguous (possibly partial or even empty) code path
14524 @item show debug entry-values
14525 @kindex show debug entry-values
14526 Show the current state of analysis messages printing for both frame argument
14527 values at function entry and tail calls.
14530 The analysis messages for tail calls can for example show why the virtual tail
14531 call frame for function @code{c} has not been recognized (due to the indirect
14532 reference by variable @code{x}):
14535 static void __attribute__((noinline, noclone)) c (void);
14536 void (*x) (void) = c;
14537 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
14538 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
14539 int main (void) @{ x (); return 0; @}
14541 Breakpoint 1, DW_OP_entry_value resolving cannot find
14542 DW_TAG_call_site 0x40039a in main
14544 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
14547 #1 0x000000000040039a in main () at t.c:5
14550 Another possibility is an ambiguous virtual tail call frames resolution:
14554 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
14555 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
14556 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
14557 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
14558 static void __attribute__((noinline, noclone)) b (void)
14559 @{ if (i) c (); else e (); @}
14560 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
14561 int main (void) @{ a (); return 0; @}
14563 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
14564 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
14565 tailcall: reduced: 0x4004d2(a) |
14568 #1 0x00000000004004d2 in a () at t.c:8
14569 #2 0x0000000000400395 in main () at t.c:9
14572 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
14573 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
14575 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
14576 @ifset HAVE_MAKEINFO_CLICK
14577 @set ARROW @click{}
14578 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
14579 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
14581 @ifclear HAVE_MAKEINFO_CLICK
14583 @set CALLSEQ1B @value{CALLSEQ1A}
14584 @set CALLSEQ2B @value{CALLSEQ2A}
14587 Frames #0 and #2 are real, #1 is a virtual tail call frame.
14588 The code can have possible execution paths @value{CALLSEQ1B} or
14589 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
14591 @code{initial:} state shows some random possible calling sequence @value{GDBN}
14592 has found. It then finds another possible calling sequence - that one is
14593 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
14594 printed as the @code{reduced:} calling sequence. That one could have many
14595 further @code{compare:} and @code{reduced:} statements as long as there remain
14596 any non-ambiguous sequence entries.
14598 For the frame of function @code{b} in both cases there are different possible
14599 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
14600 also ambiguous. The only non-ambiguous frame is the one for function @code{a},
14601 therefore this one is displayed to the user while the ambiguous frames are
14604 There can be also reasons why printing of frame argument values at function
14609 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
14610 static void __attribute__((noinline, noclone)) a (int i);
14611 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
14612 static void __attribute__((noinline, noclone)) a (int i)
14613 @{ if (i) b (i - 1); else c (0); @}
14614 int main (void) @{ a (5); return 0; @}
14617 #0 c (i=i@@entry=0) at t.c:2
14618 #1 0x0000000000400428 in a (DW_OP_entry_value resolving has found
14619 function "a" at 0x400420 can call itself via tail calls
14620 i=<optimized out>) at t.c:6
14621 #2 0x000000000040036e in main () at t.c:7
14624 @value{GDBN} cannot find out from the inferior state if and how many times did
14625 function @code{a} call itself (via function @code{b}) as these calls would be
14626 tail calls. Such tail calls would modify the @code{i} variable, therefore
14627 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
14628 prints @code{<optimized out>} instead.
14631 @chapter C Preprocessor Macros
14633 Some languages, such as C and C@t{++}, provide a way to define and invoke
14634 ``preprocessor macros'' which expand into strings of tokens.
14635 @value{GDBN} can evaluate expressions containing macro invocations, show
14636 the result of macro expansion, and show a macro's definition, including
14637 where it was defined.
14639 You may need to compile your program specially to provide @value{GDBN}
14640 with information about preprocessor macros. Most compilers do not
14641 include macros in their debugging information, even when you compile
14642 with the @option{-g} flag. @xref{Compilation}.
14644 A program may define a macro at one point, remove that definition later,
14645 and then provide a different definition after that. Thus, at different
14646 points in the program, a macro may have different definitions, or have
14647 no definition at all. If there is a current stack frame, @value{GDBN}
14648 uses the macros in scope at that frame's source code line. Otherwise,
14649 @value{GDBN} uses the macros in scope at the current listing location;
14652 Whenever @value{GDBN} evaluates an expression, it always expands any
14653 macro invocations present in the expression. @value{GDBN} also provides
14654 the following commands for working with macros explicitly.
14658 @kindex macro expand
14659 @cindex macro expansion, showing the results of preprocessor
14660 @cindex preprocessor macro expansion, showing the results of
14661 @cindex expanding preprocessor macros
14662 @item macro expand @var{expression}
14663 @itemx macro exp @var{expression}
14664 Show the results of expanding all preprocessor macro invocations in
14665 @var{expression}. Since @value{GDBN} simply expands macros, but does
14666 not parse the result, @var{expression} need not be a valid expression;
14667 it can be any string of tokens.
14670 @item macro expand-once @var{expression}
14671 @itemx macro exp1 @var{expression}
14672 @cindex expand macro once
14673 @i{(This command is not yet implemented.)} Show the results of
14674 expanding those preprocessor macro invocations that appear explicitly in
14675 @var{expression}. Macro invocations appearing in that expansion are
14676 left unchanged. This command allows you to see the effect of a
14677 particular macro more clearly, without being confused by further
14678 expansions. Since @value{GDBN} simply expands macros, but does not
14679 parse the result, @var{expression} need not be a valid expression; it
14680 can be any string of tokens.
14683 @cindex macro definition, showing
14684 @cindex definition of a macro, showing
14685 @cindex macros, from debug info
14686 @item info macro [-a|-all] [--] @var{macro}
14687 Show the current definition or all definitions of the named @var{macro},
14688 and describe the source location or compiler command-line where that
14689 definition was established. The optional double dash is to signify the end of
14690 argument processing and the beginning of @var{macro} for non C-like macros where
14691 the macro may begin with a hyphen.
14693 @kindex info macros
14694 @item info macros @var{locspec}
14695 Show all macro definitions that are in effect at the source line of
14696 the code location that results from resolving @var{locspec}, and
14697 describe the source location or compiler command-line where those
14698 definitions were established.
14700 @kindex macro define
14701 @cindex user-defined macros
14702 @cindex defining macros interactively
14703 @cindex macros, user-defined
14704 @item macro define @var{macro} @var{replacement-list}
14705 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
14706 Introduce a definition for a preprocessor macro named @var{macro},
14707 invocations of which are replaced by the tokens given in
14708 @var{replacement-list}. The first form of this command defines an
14709 ``object-like'' macro, which takes no arguments; the second form
14710 defines a ``function-like'' macro, which takes the arguments given in
14713 A definition introduced by this command is in scope in every
14714 expression evaluated in @value{GDBN}, until it is removed with the
14715 @code{macro undef} command, described below. The definition overrides
14716 all definitions for @var{macro} present in the program being debugged,
14717 as well as any previous user-supplied definition.
14719 @kindex macro undef
14720 @item macro undef @var{macro}
14721 Remove any user-supplied definition for the macro named @var{macro}.
14722 This command only affects definitions provided with the @code{macro
14723 define} command, described above; it cannot remove definitions present
14724 in the program being debugged.
14728 List all the macros defined using the @code{macro define} command.
14731 @cindex macros, example of debugging with
14732 Here is a transcript showing the above commands in action. First, we
14733 show our source files:
14738 #include "sample.h"
14741 #define ADD(x) (M + x)
14746 printf ("Hello, world!\n");
14748 printf ("We're so creative.\n");
14750 printf ("Goodbye, world!\n");
14757 Now, we compile the program using the @sc{gnu} C compiler,
14758 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
14759 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
14760 and @option{-gdwarf-4}; we recommend always choosing the most recent
14761 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
14762 includes information about preprocessor macros in the debugging
14766 $ gcc -gdwarf-2 -g3 sample.c -o sample
14770 Now, we start @value{GDBN} on our sample program:
14774 GNU gdb 2002-05-06-cvs
14775 Copyright 2002 Free Software Foundation, Inc.
14776 GDB is free software, @dots{}
14780 We can expand macros and examine their definitions, even when the
14781 program is not running. @value{GDBN} uses the current listing position
14782 to decide which macro definitions are in scope:
14785 (@value{GDBP}) list main
14788 5 #define ADD(x) (M + x)
14793 10 printf ("Hello, world!\n");
14795 12 printf ("We're so creative.\n");
14796 (@value{GDBP}) info macro ADD
14797 Defined at /home/jimb/gdb/macros/play/sample.c:5
14798 #define ADD(x) (M + x)
14799 (@value{GDBP}) info macro Q
14800 Defined at /home/jimb/gdb/macros/play/sample.h:1
14801 included at /home/jimb/gdb/macros/play/sample.c:2
14803 (@value{GDBP}) macro expand ADD(1)
14804 expands to: (42 + 1)
14805 (@value{GDBP}) macro expand-once ADD(1)
14806 expands to: once (M + 1)
14810 In the example above, note that @code{macro expand-once} expands only
14811 the macro invocation explicit in the original text --- the invocation of
14812 @code{ADD} --- but does not expand the invocation of the macro @code{M},
14813 which was introduced by @code{ADD}.
14815 Once the program is running, @value{GDBN} uses the macro definitions in
14816 force at the source line of the current stack frame:
14819 (@value{GDBP}) break main
14820 Breakpoint 1 at 0x8048370: file sample.c, line 10.
14822 Starting program: /home/jimb/gdb/macros/play/sample
14824 Breakpoint 1, main () at sample.c:10
14825 10 printf ("Hello, world!\n");
14829 At line 10, the definition of the macro @code{N} at line 9 is in force:
14832 (@value{GDBP}) info macro N
14833 Defined at /home/jimb/gdb/macros/play/sample.c:9
14835 (@value{GDBP}) macro expand N Q M
14836 expands to: 28 < 42
14837 (@value{GDBP}) print N Q M
14842 As we step over directives that remove @code{N}'s definition, and then
14843 give it a new definition, @value{GDBN} finds the definition (or lack
14844 thereof) in force at each point:
14847 (@value{GDBP}) next
14849 12 printf ("We're so creative.\n");
14850 (@value{GDBP}) info macro N
14851 The symbol `N' has no definition as a C/C++ preprocessor macro
14852 at /home/jimb/gdb/macros/play/sample.c:12
14853 (@value{GDBP}) next
14855 14 printf ("Goodbye, world!\n");
14856 (@value{GDBP}) info macro N
14857 Defined at /home/jimb/gdb/macros/play/sample.c:13
14859 (@value{GDBP}) macro expand N Q M
14860 expands to: 1729 < 42
14861 (@value{GDBP}) print N Q M
14866 In addition to source files, macros can be defined on the compilation command
14867 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
14868 such a way, @value{GDBN} displays the location of their definition as line zero
14869 of the source file submitted to the compiler.
14872 (@value{GDBP}) info macro __STDC__
14873 Defined at /home/jimb/gdb/macros/play/sample.c:0
14880 @chapter Tracepoints
14881 @c This chapter is based on the documentation written by Michael
14882 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
14884 @cindex tracepoints
14885 In some applications, it is not feasible for the debugger to interrupt
14886 the program's execution long enough for the developer to learn
14887 anything helpful about its behavior. If the program's correctness
14888 depends on its real-time behavior, delays introduced by a debugger
14889 might cause the program to change its behavior drastically, or perhaps
14890 fail, even when the code itself is correct. It is useful to be able
14891 to observe the program's behavior without interrupting it.
14893 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
14894 specify locations in the program, called @dfn{tracepoints}, and
14895 arbitrary expressions to evaluate when those tracepoints are reached.
14896 Later, using the @code{tfind} command, you can examine the values
14897 those expressions had when the program hit the tracepoints. The
14898 expressions may also denote objects in memory---structures or arrays,
14899 for example---whose values @value{GDBN} should record; while visiting
14900 a particular tracepoint, you may inspect those objects as if they were
14901 in memory at that moment. However, because @value{GDBN} records these
14902 values without interacting with you, it can do so quickly and
14903 unobtrusively, hopefully not disturbing the program's behavior.
14905 The tracepoint facility is currently available only for remote
14906 targets. @xref{Targets}. In addition, your remote target must know
14907 how to collect trace data. This functionality is implemented in the
14908 remote stub; however, none of the stubs distributed with @value{GDBN}
14909 support tracepoints as of this writing. The format of the remote
14910 packets used to implement tracepoints are described in @ref{Tracepoint
14913 It is also possible to get trace data from a file, in a manner reminiscent
14914 of corefiles; you specify the filename, and use @code{tfind} to search
14915 through the file. @xref{Trace Files}, for more details.
14917 This chapter describes the tracepoint commands and features.
14920 * Set Tracepoints::
14921 * Analyze Collected Data::
14922 * Tracepoint Variables::
14926 @node Set Tracepoints
14927 @section Commands to Set Tracepoints
14929 Before running such a @dfn{trace experiment}, an arbitrary number of
14930 tracepoints can be set. A tracepoint is actually a special type of
14931 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
14932 standard breakpoint commands. For instance, as with breakpoints,
14933 tracepoint numbers are successive integers starting from one, and many
14934 of the commands associated with tracepoints take the tracepoint number
14935 as their argument, to identify which tracepoint to work on.
14937 For each tracepoint, you can specify, in advance, some arbitrary set
14938 of data that you want the target to collect in the trace buffer when
14939 it hits that tracepoint. The collected data can include registers,
14940 local variables, or global data. Later, you can use @value{GDBN}
14941 commands to examine the values these data had at the time the
14942 tracepoint was hit.
14944 Tracepoints do not support every breakpoint feature. Ignore counts on
14945 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
14946 commands when they are hit. Tracepoints may not be thread-specific
14949 @cindex fast tracepoints
14950 Some targets may support @dfn{fast tracepoints}, which are inserted in
14951 a different way (such as with a jump instead of a trap), that is
14952 faster but possibly restricted in where they may be installed.
14954 @cindex static tracepoints
14955 @cindex markers, static tracepoints
14956 @cindex probing markers, static tracepoints
14957 Regular and fast tracepoints are dynamic tracing facilities, meaning
14958 that they can be used to insert tracepoints at (almost) any location
14959 in the target. Some targets may also support controlling @dfn{static
14960 tracepoints} from @value{GDBN}. With static tracing, a set of
14961 instrumentation points, also known as @dfn{markers}, are embedded in
14962 the target program, and can be activated or deactivated by name or
14963 address. These are usually placed at locations which facilitate
14964 investigating what the target is actually doing. @value{GDBN}'s
14965 support for static tracing includes being able to list instrumentation
14966 points, and attach them with @value{GDBN} defined high level
14967 tracepoints that expose the whole range of convenience of
14968 @value{GDBN}'s tracepoints support. Namely, support for collecting
14969 registers values and values of global or local (to the instrumentation
14970 point) variables; tracepoint conditions and trace state variables.
14971 The act of installing a @value{GDBN} static tracepoint on an
14972 instrumentation point, or marker, is referred to as @dfn{probing} a
14973 static tracepoint marker.
14975 @code{gdbserver} supports tracepoints on some target systems.
14976 @xref{Server,,Tracepoints support in @code{gdbserver}}.
14978 This section describes commands to set tracepoints and associated
14979 conditions and actions.
14982 * Create and Delete Tracepoints::
14983 * Enable and Disable Tracepoints::
14984 * Tracepoint Passcounts::
14985 * Tracepoint Conditions::
14986 * Trace State Variables::
14987 * Tracepoint Actions::
14988 * Listing Tracepoints::
14989 * Listing Static Tracepoint Markers::
14990 * Starting and Stopping Trace Experiments::
14991 * Tracepoint Restrictions::
14994 @node Create and Delete Tracepoints
14995 @subsection Create and Delete Tracepoints
14998 @cindex set tracepoint
15000 @item trace @var{locspec}
15001 The @code{trace} command is very similar to the @code{break} command.
15002 Its argument @var{locspec} can be any valid location specification.
15003 @xref{Location Specifications}. The @code{trace} command defines a tracepoint,
15004 which is a point in the target program where the debugger will briefly stop,
15005 collect some data, and then allow the program to continue. Setting a tracepoint
15006 or changing its actions takes effect immediately if the remote stub
15007 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
15009 If remote stub doesn't support the @samp{InstallInTrace} feature, all
15010 these changes don't take effect until the next @code{tstart}
15011 command, and once a trace experiment is running, further changes will
15012 not have any effect until the next trace experiment starts. In addition,
15013 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
15014 address is not yet resolved. (This is similar to pending breakpoints.)
15015 Pending tracepoints are not downloaded to the target and not installed
15016 until they are resolved. The resolution of pending tracepoints requires
15017 @value{GDBN} support---when debugging with the remote target, and
15018 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
15019 tracing}), pending tracepoints can not be resolved (and downloaded to
15020 the remote stub) while @value{GDBN} is disconnected.
15022 Here are some examples of using the @code{trace} command:
15025 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
15027 (@value{GDBP}) @b{trace +2} // 2 lines forward
15029 (@value{GDBP}) @b{trace my_function} // first source line of function
15031 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
15033 (@value{GDBP}) @b{trace *0x2117c4} // an address
15037 You can abbreviate @code{trace} as @code{tr}.
15039 @item trace @var{locspec} if @var{cond}
15040 Set a tracepoint with condition @var{cond}; evaluate the expression
15041 @var{cond} each time the tracepoint is reached, and collect data only
15042 if the value is nonzero---that is, if @var{cond} evaluates as true.
15043 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
15044 information on tracepoint conditions.
15046 @item ftrace @var{locspec} [ if @var{cond} ]
15047 @cindex set fast tracepoint
15048 @cindex fast tracepoints, setting
15050 The @code{ftrace} command sets a fast tracepoint. For targets that
15051 support them, fast tracepoints will use a more efficient but possibly
15052 less general technique to trigger data collection, such as a jump
15053 instruction instead of a trap, or some sort of hardware support. It
15054 may not be possible to create a fast tracepoint at the desired
15055 location, in which case the command will exit with an explanatory
15058 @value{GDBN} handles arguments to @code{ftrace} exactly as for
15061 On 32-bit x86-architecture systems, fast tracepoints normally need to
15062 be placed at an instruction that is 5 bytes or longer, but can be
15063 placed at 4-byte instructions if the low 64K of memory of the target
15064 program is available to install trampolines. Some Unix-type systems,
15065 such as @sc{gnu}/Linux, exclude low addresses from the program's
15066 address space; but for instance with the Linux kernel it is possible
15067 to let @value{GDBN} use this area by doing a @command{sysctl} command
15068 to set the @code{mmap_min_addr} kernel parameter, as in
15071 sudo sysctl -w vm.mmap_min_addr=32768
15075 which sets the low address to 32K, which leaves plenty of room for
15076 trampolines. The minimum address should be set to a page boundary.
15078 @item strace [@var{locspec} | -m @var{marker}] [ if @var{cond} ]
15079 @cindex set static tracepoint
15080 @cindex static tracepoints, setting
15081 @cindex probe static tracepoint marker
15083 The @code{strace} command sets a static tracepoint. For targets that
15084 support it, setting a static tracepoint probes a static
15085 instrumentation point, or marker, found at the code locations that
15086 result from resolving @var{locspec}. It may not be possible to set a
15087 static tracepoint at the desired code location, in which case the
15088 command will exit with an explanatory message.
15090 @value{GDBN} handles arguments to @code{strace} exactly as for
15091 @code{trace}, with the addition that the user can also specify
15092 @code{-m @var{marker}} instead of a location spec. This probes the marker
15093 identified by the @var{marker} string identifier. This identifier
15094 depends on the static tracepoint backend library your program is
15095 using. You can find all the marker identifiers in the @samp{ID} field
15096 of the @code{info static-tracepoint-markers} command output.
15097 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
15098 Markers}. For example, in the following small program using the UST
15104 trace_mark(ust, bar33, "str %s", "FOOBAZ");
15109 the marker id is composed of joining the first two arguments to the
15110 @code{trace_mark} call with a slash, which translates to:
15113 (@value{GDBP}) info static-tracepoint-markers
15114 Cnt Enb ID Address What
15115 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
15121 so you may probe the marker above with:
15124 (@value{GDBP}) strace -m ust/bar33
15127 Static tracepoints accept an extra collect action --- @code{collect
15128 $_sdata}. This collects arbitrary user data passed in the probe point
15129 call to the tracing library. In the UST example above, you'll see
15130 that the third argument to @code{trace_mark} is a printf-like format
15131 string. The user data is then the result of running that formatting
15132 string against the following arguments. Note that @code{info
15133 static-tracepoint-markers} command output lists that format string in
15134 the @samp{Data:} field.
15136 You can inspect this data when analyzing the trace buffer, by printing
15137 the $_sdata variable like any other variable available to
15138 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
15141 @cindex last tracepoint number
15142 @cindex recent tracepoint number
15143 @cindex tracepoint number
15144 The convenience variable @code{$tpnum} records the tracepoint number
15145 of the most recently set tracepoint.
15147 @kindex delete tracepoint
15148 @cindex tracepoint deletion
15149 @item delete tracepoint @r{[}@var{num}@r{]}
15150 Permanently delete one or more tracepoints. With no argument, the
15151 default is to delete all tracepoints. Note that the regular
15152 @code{delete} command can remove tracepoints also.
15157 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
15159 (@value{GDBP}) @b{delete trace} // remove all tracepoints
15163 You can abbreviate this command as @code{del tr}.
15166 @node Enable and Disable Tracepoints
15167 @subsection Enable and Disable Tracepoints
15169 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
15172 @kindex disable tracepoint
15173 @item disable tracepoint @r{[}@var{num}@r{]}
15174 Disable tracepoint @var{num}, or all tracepoints if no argument
15175 @var{num} is given. A disabled tracepoint will have no effect during
15176 a trace experiment, but it is not forgotten. You can re-enable
15177 a disabled tracepoint using the @code{enable tracepoint} command.
15178 If the command is issued during a trace experiment and the debug target
15179 has support for disabling tracepoints during a trace experiment, then the
15180 change will be effective immediately. Otherwise, it will be applied to the
15181 next trace experiment.
15183 @kindex enable tracepoint
15184 @item enable tracepoint @r{[}@var{num}@r{]}
15185 Enable tracepoint @var{num}, or all tracepoints. If this command is
15186 issued during a trace experiment and the debug target supports enabling
15187 tracepoints during a trace experiment, then the enabled tracepoints will
15188 become effective immediately. Otherwise, they will become effective the
15189 next time a trace experiment is run.
15192 @node Tracepoint Passcounts
15193 @subsection Tracepoint Passcounts
15197 @cindex tracepoint pass count
15198 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
15199 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
15200 automatically stop a trace experiment. If a tracepoint's passcount is
15201 @var{n}, then the trace experiment will be automatically stopped on
15202 the @var{n}'th time that tracepoint is hit. If the tracepoint number
15203 @var{num} is not specified, the @code{passcount} command sets the
15204 passcount of the most recently defined tracepoint. If no passcount is
15205 given, the trace experiment will run until stopped explicitly by the
15211 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
15212 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
15214 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
15215 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
15216 (@value{GDBP}) @b{trace foo}
15217 (@value{GDBP}) @b{pass 3}
15218 (@value{GDBP}) @b{trace bar}
15219 (@value{GDBP}) @b{pass 2}
15220 (@value{GDBP}) @b{trace baz}
15221 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
15222 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
15223 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
15224 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
15228 @node Tracepoint Conditions
15229 @subsection Tracepoint Conditions
15230 @cindex conditional tracepoints
15231 @cindex tracepoint conditions
15233 The simplest sort of tracepoint collects data every time your program
15234 reaches a specified place. You can also specify a @dfn{condition} for
15235 a tracepoint. A condition is just a Boolean expression in your
15236 programming language (@pxref{Expressions, ,Expressions}). A
15237 tracepoint with a condition evaluates the expression each time your
15238 program reaches it, and data collection happens only if the condition
15241 Tracepoint conditions can be specified when a tracepoint is set, by
15242 using @samp{if} in the arguments to the @code{trace} command.
15243 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
15244 also be set or changed at any time with the @code{condition} command,
15245 just as with breakpoints.
15247 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
15248 the conditional expression itself. Instead, @value{GDBN} encodes the
15249 expression into an agent expression (@pxref{Agent Expressions})
15250 suitable for execution on the target, independently of @value{GDBN}.
15251 Global variables become raw memory locations, locals become stack
15252 accesses, and so forth.
15254 For instance, suppose you have a function that is usually called
15255 frequently, but should not be called after an error has occurred. You
15256 could use the following tracepoint command to collect data about calls
15257 of that function that happen while the error code is propagating
15258 through the program; an unconditional tracepoint could end up
15259 collecting thousands of useless trace frames that you would have to
15263 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
15266 @node Trace State Variables
15267 @subsection Trace State Variables
15268 @cindex trace state variables
15270 A @dfn{trace state variable} is a special type of variable that is
15271 created and managed by target-side code. The syntax is the same as
15272 that for GDB's convenience variables (a string prefixed with ``$''),
15273 but they are stored on the target. They must be created explicitly,
15274 using a @code{tvariable} command. They are always 64-bit signed
15277 Trace state variables are remembered by @value{GDBN}, and downloaded
15278 to the target along with tracepoint information when the trace
15279 experiment starts. There are no intrinsic limits on the number of
15280 trace state variables, beyond memory limitations of the target.
15282 @cindex convenience variables, and trace state variables
15283 Although trace state variables are managed by the target, you can use
15284 them in print commands and expressions as if they were convenience
15285 variables; @value{GDBN} will get the current value from the target
15286 while the trace experiment is running. Trace state variables share
15287 the same namespace as other ``$'' variables, which means that you
15288 cannot have trace state variables with names like @code{$23} or
15289 @code{$pc}, nor can you have a trace state variable and a convenience
15290 variable with the same name.
15294 @item tvariable $@var{name} [ = @var{expression} ]
15296 The @code{tvariable} command creates a new trace state variable named
15297 @code{$@var{name}}, and optionally gives it an initial value of
15298 @var{expression}. The @var{expression} is evaluated when this command is
15299 entered; the result will be converted to an integer if possible,
15300 otherwise @value{GDBN} will report an error. A subsequent
15301 @code{tvariable} command specifying the same name does not create a
15302 variable, but instead assigns the supplied initial value to the
15303 existing variable of that name, overwriting any previous initial
15304 value. The default initial value is 0.
15306 @item info tvariables
15307 @kindex info tvariables
15308 List all the trace state variables along with their initial values.
15309 Their current values may also be displayed, if the trace experiment is
15312 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
15313 @kindex delete tvariable
15314 Delete the given trace state variables, or all of them if no arguments
15319 @node Tracepoint Actions
15320 @subsection Tracepoint Action Lists
15324 @cindex tracepoint actions
15325 @item actions @r{[}@var{num}@r{]}
15326 This command will prompt for a list of actions to be taken when the
15327 tracepoint is hit. If the tracepoint number @var{num} is not
15328 specified, this command sets the actions for the one that was most
15329 recently defined (so that you can define a tracepoint and then say
15330 @code{actions} without bothering about its number). You specify the
15331 actions themselves on the following lines, one action at a time, and
15332 terminate the actions list with a line containing just @code{end}. So
15333 far, the only defined actions are @code{collect}, @code{teval}, and
15334 @code{while-stepping}.
15336 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
15337 Commands, ,Breakpoint Command Lists}), except that only the defined
15338 actions are allowed; any other @value{GDBN} command is rejected.
15340 @cindex remove actions from a tracepoint
15341 To remove all actions from a tracepoint, type @samp{actions @var{num}}
15342 and follow it immediately with @samp{end}.
15345 (@value{GDBP}) @b{collect @var{data}} // collect some data
15347 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
15349 (@value{GDBP}) @b{end} // signals the end of actions.
15352 In the following example, the action list begins with @code{collect}
15353 commands indicating the things to be collected when the tracepoint is
15354 hit. Then, in order to single-step and collect additional data
15355 following the tracepoint, a @code{while-stepping} command is used,
15356 followed by the list of things to be collected after each step in a
15357 sequence of single steps. The @code{while-stepping} command is
15358 terminated by its own separate @code{end} command. Lastly, the action
15359 list is terminated by an @code{end} command.
15362 (@value{GDBP}) @b{trace foo}
15363 (@value{GDBP}) @b{actions}
15364 Enter actions for tracepoint 1, one per line:
15367 > while-stepping 12
15368 > collect $pc, arr[i]
15373 @kindex collect @r{(tracepoints)}
15374 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
15375 Collect values of the given expressions when the tracepoint is hit.
15376 This command accepts a comma-separated list of any valid expressions.
15377 In addition to global, static, or local variables, the following
15378 special arguments are supported:
15382 Collect all registers.
15385 Collect all function arguments.
15388 Collect all local variables.
15391 Collect the return address. This is helpful if you want to see more
15394 @emph{Note:} The return address location can not always be reliably
15395 determined up front, and the wrong address / registers may end up
15396 collected instead. On some architectures the reliability is higher
15397 for tracepoints at function entry, while on others it's the opposite.
15398 When this happens, backtracing will stop because the return address is
15399 found unavailable (unless another collect rule happened to match it).
15402 Collects the number of arguments from the static probe at which the
15403 tracepoint is located.
15404 @xref{Static Probe Points}.
15406 @item $_probe_arg@var{n}
15407 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
15408 from the static probe at which the tracepoint is located.
15409 @xref{Static Probe Points}.
15412 @vindex $_sdata@r{, collect}
15413 Collect static tracepoint marker specific data. Only available for
15414 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
15415 Lists}. On the UST static tracepoints library backend, an
15416 instrumentation point resembles a @code{printf} function call. The
15417 tracing library is able to collect user specified data formatted to a
15418 character string using the format provided by the programmer that
15419 instrumented the program. Other backends have similar mechanisms.
15420 Here's an example of a UST marker call:
15423 const char master_name[] = "$your_name";
15424 trace_mark(channel1, marker1, "hello %s", master_name)
15427 In this case, collecting @code{$_sdata} collects the string
15428 @samp{hello $yourname}. When analyzing the trace buffer, you can
15429 inspect @samp{$_sdata} like any other variable available to
15433 You can give several consecutive @code{collect} commands, each one
15434 with a single argument, or one @code{collect} command with several
15435 arguments separated by commas; the effect is the same.
15437 The optional @var{mods} changes the usual handling of the arguments.
15438 @code{s} requests that pointers to chars be handled as strings, in
15439 particular collecting the contents of the memory being pointed at, up
15440 to the first zero. The upper bound is by default the value of the
15441 @code{print characters} variable; if @code{s} is followed by a decimal
15442 number, that is the upper bound instead. So for instance
15443 @samp{collect/s25 mystr} collects as many as 25 characters at
15446 The command @code{info scope} (@pxref{Symbols, info scope}) is
15447 particularly useful for figuring out what data to collect.
15449 @kindex teval @r{(tracepoints)}
15450 @item teval @var{expr1}, @var{expr2}, @dots{}
15451 Evaluate the given expressions when the tracepoint is hit. This
15452 command accepts a comma-separated list of expressions. The results
15453 are discarded, so this is mainly useful for assigning values to trace
15454 state variables (@pxref{Trace State Variables}) without adding those
15455 values to the trace buffer, as would be the case if the @code{collect}
15458 @kindex while-stepping @r{(tracepoints)}
15459 @item while-stepping @var{n}
15460 Perform @var{n} single-step instruction traces after the tracepoint,
15461 collecting new data after each step. The @code{while-stepping}
15462 command is followed by the list of what to collect while stepping
15463 (followed by its own @code{end} command):
15466 > while-stepping 12
15467 > collect $regs, myglobal
15473 Note that @code{$pc} is not automatically collected by
15474 @code{while-stepping}; you need to explicitly collect that register if
15475 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
15478 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
15479 @kindex set default-collect
15480 @cindex default collection action
15481 This variable is a list of expressions to collect at each tracepoint
15482 hit. It is effectively an additional @code{collect} action prepended
15483 to every tracepoint action list. The expressions are parsed
15484 individually for each tracepoint, so for instance a variable named
15485 @code{xyz} may be interpreted as a global for one tracepoint, and a
15486 local for another, as appropriate to the tracepoint's location.
15488 @item show default-collect
15489 @kindex show default-collect
15490 Show the list of expressions that are collected by default at each
15495 @node Listing Tracepoints
15496 @subsection Listing Tracepoints
15499 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
15500 @kindex info tp @r{[}@var{n}@dots{}@r{]}
15501 @cindex information about tracepoints
15502 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
15503 Display information about the tracepoint @var{num}. If you don't
15504 specify a tracepoint number, displays information about all the
15505 tracepoints defined so far. The format is similar to that used for
15506 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
15507 command, simply restricting itself to tracepoints.
15509 A tracepoint's listing may include additional information specific to
15514 its passcount as given by the @code{passcount @var{n}} command
15517 the state about installed on target of each location
15521 (@value{GDBP}) @b{info trace}
15522 Num Type Disp Enb Address What
15523 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
15525 collect globfoo, $regs
15530 2 tracepoint keep y <MULTIPLE>
15532 2.1 y 0x0804859c in func4 at change-loc.h:35
15533 installed on target
15534 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
15535 installed on target
15536 2.3 y <PENDING> set_tracepoint
15537 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
15538 not installed on target
15543 This command can be abbreviated @code{info tp}.
15546 @node Listing Static Tracepoint Markers
15547 @subsection Listing Static Tracepoint Markers
15550 @kindex info static-tracepoint-markers
15551 @cindex information about static tracepoint markers
15552 @item info static-tracepoint-markers
15553 Display information about all static tracepoint markers defined in the
15556 For each marker, the following columns are printed:
15560 An incrementing counter, output to help readability. This is not a
15563 The marker ID, as reported by the target.
15564 @item Enabled or Disabled
15565 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
15566 that are not enabled.
15568 Where the marker is in your program, as a memory address.
15570 Where the marker is in the source for your program, as a file and line
15571 number. If the debug information included in the program does not
15572 allow @value{GDBN} to locate the source of the marker, this column
15573 will be left blank.
15577 In addition, the following information may be printed for each marker:
15581 User data passed to the tracing library by the marker call. In the
15582 UST backend, this is the format string passed as argument to the
15584 @item Static tracepoints probing the marker
15585 The list of static tracepoints attached to the marker.
15589 (@value{GDBP}) info static-tracepoint-markers
15590 Cnt ID Enb Address What
15591 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
15592 Data: number1 %d number2 %d
15593 Probed by static tracepoints: #2
15594 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
15600 @node Starting and Stopping Trace Experiments
15601 @subsection Starting and Stopping Trace Experiments
15604 @kindex tstart [ @var{notes} ]
15605 @cindex start a new trace experiment
15606 @cindex collected data discarded
15608 This command starts the trace experiment, and begins collecting data.
15609 It has the side effect of discarding all the data collected in the
15610 trace buffer during the previous trace experiment. If any arguments
15611 are supplied, they are taken as a note and stored with the trace
15612 experiment's state. The notes may be arbitrary text, and are
15613 especially useful with disconnected tracing in a multi-user context;
15614 the notes can explain what the trace is doing, supply user contact
15615 information, and so forth.
15617 @kindex tstop [ @var{notes} ]
15618 @cindex stop a running trace experiment
15620 This command stops the trace experiment. If any arguments are
15621 supplied, they are recorded with the experiment as a note. This is
15622 useful if you are stopping a trace started by someone else, for
15623 instance if the trace is interfering with the system's behavior and
15624 needs to be stopped quickly.
15626 @strong{Note}: a trace experiment and data collection may stop
15627 automatically if any tracepoint's passcount is reached
15628 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
15631 @cindex status of trace data collection
15632 @cindex trace experiment, status of
15634 This command displays the status of the current trace data
15638 Here is an example of the commands we described so far:
15641 (@value{GDBP}) @b{trace gdb_c_test}
15642 (@value{GDBP}) @b{actions}
15643 Enter actions for tracepoint #1, one per line.
15644 > collect $regs,$locals,$args
15645 > while-stepping 11
15649 (@value{GDBP}) @b{tstart}
15650 [time passes @dots{}]
15651 (@value{GDBP}) @b{tstop}
15654 @anchor{disconnected tracing}
15655 @cindex disconnected tracing
15656 You can choose to continue running the trace experiment even if
15657 @value{GDBN} disconnects from the target, voluntarily or
15658 involuntarily. For commands such as @code{detach}, the debugger will
15659 ask what you want to do with the trace. But for unexpected
15660 terminations (@value{GDBN} crash, network outage), it would be
15661 unfortunate to lose hard-won trace data, so the variable
15662 @code{disconnected-tracing} lets you decide whether the trace should
15663 continue running without @value{GDBN}.
15666 @item set disconnected-tracing on
15667 @itemx set disconnected-tracing off
15668 @kindex set disconnected-tracing
15669 Choose whether a tracing run should continue to run if @value{GDBN}
15670 has disconnected from the target. Note that @code{detach} or
15671 @code{quit} will ask you directly what to do about a running trace no
15672 matter what this variable's setting, so the variable is mainly useful
15673 for handling unexpected situations, such as loss of the network.
15675 @item show disconnected-tracing
15676 @kindex show disconnected-tracing
15677 Show the current choice for disconnected tracing.
15681 When you reconnect to the target, the trace experiment may or may not
15682 still be running; it might have filled the trace buffer in the
15683 meantime, or stopped for one of the other reasons. If it is running,
15684 it will continue after reconnection.
15686 Upon reconnection, the target will upload information about the
15687 tracepoints in effect. @value{GDBN} will then compare that
15688 information to the set of tracepoints currently defined, and attempt
15689 to match them up, allowing for the possibility that the numbers may
15690 have changed due to creation and deletion in the meantime. If one of
15691 the target's tracepoints does not match any in @value{GDBN}, the
15692 debugger will create a new tracepoint, so that you have a number with
15693 which to specify that tracepoint. This matching-up process is
15694 necessarily heuristic, and it may result in useless tracepoints being
15695 created; you may simply delete them if they are of no use.
15697 @cindex circular trace buffer
15698 If your target agent supports a @dfn{circular trace buffer}, then you
15699 can run a trace experiment indefinitely without filling the trace
15700 buffer; when space runs out, the agent deletes already-collected trace
15701 frames, oldest first, until there is enough room to continue
15702 collecting. This is especially useful if your tracepoints are being
15703 hit too often, and your trace gets terminated prematurely because the
15704 buffer is full. To ask for a circular trace buffer, simply set
15705 @samp{circular-trace-buffer} to on. You can set this at any time,
15706 including during tracing; if the agent can do it, it will change
15707 buffer handling on the fly, otherwise it will not take effect until
15711 @item set circular-trace-buffer on
15712 @itemx set circular-trace-buffer off
15713 @kindex set circular-trace-buffer
15714 Choose whether a tracing run should use a linear or circular buffer
15715 for trace data. A linear buffer will not lose any trace data, but may
15716 fill up prematurely, while a circular buffer will discard old trace
15717 data, but it will have always room for the latest tracepoint hits.
15719 @item show circular-trace-buffer
15720 @kindex show circular-trace-buffer
15721 Show the current choice for the trace buffer. Note that this may not
15722 match the agent's current buffer handling, nor is it guaranteed to
15723 match the setting that might have been in effect during a past run,
15724 for instance if you are looking at frames from a trace file.
15729 @item set trace-buffer-size @var{n}
15730 @itemx set trace-buffer-size unlimited
15731 @kindex set trace-buffer-size
15732 Request that the target use a trace buffer of @var{n} bytes. Not all
15733 targets will honor the request; they may have a compiled-in size for
15734 the trace buffer, or some other limitation. Set to a value of
15735 @code{unlimited} or @code{-1} to let the target use whatever size it
15736 likes. This is also the default.
15738 @item show trace-buffer-size
15739 @kindex show trace-buffer-size
15740 Show the current requested size for the trace buffer. Note that this
15741 will only match the actual size if the target supports size-setting,
15742 and was able to handle the requested size. For instance, if the
15743 target can only change buffer size between runs, this variable will
15744 not reflect the change until the next run starts. Use @code{tstatus}
15745 to get a report of the actual buffer size.
15749 @item set trace-user @var{text}
15750 @kindex set trace-user
15752 @item show trace-user
15753 @kindex show trace-user
15755 @item set trace-notes @var{text}
15756 @kindex set trace-notes
15757 Set the trace run's notes.
15759 @item show trace-notes
15760 @kindex show trace-notes
15761 Show the trace run's notes.
15763 @item set trace-stop-notes @var{text}
15764 @kindex set trace-stop-notes
15765 Set the trace run's stop notes. The handling of the note is as for
15766 @code{tstop} arguments; the set command is convenient way to fix a
15767 stop note that is mistaken or incomplete.
15769 @item show trace-stop-notes
15770 @kindex show trace-stop-notes
15771 Show the trace run's stop notes.
15775 @node Tracepoint Restrictions
15776 @subsection Tracepoint Restrictions
15778 @cindex tracepoint restrictions
15779 There are a number of restrictions on the use of tracepoints. As
15780 described above, tracepoint data gathering occurs on the target
15781 without interaction from @value{GDBN}. Thus the full capabilities of
15782 the debugger are not available during data gathering, and then at data
15783 examination time, you will be limited by only having what was
15784 collected. The following items describe some common problems, but it
15785 is not exhaustive, and you may run into additional difficulties not
15791 Tracepoint expressions are intended to gather objects (lvalues). Thus
15792 the full flexibility of GDB's expression evaluator is not available.
15793 You cannot call functions, cast objects to aggregate types, access
15794 convenience variables or modify values (except by assignment to trace
15795 state variables). Some language features may implicitly call
15796 functions (for instance Objective-C fields with accessors), and therefore
15797 cannot be collected either.
15800 Collection of local variables, either individually or in bulk with
15801 @code{$locals} or @code{$args}, during @code{while-stepping} may
15802 behave erratically. The stepping action may enter a new scope (for
15803 instance by stepping into a function), or the location of the variable
15804 may change (for instance it is loaded into a register). The
15805 tracepoint data recorded uses the location information for the
15806 variables that is correct for the tracepoint location. When the
15807 tracepoint is created, it is not possible, in general, to determine
15808 where the steps of a @code{while-stepping} sequence will advance the
15809 program---particularly if a conditional branch is stepped.
15812 Collection of an incompletely-initialized or partially-destroyed object
15813 may result in something that @value{GDBN} cannot display, or displays
15814 in a misleading way.
15817 When @value{GDBN} displays a pointer to character it automatically
15818 dereferences the pointer to also display characters of the string
15819 being pointed to. However, collecting the pointer during tracing does
15820 not automatically collect the string. You need to explicitly
15821 dereference the pointer and provide size information if you want to
15822 collect not only the pointer, but the memory pointed to. For example,
15823 @code{*ptr@@50} can be used to collect the 50 element array pointed to
15827 It is not possible to collect a complete stack backtrace at a
15828 tracepoint. Instead, you may collect the registers and a few hundred
15829 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
15830 (adjust to use the name of the actual stack pointer register on your
15831 target architecture, and the amount of stack you wish to capture).
15832 Then the @code{backtrace} command will show a partial backtrace when
15833 using a trace frame. The number of stack frames that can be examined
15834 depends on the sizes of the frames in the collected stack. Note that
15835 if you ask for a block so large that it goes past the bottom of the
15836 stack, the target agent may report an error trying to read from an
15840 If you do not collect registers at a tracepoint, @value{GDBN} can
15841 infer that the value of @code{$pc} must be the same as the address of
15842 the tracepoint and use that when you are looking at a trace frame
15843 for that tracepoint. However, this cannot work if the tracepoint has
15844 multiple locations (for instance if it was set in a function that was
15845 inlined), or if it has a @code{while-stepping} loop. In those cases
15846 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
15851 @node Analyze Collected Data
15852 @section Using the Collected Data
15854 After the tracepoint experiment ends, you use @value{GDBN} commands
15855 for examining the trace data. The basic idea is that each tracepoint
15856 collects a trace @dfn{snapshot} every time it is hit and another
15857 snapshot every time it single-steps. All these snapshots are
15858 consecutively numbered from zero and go into a buffer, and you can
15859 examine them later. The way you examine them is to @dfn{focus} on a
15860 specific trace snapshot. When the remote stub is focused on a trace
15861 snapshot, it will respond to all @value{GDBN} requests for memory and
15862 registers by reading from the buffer which belongs to that snapshot,
15863 rather than from @emph{real} memory or registers of the program being
15864 debugged. This means that @strong{all} @value{GDBN} commands
15865 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
15866 behave as if we were currently debugging the program state as it was
15867 when the tracepoint occurred. Any requests for data that are not in
15868 the buffer will fail.
15871 * tfind:: How to select a trace snapshot
15872 * tdump:: How to display all data for a snapshot
15873 * save tracepoints:: How to save tracepoints for a future run
15877 @subsection @code{tfind @var{n}}
15880 @cindex select trace snapshot
15881 @cindex find trace snapshot
15882 The basic command for selecting a trace snapshot from the buffer is
15883 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
15884 counting from zero. If no argument @var{n} is given, the next
15885 snapshot is selected.
15887 Here are the various forms of using the @code{tfind} command.
15891 Find the first snapshot in the buffer. This is a synonym for
15892 @code{tfind 0} (since 0 is the number of the first snapshot).
15895 Stop debugging trace snapshots, resume @emph{live} debugging.
15898 Same as @samp{tfind none}.
15901 No argument means find the next trace snapshot or find the first
15902 one if no trace snapshot is selected.
15905 Find the previous trace snapshot before the current one. This permits
15906 retracing earlier steps.
15908 @item tfind tracepoint @var{num}
15909 Find the next snapshot associated with tracepoint @var{num}. Search
15910 proceeds forward from the last examined trace snapshot. If no
15911 argument @var{num} is given, it means find the next snapshot collected
15912 for the same tracepoint as the current snapshot.
15914 @item tfind pc @var{addr}
15915 Find the next snapshot associated with the value @var{addr} of the
15916 program counter. Search proceeds forward from the last examined trace
15917 snapshot. If no argument @var{addr} is given, it means find the next
15918 snapshot with the same value of PC as the current snapshot.
15920 @item tfind outside @var{addr1}, @var{addr2}
15921 Find the next snapshot whose PC is outside the given range of
15922 addresses (exclusive).
15924 @item tfind range @var{addr1}, @var{addr2}
15925 Find the next snapshot whose PC is between @var{addr1} and
15926 @var{addr2} (inclusive).
15928 @item tfind line @r{[}@var{file}:@r{]}@var{n}
15929 Find the next snapshot associated with the source line @var{n}. If
15930 the optional argument @var{file} is given, refer to line @var{n} in
15931 that source file. Search proceeds forward from the last examined
15932 trace snapshot. If no argument @var{n} is given, it means find the
15933 next line other than the one currently being examined; thus saying
15934 @code{tfind line} repeatedly can appear to have the same effect as
15935 stepping from line to line in a @emph{live} debugging session.
15938 The default arguments for the @code{tfind} commands are specifically
15939 designed to make it easy to scan through the trace buffer. For
15940 instance, @code{tfind} with no argument selects the next trace
15941 snapshot, and @code{tfind -} with no argument selects the previous
15942 trace snapshot. So, by giving one @code{tfind} command, and then
15943 simply hitting @key{RET} repeatedly you can examine all the trace
15944 snapshots in order. Or, by saying @code{tfind -} and then hitting
15945 @key{RET} repeatedly you can examine the snapshots in reverse order.
15946 The @code{tfind line} command with no argument selects the snapshot
15947 for the next source line executed. The @code{tfind pc} command with
15948 no argument selects the next snapshot with the same program counter
15949 (PC) as the current frame. The @code{tfind tracepoint} command with
15950 no argument selects the next trace snapshot collected by the same
15951 tracepoint as the current one.
15953 In addition to letting you scan through the trace buffer manually,
15954 these commands make it easy to construct @value{GDBN} scripts that
15955 scan through the trace buffer and print out whatever collected data
15956 you are interested in. Thus, if we want to examine the PC, FP, and SP
15957 registers from each trace frame in the buffer, we can say this:
15960 (@value{GDBP}) @b{tfind start}
15961 (@value{GDBP}) @b{while ($trace_frame != -1)}
15962 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
15963 $trace_frame, $pc, $sp, $fp
15967 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
15968 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
15969 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
15970 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
15971 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
15972 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
15973 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
15974 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
15975 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
15976 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
15977 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
15980 Or, if we want to examine the variable @code{X} at each source line in
15984 (@value{GDBP}) @b{tfind start}
15985 (@value{GDBP}) @b{while ($trace_frame != -1)}
15986 > printf "Frame %d, X == %d\n", $trace_frame, X
15996 @subsection @code{tdump}
15998 @cindex dump all data collected at tracepoint
15999 @cindex tracepoint data, display
16001 This command takes no arguments. It prints all the data collected at
16002 the current trace snapshot.
16005 (@value{GDBP}) @b{trace 444}
16006 (@value{GDBP}) @b{actions}
16007 Enter actions for tracepoint #2, one per line:
16008 > collect $regs, $locals, $args, gdb_long_test
16011 (@value{GDBP}) @b{tstart}
16013 (@value{GDBP}) @b{tfind line 444}
16014 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
16016 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
16018 (@value{GDBP}) @b{tdump}
16019 Data collected at tracepoint 2, trace frame 1:
16020 d0 0xc4aa0085 -995491707
16024 d4 0x71aea3d 119204413
16027 d7 0x380035 3670069
16028 a0 0x19e24a 1696330
16029 a1 0x3000668 50333288
16031 a3 0x322000 3284992
16032 a4 0x3000698 50333336
16033 a5 0x1ad3cc 1758156
16034 fp 0x30bf3c 0x30bf3c
16035 sp 0x30bf34 0x30bf34
16037 pc 0x20b2c8 0x20b2c8
16041 p = 0x20e5b4 "gdb-test"
16048 gdb_long_test = 17 '\021'
16053 @code{tdump} works by scanning the tracepoint's current collection
16054 actions and printing the value of each expression listed. So
16055 @code{tdump} can fail, if after a run, you change the tracepoint's
16056 actions to mention variables that were not collected during the run.
16058 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
16059 uses the collected value of @code{$pc} to distinguish between trace
16060 frames that were collected at the tracepoint hit, and frames that were
16061 collected while stepping. This allows it to correctly choose whether
16062 to display the basic list of collections, or the collections from the
16063 body of the while-stepping loop. However, if @code{$pc} was not collected,
16064 then @code{tdump} will always attempt to dump using the basic collection
16065 list, and may fail if a while-stepping frame does not include all the
16066 same data that is collected at the tracepoint hit.
16067 @c This is getting pretty arcane, example would be good.
16069 @node save tracepoints
16070 @subsection @code{save tracepoints @var{filename}}
16071 @kindex save tracepoints
16072 @kindex save-tracepoints
16073 @cindex save tracepoints for future sessions
16075 This command saves all current tracepoint definitions together with
16076 their actions and passcounts, into a file @file{@var{filename}}
16077 suitable for use in a later debugging session. To read the saved
16078 tracepoint definitions, use the @code{source} command (@pxref{Command
16079 Files}). The @w{@code{save-tracepoints}} command is a deprecated
16080 alias for @w{@code{save tracepoints}}
16082 @node Tracepoint Variables
16083 @section Convenience Variables for Tracepoints
16084 @cindex tracepoint variables
16085 @cindex convenience variables for tracepoints
16088 @vindex $trace_frame
16089 @item (int) $trace_frame
16090 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
16091 snapshot is selected.
16093 @vindex $tracepoint
16094 @item (int) $tracepoint
16095 The tracepoint for the current trace snapshot.
16097 @vindex $trace_line
16098 @item (int) $trace_line
16099 The line number for the current trace snapshot.
16101 @vindex $trace_file
16102 @item (char []) $trace_file
16103 The source file for the current trace snapshot.
16105 @vindex $trace_func
16106 @item (char []) $trace_func
16107 The name of the function containing @code{$tracepoint}.
16110 Note: @code{$trace_file} is not suitable for use in @code{printf},
16111 use @code{output} instead.
16113 Here's a simple example of using these convenience variables for
16114 stepping through all the trace snapshots and printing some of their
16115 data. Note that these are not the same as trace state variables,
16116 which are managed by the target.
16119 (@value{GDBP}) @b{tfind start}
16121 (@value{GDBP}) @b{while $trace_frame != -1}
16122 > output $trace_file
16123 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
16129 @section Using Trace Files
16130 @cindex trace files
16132 In some situations, the target running a trace experiment may no
16133 longer be available; perhaps it crashed, or the hardware was needed
16134 for a different activity. To handle these cases, you can arrange to
16135 dump the trace data into a file, and later use that file as a source
16136 of trace data, via the @code{target tfile} command.
16141 @item tsave [ -r ] @var{filename}
16142 @itemx tsave [-ctf] @var{dirname}
16143 Save the trace data to @var{filename}. By default, this command
16144 assumes that @var{filename} refers to the host filesystem, so if
16145 necessary @value{GDBN} will copy raw trace data up from the target and
16146 then save it. If the target supports it, you can also supply the
16147 optional argument @code{-r} (``remote'') to direct the target to save
16148 the data directly into @var{filename} in its own filesystem, which may be
16149 more efficient if the trace buffer is very large. (Note, however, that
16150 @code{target tfile} can only read from files accessible to the host.)
16151 By default, this command will save trace frame in tfile format.
16152 You can supply the optional argument @code{-ctf} to save data in CTF
16153 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
16154 that can be shared by multiple debugging and tracing tools. Please go to
16155 @indicateurl{http://www.efficios.com/ctf} to get more information.
16157 @kindex target tfile
16161 @item target tfile @var{filename}
16162 @itemx target ctf @var{dirname}
16163 Use the file named @var{filename} or directory named @var{dirname} as
16164 a source of trace data. Commands that examine data work as they do with
16165 a live target, but it is not possible to run any new trace experiments.
16166 @code{tstatus} will report the state of the trace run at the moment
16167 the data was saved, as well as the current trace frame you are examining.
16168 Both @var{filename} and @var{dirname} must be on a filesystem accessible to
16172 (@value{GDBP}) target ctf ctf.ctf
16173 (@value{GDBP}) tfind
16174 Found trace frame 0, tracepoint 2
16175 39 ++a; /* set tracepoint 1 here */
16176 (@value{GDBP}) tdump
16177 Data collected at tracepoint 2, trace frame 0:
16181 c = @{"123", "456", "789", "123", "456", "789"@}
16182 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
16190 @chapter Debugging Programs That Use Overlays
16193 If your program is too large to fit completely in your target system's
16194 memory, you can sometimes use @dfn{overlays} to work around this
16195 problem. @value{GDBN} provides some support for debugging programs that
16199 * How Overlays Work:: A general explanation of overlays.
16200 * Overlay Commands:: Managing overlays in @value{GDBN}.
16201 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
16202 mapped by asking the inferior.
16203 * Overlay Sample Program:: A sample program using overlays.
16206 @node How Overlays Work
16207 @section How Overlays Work
16208 @cindex mapped overlays
16209 @cindex unmapped overlays
16210 @cindex load address, overlay's
16211 @cindex mapped address
16212 @cindex overlay area
16214 Suppose you have a computer whose instruction address space is only 64
16215 kilobytes long, but which has much more memory which can be accessed by
16216 other means: special instructions, segment registers, or memory
16217 management hardware, for example. Suppose further that you want to
16218 adapt a program which is larger than 64 kilobytes to run on this system.
16220 One solution is to identify modules of your program which are relatively
16221 independent, and need not call each other directly; call these modules
16222 @dfn{overlays}. Separate the overlays from the main program, and place
16223 their machine code in the larger memory. Place your main program in
16224 instruction memory, but leave at least enough space there to hold the
16225 largest overlay as well.
16227 Now, to call a function located in an overlay, you must first copy that
16228 overlay's machine code from the large memory into the space set aside
16229 for it in the instruction memory, and then jump to its entry point
16232 @c NB: In the below the mapped area's size is greater or equal to the
16233 @c size of all overlays. This is intentional to remind the developer
16234 @c that overlays don't necessarily need to be the same size.
16238 Data Instruction Larger
16239 Address Space Address Space Address Space
16240 +-----------+ +-----------+ +-----------+
16242 +-----------+ +-----------+ +-----------+<-- overlay 1
16243 | program | | main | .----| overlay 1 | load address
16244 | variables | | program | | +-----------+
16245 | and heap | | | | | |
16246 +-----------+ | | | +-----------+<-- overlay 2
16247 | | +-----------+ | | | load address
16248 +-----------+ | | | .-| overlay 2 |
16250 mapped --->+-----------+ | | +-----------+
16251 address | | | | | |
16252 | overlay | <-' | | |
16253 | area | <---' +-----------+<-- overlay 3
16254 | | <---. | | load address
16255 +-----------+ `--| overlay 3 |
16262 @anchor{A code overlay}A code overlay
16266 The diagram (@pxref{A code overlay}) shows a system with separate data
16267 and instruction address spaces. To map an overlay, the program copies
16268 its code from the larger address space to the instruction address space.
16269 Since the overlays shown here all use the same mapped address, only one
16270 may be mapped at a time. For a system with a single address space for
16271 data and instructions, the diagram would be similar, except that the
16272 program variables and heap would share an address space with the main
16273 program and the overlay area.
16275 An overlay loaded into instruction memory and ready for use is called a
16276 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
16277 instruction memory. An overlay not present (or only partially present)
16278 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
16279 is its address in the larger memory. The mapped address is also called
16280 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
16281 called the @dfn{load memory address}, or @dfn{LMA}.
16283 Unfortunately, overlays are not a completely transparent way to adapt a
16284 program to limited instruction memory. They introduce a new set of
16285 global constraints you must keep in mind as you design your program:
16290 Before calling or returning to a function in an overlay, your program
16291 must make sure that overlay is actually mapped. Otherwise, the call or
16292 return will transfer control to the right address, but in the wrong
16293 overlay, and your program will probably crash.
16296 If the process of mapping an overlay is expensive on your system, you
16297 will need to choose your overlays carefully to minimize their effect on
16298 your program's performance.
16301 The executable file you load onto your system must contain each
16302 overlay's instructions, appearing at the overlay's load address, not its
16303 mapped address. However, each overlay's instructions must be relocated
16304 and its symbols defined as if the overlay were at its mapped address.
16305 You can use GNU linker scripts to specify different load and relocation
16306 addresses for pieces of your program; see @ref{Overlay Description,,,
16307 ld.info, Using ld: the GNU linker}.
16310 The procedure for loading executable files onto your system must be able
16311 to load their contents into the larger address space as well as the
16312 instruction and data spaces.
16316 The overlay system described above is rather simple, and could be
16317 improved in many ways:
16322 If your system has suitable bank switch registers or memory management
16323 hardware, you could use those facilities to make an overlay's load area
16324 contents simply appear at their mapped address in instruction space.
16325 This would probably be faster than copying the overlay to its mapped
16326 area in the usual way.
16329 If your overlays are small enough, you could set aside more than one
16330 overlay area, and have more than one overlay mapped at a time.
16333 You can use overlays to manage data, as well as instructions. In
16334 general, data overlays are even less transparent to your design than
16335 code overlays: whereas code overlays only require care when you call or
16336 return to functions, data overlays require care every time you access
16337 the data. Also, if you change the contents of a data overlay, you
16338 must copy its contents back out to its load address before you can copy a
16339 different data overlay into the same mapped area.
16344 @node Overlay Commands
16345 @section Overlay Commands
16347 To use @value{GDBN}'s overlay support, each overlay in your program must
16348 correspond to a separate section of the executable file. The section's
16349 virtual memory address and load memory address must be the overlay's
16350 mapped and load addresses. Identifying overlays with sections allows
16351 @value{GDBN} to determine the appropriate address of a function or
16352 variable, depending on whether the overlay is mapped or not.
16354 @value{GDBN}'s overlay commands all start with the word @code{overlay};
16355 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
16360 Disable @value{GDBN}'s overlay support. When overlay support is
16361 disabled, @value{GDBN} assumes that all functions and variables are
16362 always present at their mapped addresses. By default, @value{GDBN}'s
16363 overlay support is disabled.
16365 @item overlay manual
16366 @cindex manual overlay debugging
16367 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
16368 relies on you to tell it which overlays are mapped, and which are not,
16369 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
16370 commands described below.
16372 @item overlay map-overlay @var{overlay}
16373 @itemx overlay map @var{overlay}
16374 @cindex map an overlay
16375 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
16376 be the name of the object file section containing the overlay. When an
16377 overlay is mapped, @value{GDBN} assumes it can find the overlay's
16378 functions and variables at their mapped addresses. @value{GDBN} assumes
16379 that any other overlays whose mapped ranges overlap that of
16380 @var{overlay} are now unmapped.
16382 @item overlay unmap-overlay @var{overlay}
16383 @itemx overlay unmap @var{overlay}
16384 @cindex unmap an overlay
16385 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
16386 must be the name of the object file section containing the overlay.
16387 When an overlay is unmapped, @value{GDBN} assumes it can find the
16388 overlay's functions and variables at their load addresses.
16391 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
16392 consults a data structure the overlay manager maintains in the inferior
16393 to see which overlays are mapped. For details, see @ref{Automatic
16394 Overlay Debugging}.
16396 @item overlay load-target
16397 @itemx overlay load
16398 @cindex reloading the overlay table
16399 Re-read the overlay table from the inferior. Normally, @value{GDBN}
16400 re-reads the table @value{GDBN} automatically each time the inferior
16401 stops, so this command should only be necessary if you have changed the
16402 overlay mapping yourself using @value{GDBN}. This command is only
16403 useful when using automatic overlay debugging.
16405 @item overlay list-overlays
16406 @itemx overlay list
16407 @cindex listing mapped overlays
16408 Display a list of the overlays currently mapped, along with their mapped
16409 addresses, load addresses, and sizes.
16413 Normally, when @value{GDBN} prints a code address, it includes the name
16414 of the function the address falls in:
16417 (@value{GDBP}) print main
16418 $3 = @{int ()@} 0x11a0 <main>
16421 When overlay debugging is enabled, @value{GDBN} recognizes code in
16422 unmapped overlays, and prints the names of unmapped functions with
16423 asterisks around them. For example, if @code{foo} is a function in an
16424 unmapped overlay, @value{GDBN} prints it this way:
16427 (@value{GDBP}) overlay list
16428 No sections are mapped.
16429 (@value{GDBP}) print foo
16430 $5 = @{int (int)@} 0x100000 <*foo*>
16433 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
16437 (@value{GDBP}) overlay list
16438 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
16439 mapped at 0x1016 - 0x104a
16440 (@value{GDBP}) print foo
16441 $6 = @{int (int)@} 0x1016 <foo>
16444 When overlay debugging is enabled, @value{GDBN} can find the correct
16445 address for functions and variables in an overlay, whether or not the
16446 overlay is mapped. This allows most @value{GDBN} commands, like
16447 @code{break} and @code{disassemble}, to work normally, even on unmapped
16448 code. However, @value{GDBN}'s breakpoint support has some limitations:
16452 @cindex breakpoints in overlays
16453 @cindex overlays, setting breakpoints in
16454 You can set breakpoints in functions in unmapped overlays, as long as
16455 @value{GDBN} can write to the overlay at its load address.
16457 @value{GDBN} can not set hardware or simulator-based breakpoints in
16458 unmapped overlays. However, if you set a breakpoint at the end of your
16459 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
16460 you are using manual overlay management), @value{GDBN} will re-set its
16461 breakpoints properly.
16465 @node Automatic Overlay Debugging
16466 @section Automatic Overlay Debugging
16467 @cindex automatic overlay debugging
16469 @value{GDBN} can automatically track which overlays are mapped and which
16470 are not, given some simple co-operation from the overlay manager in the
16471 inferior. If you enable automatic overlay debugging with the
16472 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
16473 looks in the inferior's memory for certain variables describing the
16474 current state of the overlays.
16476 Here are the variables your overlay manager must define to support
16477 @value{GDBN}'s automatic overlay debugging:
16481 @item @code{_ovly_table}:
16482 This variable must be an array of the following structures:
16487 /* The overlay's mapped address. */
16490 /* The size of the overlay, in bytes. */
16491 unsigned long size;
16493 /* The overlay's load address. */
16496 /* Non-zero if the overlay is currently mapped;
16498 unsigned long mapped;
16502 @item @code{_novlys}:
16503 This variable must be a four-byte signed integer, holding the total
16504 number of elements in @code{_ovly_table}.
16508 To decide whether a particular overlay is mapped or not, @value{GDBN}
16509 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
16510 @code{lma} members equal the VMA and LMA of the overlay's section in the
16511 executable file. When @value{GDBN} finds a matching entry, it consults
16512 the entry's @code{mapped} member to determine whether the overlay is
16515 In addition, your overlay manager may define a function called
16516 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
16517 will silently set a breakpoint there. If the overlay manager then
16518 calls this function whenever it has changed the overlay table, this
16519 will enable @value{GDBN} to accurately keep track of which overlays
16520 are in program memory, and update any breakpoints that may be set
16521 in overlays. This will allow breakpoints to work even if the
16522 overlays are kept in ROM or other non-writable memory while they
16523 are not being executed.
16525 @node Overlay Sample Program
16526 @section Overlay Sample Program
16527 @cindex overlay example program
16529 When linking a program which uses overlays, you must place the overlays
16530 at their load addresses, while relocating them to run at their mapped
16531 addresses. To do this, you must write a linker script (@pxref{Overlay
16532 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
16533 since linker scripts are specific to a particular host system, target
16534 architecture, and target memory layout, this manual cannot provide
16535 portable sample code demonstrating @value{GDBN}'s overlay support.
16537 However, the @value{GDBN} source distribution does contain an overlaid
16538 program, with linker scripts for a few systems, as part of its test
16539 suite. The program consists of the following files from
16540 @file{gdb/testsuite/gdb.base}:
16544 The main program file.
16546 A simple overlay manager, used by @file{overlays.c}.
16551 Overlay modules, loaded and used by @file{overlays.c}.
16554 Linker scripts for linking the test program on the @code{d10v-elf}
16555 and @code{m32r-elf} targets.
16558 You can build the test program using the @code{d10v-elf} GCC
16559 cross-compiler like this:
16562 $ d10v-elf-gcc -g -c overlays.c
16563 $ d10v-elf-gcc -g -c ovlymgr.c
16564 $ d10v-elf-gcc -g -c foo.c
16565 $ d10v-elf-gcc -g -c bar.c
16566 $ d10v-elf-gcc -g -c baz.c
16567 $ d10v-elf-gcc -g -c grbx.c
16568 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
16569 baz.o grbx.o -Wl,-Td10v.ld -o overlays
16572 The build process is identical for any other architecture, except that
16573 you must substitute the appropriate compiler and linker script for the
16574 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
16578 @chapter Using @value{GDBN} with Different Languages
16581 Although programming languages generally have common aspects, they are
16582 rarely expressed in the same manner. For instance, in ANSI C,
16583 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
16584 Modula-2, it is accomplished by @code{p^}. Values can also be
16585 represented (and displayed) differently. Hex numbers in C appear as
16586 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
16588 @cindex working language
16589 Language-specific information is built into @value{GDBN} for some languages,
16590 allowing you to express operations like the above in your program's
16591 native language, and allowing @value{GDBN} to output values in a manner
16592 consistent with the syntax of your program's native language. The
16593 language you use to build expressions is called the @dfn{working
16597 * Setting:: Switching between source languages
16598 * Show:: Displaying the language
16599 * Checks:: Type and range checks
16600 * Supported Languages:: Supported languages
16601 * Unsupported Languages:: Unsupported languages
16605 @section Switching Between Source Languages
16607 There are two ways to control the working language---either have @value{GDBN}
16608 set it automatically, or select it manually yourself. You can use the
16609 @code{set language} command for either purpose. On startup, @value{GDBN}
16610 defaults to setting the language automatically. The working language is
16611 used to determine how expressions you type are interpreted, how values
16614 In addition to the working language, every source file that
16615 @value{GDBN} knows about has its own working language. For some object
16616 file formats, the compiler might indicate which language a particular
16617 source file is in. However, most of the time @value{GDBN} infers the
16618 language from the name of the file. The language of a source file
16619 controls whether C@t{++} names are demangled---this way @code{backtrace} can
16620 show each frame appropriately for its own language. There is no way to
16621 set the language of a source file from within @value{GDBN}, but you can
16622 set the language associated with a filename extension. @xref{Show, ,
16623 Displaying the Language}.
16625 This is most commonly a problem when you use a program, such
16626 as @code{cfront} or @code{f2c}, that generates C but is written in
16627 another language. In that case, make the
16628 program use @code{#line} directives in its C output; that way
16629 @value{GDBN} will know the correct language of the source code of the original
16630 program, and will display that source code, not the generated C code.
16633 * Filenames:: Filename extensions and languages.
16634 * Manually:: Setting the working language manually
16635 * Automatically:: Having @value{GDBN} infer the source language
16639 @subsection List of Filename Extensions and Languages
16641 If a source file name ends in one of the following extensions, then
16642 @value{GDBN} infers that its language is the one indicated.
16660 C@t{++} source file
16666 Objective-C source file
16670 Fortran source file
16673 Modula-2 source file
16677 Assembler source file. This actually behaves almost like C, but
16678 @value{GDBN} does not skip over function prologues when stepping.
16681 In addition, you may set the language associated with a filename
16682 extension. @xref{Show, , Displaying the Language}.
16685 @subsection Setting the Working Language
16687 If you allow @value{GDBN} to set the language automatically,
16688 expressions are interpreted the same way in your debugging session and
16691 @kindex set language
16692 If you wish, you may set the language manually. To do this, issue the
16693 command @samp{set language @var{lang}}, where @var{lang} is the name of
16694 a language, such as
16695 @code{c} or @code{modula-2}.
16696 For a list of the supported languages, type @samp{set language}.
16698 Setting the language manually prevents @value{GDBN} from updating the working
16699 language automatically. This can lead to confusion if you try
16700 to debug a program when the working language is not the same as the
16701 source language, when an expression is acceptable to both
16702 languages---but means different things. For instance, if the current
16703 source file were written in C, and @value{GDBN} was parsing Modula-2, a
16711 might not have the effect you intended. In C, this means to add
16712 @code{b} and @code{c} and place the result in @code{a}. The result
16713 printed would be the value of @code{a}. In Modula-2, this means to compare
16714 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
16716 @node Automatically
16717 @subsection Having @value{GDBN} Infer the Source Language
16719 To have @value{GDBN} set the working language automatically, use
16720 @samp{set language local} or @samp{set language auto}. @value{GDBN}
16721 then infers the working language. That is, when your program stops in a
16722 frame (usually by encountering a breakpoint), @value{GDBN} sets the
16723 working language to the language recorded for the function in that
16724 frame. If the language for a frame is unknown (that is, if the function
16725 or block corresponding to the frame was defined in a source file that
16726 does not have a recognized extension), the current working language is
16727 not changed, and @value{GDBN} issues a warning.
16729 This may not seem necessary for most programs, which are written
16730 entirely in one source language. However, program modules and libraries
16731 written in one source language can be used by a main program written in
16732 a different source language. Using @samp{set language auto} in this
16733 case frees you from having to set the working language manually.
16736 @section Displaying the Language
16738 The following commands help you find out which language is the
16739 working language, and also what language source files were written in.
16742 @item show language
16743 @anchor{show language}
16744 @kindex show language
16745 Display the current working language. This is the
16746 language you can use with commands such as @code{print} to
16747 build and compute expressions that may involve variables in your program.
16750 @kindex info frame@r{, show the source language}
16751 Display the source language for this frame. This language becomes the
16752 working language if you use an identifier from this frame.
16753 @xref{Frame Info, ,Information about a Frame}, to identify the other
16754 information listed here.
16757 @kindex info source@r{, show the source language}
16758 Display the source language of this source file.
16759 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
16760 information listed here.
16763 In unusual circumstances, you may have source files with extensions
16764 not in the standard list. You can then set the extension associated
16765 with a language explicitly:
16768 @item set extension-language @var{ext} @var{language}
16769 @kindex set extension-language
16770 Tell @value{GDBN} that source files with extension @var{ext} are to be
16771 assumed as written in the source language @var{language}.
16773 @item info extensions
16774 @kindex info extensions
16775 List all the filename extensions and the associated languages.
16779 @section Type and Range Checking
16781 Some languages are designed to guard you against making seemingly common
16782 errors through a series of compile- and run-time checks. These include
16783 checking the type of arguments to functions and operators and making
16784 sure mathematical overflows are caught at run time. Checks such as
16785 these help to ensure a program's correctness once it has been compiled
16786 by eliminating type mismatches and providing active checks for range
16787 errors when your program is running.
16789 By default @value{GDBN} checks for these errors according to the
16790 rules of the current source language. Although @value{GDBN} does not check
16791 the statements in your program, it can check expressions entered directly
16792 into @value{GDBN} for evaluation via the @code{print} command, for example.
16795 * Type Checking:: An overview of type checking
16796 * Range Checking:: An overview of range checking
16799 @cindex type checking
16800 @cindex checks, type
16801 @node Type Checking
16802 @subsection An Overview of Type Checking
16804 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
16805 arguments to operators and functions have to be of the correct type,
16806 otherwise an error occurs. These checks prevent type mismatch
16807 errors from ever causing any run-time problems. For example,
16810 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
16812 (@value{GDBP}) print obj.my_method (0)
16815 (@value{GDBP}) print obj.my_method (0x1234)
16816 Cannot resolve method klass::my_method to any overloaded instance
16819 The second example fails because in C@t{++} the integer constant
16820 @samp{0x1234} is not type-compatible with the pointer parameter type.
16822 For the expressions you use in @value{GDBN} commands, you can tell
16823 @value{GDBN} to not enforce strict type checking or
16824 to treat any mismatches as errors and abandon the expression;
16825 When type checking is disabled, @value{GDBN} successfully evaluates
16826 expressions like the second example above.
16828 Even if type checking is off, there may be other reasons
16829 related to type that prevent @value{GDBN} from evaluating an expression.
16830 For instance, @value{GDBN} does not know how to add an @code{int} and
16831 a @code{struct foo}. These particular type errors have nothing to do
16832 with the language in use and usually arise from expressions which make
16833 little sense to evaluate anyway.
16835 @value{GDBN} provides some additional commands for controlling type checking:
16837 @kindex set check type
16838 @kindex show check type
16840 @item set check type on
16841 @itemx set check type off
16842 Set strict type checking on or off. If any type mismatches occur in
16843 evaluating an expression while type checking is on, @value{GDBN} prints a
16844 message and aborts evaluation of the expression.
16846 @item show check type
16847 Show the current setting of type checking and whether @value{GDBN}
16848 is enforcing strict type checking rules.
16851 @cindex range checking
16852 @cindex checks, range
16853 @node Range Checking
16854 @subsection An Overview of Range Checking
16856 In some languages (such as Modula-2), it is an error to exceed the
16857 bounds of a type; this is enforced with run-time checks. Such range
16858 checking is meant to ensure program correctness by making sure
16859 computations do not overflow, or indices on an array element access do
16860 not exceed the bounds of the array.
16862 For expressions you use in @value{GDBN} commands, you can tell
16863 @value{GDBN} to treat range errors in one of three ways: ignore them,
16864 always treat them as errors and abandon the expression, or issue
16865 warnings but evaluate the expression anyway.
16867 A range error can result from numerical overflow, from exceeding an
16868 array index bound, or when you type a constant that is not a member
16869 of any type. Some languages, however, do not treat overflows as an
16870 error. In many implementations of C, mathematical overflow causes the
16871 result to ``wrap around'' to lower values---for example, if @var{m} is
16872 the largest integer value, and @var{s} is the smallest, then
16875 @var{m} + 1 @result{} @var{s}
16878 This, too, is specific to individual languages, and in some cases
16879 specific to individual compilers or machines. @xref{Supported Languages, ,
16880 Supported Languages}, for further details on specific languages.
16882 @value{GDBN} provides some additional commands for controlling the range checker:
16884 @kindex set check range
16885 @kindex show check range
16887 @item set check range auto
16888 Set range checking on or off based on the current working language.
16889 @xref{Supported Languages, ,Supported Languages}, for the default settings for
16892 @item set check range on
16893 @itemx set check range off
16894 Set range checking on or off, overriding the default setting for the
16895 current working language. A warning is issued if the setting does not
16896 match the language default. If a range error occurs and range checking is on,
16897 then a message is printed and evaluation of the expression is aborted.
16899 @item set check range warn
16900 Output messages when the @value{GDBN} range checker detects a range error,
16901 but attempt to evaluate the expression anyway. Evaluating the
16902 expression may still be impossible for other reasons, such as accessing
16903 memory that the process does not own (a typical example from many Unix
16906 @item show check range
16907 Show the current setting of the range checker, and whether or not it is
16908 being set automatically by @value{GDBN}.
16911 @node Supported Languages
16912 @section Supported Languages
16914 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran,
16915 OpenCL C, Pascal, Rust, assembly, Modula-2, and Ada.
16916 @c This is false ...
16917 Some @value{GDBN} features may be used in expressions regardless of the
16918 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
16919 and the @samp{@{type@}addr} construct (@pxref{Expressions,
16920 ,Expressions}) can be used with the constructs of any supported
16923 The following sections detail to what degree each source language is
16924 supported by @value{GDBN}. These sections are not meant to be language
16925 tutorials or references, but serve only as a reference guide to what the
16926 @value{GDBN} expression parser accepts, and what input and output
16927 formats should look like for different languages. There are many good
16928 books written on each of these languages; please look to these for a
16929 language reference or tutorial.
16932 * C:: C and C@t{++}
16935 * Objective-C:: Objective-C
16936 * OpenCL C:: OpenCL C
16937 * Fortran:: Fortran
16940 * Modula-2:: Modula-2
16945 @subsection C and C@t{++}
16947 @cindex C and C@t{++}
16948 @cindex expressions in C or C@t{++}
16950 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
16951 to both languages. Whenever this is the case, we discuss those languages
16955 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
16956 @cindex @sc{gnu} C@t{++}
16957 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
16958 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
16959 effectively, you must compile your C@t{++} programs with a supported
16960 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
16961 compiler (@code{aCC}).
16964 * C Operators:: C and C@t{++} operators
16965 * C Constants:: C and C@t{++} constants
16966 * C Plus Plus Expressions:: C@t{++} expressions
16967 * C Defaults:: Default settings for C and C@t{++}
16968 * C Checks:: C and C@t{++} type and range checks
16969 * Debugging C:: @value{GDBN} and C
16970 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
16971 * Decimal Floating Point:: Numbers in Decimal Floating Point format
16975 @subsubsection C and C@t{++} Operators
16977 @cindex C and C@t{++} operators
16979 Operators must be defined on values of specific types. For instance,
16980 @code{+} is defined on numbers, but not on structures. Operators are
16981 often defined on groups of types.
16983 For the purposes of C and C@t{++}, the following definitions hold:
16988 @emph{Integral types} include @code{int} with any of its storage-class
16989 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
16992 @emph{Floating-point types} include @code{float}, @code{double}, and
16993 @code{long double} (if supported by the target platform).
16996 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
16999 @emph{Scalar types} include all of the above.
17004 The following operators are supported. They are listed here
17005 in order of increasing precedence:
17009 The comma or sequencing operator. Expressions in a comma-separated list
17010 are evaluated from left to right, with the result of the entire
17011 expression being the last expression evaluated.
17014 Assignment. The value of an assignment expression is the value
17015 assigned. Defined on scalar types.
17018 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
17019 and translated to @w{@code{@var{a} = @var{a op b}}}.
17020 @w{@code{@var{op}=}} and @code{=} have the same precedence. The operator
17021 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
17022 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
17025 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
17026 of as: if @var{a} then @var{b} else @var{c}. The argument @var{a}
17027 should be of an integral type.
17030 Logical @sc{or}. Defined on integral types.
17033 Logical @sc{and}. Defined on integral types.
17036 Bitwise @sc{or}. Defined on integral types.
17039 Bitwise exclusive-@sc{or}. Defined on integral types.
17042 Bitwise @sc{and}. Defined on integral types.
17045 Equality and inequality. Defined on scalar types. The value of these
17046 expressions is 0 for false and non-zero for true.
17048 @item <@r{, }>@r{, }<=@r{, }>=
17049 Less than, greater than, less than or equal, greater than or equal.
17050 Defined on scalar types. The value of these expressions is 0 for false
17051 and non-zero for true.
17054 left shift, and right shift. Defined on integral types.
17057 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
17060 Addition and subtraction. Defined on integral types, floating-point types and
17063 @item *@r{, }/@r{, }%
17064 Multiplication, division, and modulus. Multiplication and division are
17065 defined on integral and floating-point types. Modulus is defined on
17069 Increment and decrement. When appearing before a variable, the
17070 operation is performed before the variable is used in an expression;
17071 when appearing after it, the variable's value is used before the
17072 operation takes place.
17075 Pointer dereferencing. Defined on pointer types. Same precedence as
17079 Address operator. Defined on variables. Same precedence as @code{++}.
17081 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
17082 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
17083 to examine the address
17084 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
17088 Negative. Defined on integral and floating-point types. Same
17089 precedence as @code{++}.
17092 Logical negation. Defined on integral types. Same precedence as
17096 Bitwise complement operator. Defined on integral types. Same precedence as
17101 Structure member, and pointer-to-structure member. For convenience,
17102 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
17103 pointer based on the stored type information.
17104 Defined on @code{struct} and @code{union} data.
17107 Dereferences of pointers to members.
17110 Array indexing. @code{@var{a}[@var{i}]} is defined as
17111 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
17114 Function parameter list. Same precedence as @code{->}.
17117 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
17118 and @code{class} types.
17121 Doubled colons also represent the @value{GDBN} scope operator
17122 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
17126 If an operator is redefined in the user code, @value{GDBN} usually
17127 attempts to invoke the redefined version instead of using the operator's
17128 predefined meaning.
17131 @subsubsection C and C@t{++} Constants
17133 @cindex C and C@t{++} constants
17135 @value{GDBN} allows you to express the constants of C and C@t{++} in the
17140 Integer constants are a sequence of digits. Octal constants are
17141 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
17142 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
17143 @samp{l}, specifying that the constant should be treated as a
17147 Floating point constants are a sequence of digits, followed by a decimal
17148 point, followed by a sequence of digits, and optionally followed by an
17149 exponent. An exponent is of the form:
17150 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
17151 sequence of digits. The @samp{+} is optional for positive exponents.
17152 A floating-point constant may also end with a letter @samp{f} or
17153 @samp{F}, specifying that the constant should be treated as being of
17154 the @code{float} (as opposed to the default @code{double}) type; or with
17155 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
17159 Enumerated constants consist of enumerated identifiers, or their
17160 integral equivalents.
17163 Character constants are a single character surrounded by single quotes
17164 (@code{'}), or a number---the ordinal value of the corresponding character
17165 (usually its @sc{ascii} value). Within quotes, the single character may
17166 be represented by a letter or by @dfn{escape sequences}, which are of
17167 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
17168 of the character's ordinal value; or of the form @samp{\@var{x}}, where
17169 @samp{@var{x}} is a predefined special character---for example,
17170 @samp{\n} for newline.
17172 Wide character constants can be written by prefixing a character
17173 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
17174 form of @samp{x}. The target wide character set is used when
17175 computing the value of this constant (@pxref{Character Sets}).
17178 String constants are a sequence of character constants surrounded by
17179 double quotes (@code{"}). Any valid character constant (as described
17180 above) may appear. Double quotes within the string must be preceded by
17181 a backslash, so for instance @samp{"a\"b'c"} is a string of five
17184 Wide string constants can be written by prefixing a string constant
17185 with @samp{L}, as in C. The target wide character set is used when
17186 computing the value of this constant (@pxref{Character Sets}).
17189 Pointer constants are an integral value. You can also write pointers
17190 to constants using the C operator @samp{&}.
17193 Array constants are comma-separated lists surrounded by braces @samp{@{}
17194 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
17195 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
17196 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
17199 @node C Plus Plus Expressions
17200 @subsubsection C@t{++} Expressions
17202 @cindex expressions in C@t{++}
17203 @value{GDBN} expression handling can interpret most C@t{++} expressions.
17205 @cindex debugging C@t{++} programs
17206 @cindex C@t{++} compilers
17207 @cindex debug formats and C@t{++}
17208 @cindex @value{NGCC} and C@t{++}
17210 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
17211 the proper compiler and the proper debug format. Currently,
17212 @value{GDBN} works best when debugging C@t{++} code that is compiled
17213 with the most recent version of @value{NGCC} possible. The DWARF
17214 debugging format is preferred; @value{NGCC} defaults to this on most
17215 popular platforms. Other compilers and/or debug formats are likely to
17216 work badly or not at all when using @value{GDBN} to debug C@t{++}
17217 code. @xref{Compilation}.
17222 @cindex member functions
17224 Member function calls are allowed; you can use expressions like
17227 count = aml->GetOriginal(x, y)
17230 @vindex this@r{, inside C@t{++} member functions}
17231 @cindex namespace in C@t{++}
17233 While a member function is active (in the selected stack frame), your
17234 expressions have the same namespace available as the member function;
17235 that is, @value{GDBN} allows implicit references to the class instance
17236 pointer @code{this} following the same rules as C@t{++}. @code{using}
17237 declarations in the current scope are also respected by @value{GDBN}.
17239 @cindex call overloaded functions
17240 @cindex overloaded functions, calling
17241 @cindex type conversions in C@t{++}
17243 You can call overloaded functions; @value{GDBN} resolves the function
17244 call to the right definition, with some restrictions. @value{GDBN} does not
17245 perform overload resolution involving user-defined type conversions,
17246 calls to constructors, or instantiations of templates that do not exist
17247 in the program. It also cannot handle ellipsis argument lists or
17250 It does perform integral conversions and promotions, floating-point
17251 promotions, arithmetic conversions, pointer conversions, conversions of
17252 class objects to base classes, and standard conversions such as those of
17253 functions or arrays to pointers; it requires an exact match on the
17254 number of function arguments.
17256 Overload resolution is always performed, unless you have specified
17257 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
17258 ,@value{GDBN} Features for C@t{++}}.
17260 You must specify @code{set overload-resolution off} in order to use an
17261 explicit function signature to call an overloaded function, as in
17263 p 'foo(char,int)'('x', 13)
17266 The @value{GDBN} command-completion facility can simplify this;
17267 see @ref{Completion, ,Command Completion}.
17269 @cindex reference declarations
17271 @value{GDBN} understands variables declared as C@t{++} lvalue or rvalue
17272 references; you can use them in expressions just as you do in C@t{++}
17273 source---they are automatically dereferenced.
17275 In the parameter list shown when @value{GDBN} displays a frame, the values of
17276 reference variables are not displayed (unlike other variables); this
17277 avoids clutter, since references are often used for large structures.
17278 The @emph{address} of a reference variable is always shown, unless
17279 you have specified @samp{set print address off}.
17282 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
17283 expressions can use it just as expressions in your program do. Since
17284 one scope may be defined in another, you can use @code{::} repeatedly if
17285 necessary, for example in an expression like
17286 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
17287 resolving name scope by reference to source files, in both C and C@t{++}
17288 debugging (@pxref{Variables, ,Program Variables}).
17291 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
17296 @subsubsection C and C@t{++} Defaults
17298 @cindex C and C@t{++} defaults
17300 If you allow @value{GDBN} to set range checking automatically, it
17301 defaults to @code{off} whenever the working language changes to
17302 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
17303 selects the working language.
17305 If you allow @value{GDBN} to set the language automatically, it
17306 recognizes source files whose names end with @file{.c}, @file{.C}, or
17307 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
17308 these files, it sets the working language to C or C@t{++}.
17309 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
17310 for further details.
17313 @subsubsection C and C@t{++} Type and Range Checks
17315 @cindex C and C@t{++} checks
17317 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
17318 checking is used. However, if you turn type checking off, @value{GDBN}
17319 will allow certain non-standard conversions, such as promoting integer
17320 constants to pointers.
17322 Range checking, if turned on, is done on mathematical operations. Array
17323 indices are not checked, since they are often used to index a pointer
17324 that is not itself an array.
17327 @subsubsection @value{GDBN} and C
17329 The @code{set print union} and @code{show print union} commands apply to
17330 the @code{union} type. When set to @samp{on}, any @code{union} that is
17331 inside a @code{struct} or @code{class} is also printed. Otherwise, it
17332 appears as @samp{@{...@}}.
17334 The @code{@@} operator aids in the debugging of dynamic arrays, formed
17335 with pointers and a memory allocation function. @xref{Expressions,
17338 @node Debugging C Plus Plus
17339 @subsubsection @value{GDBN} Features for C@t{++}
17341 @cindex commands for C@t{++}
17343 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
17344 designed specifically for use with C@t{++}. Here is a summary:
17347 @cindex break in overloaded functions
17348 @item @r{breakpoint menus}
17349 When you want a breakpoint in a function whose name is overloaded,
17350 @value{GDBN} has the capability to display a menu of possible breakpoint
17351 locations to help you specify which function definition you want.
17352 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
17354 @cindex overloading in C@t{++}
17355 @item rbreak @var{regex}
17356 Setting breakpoints using regular expressions is helpful for setting
17357 breakpoints on overloaded functions that are not members of any special
17359 @xref{Set Breaks, ,Setting Breakpoints}.
17361 @cindex C@t{++} exception handling
17363 @itemx catch rethrow
17365 Debug C@t{++} exception handling using these commands. @xref{Set
17366 Catchpoints, , Setting Catchpoints}.
17368 @cindex inheritance
17369 @item ptype @var{typename}
17370 Print inheritance relationships as well as other information for type
17372 @xref{Symbols, ,Examining the Symbol Table}.
17374 @item info vtbl @var{expression}.
17375 The @code{info vtbl} command can be used to display the virtual
17376 method tables of the object computed by @var{expression}. This shows
17377 one entry per virtual table; there may be multiple virtual tables when
17378 multiple inheritance is in use.
17380 @cindex C@t{++} demangling
17381 @item demangle @var{name}
17382 Demangle @var{name}.
17383 @xref{Symbols}, for a more complete description of the @code{demangle} command.
17385 @cindex C@t{++} symbol display
17386 @item set print demangle
17387 @itemx show print demangle
17388 @itemx set print asm-demangle
17389 @itemx show print asm-demangle
17390 Control whether C@t{++} symbols display in their source form, both when
17391 displaying code as C@t{++} source and when displaying disassemblies.
17392 @xref{Print Settings, ,Print Settings}.
17394 @item set print object
17395 @itemx show print object
17396 Choose whether to print derived (actual) or declared types of objects.
17397 @xref{Print Settings, ,Print Settings}.
17399 @item set print vtbl
17400 @itemx show print vtbl
17401 Control the format for printing virtual function tables.
17402 @xref{Print Settings, ,Print Settings}.
17403 (The @code{vtbl} commands do not work on programs compiled with the HP
17404 ANSI C@t{++} compiler (@code{aCC}).)
17406 @kindex set overload-resolution
17407 @cindex overloaded functions, overload resolution
17408 @item set overload-resolution on
17409 Enable overload resolution for C@t{++} expression evaluation. The default
17410 is on. For overloaded functions, @value{GDBN} evaluates the arguments
17411 and searches for a function whose signature matches the argument types,
17412 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
17413 Expressions, ,C@t{++} Expressions}, for details).
17414 If it cannot find a match, it emits a message.
17416 @item set overload-resolution off
17417 Disable overload resolution for C@t{++} expression evaluation. For
17418 overloaded functions that are not class member functions, @value{GDBN}
17419 chooses the first function of the specified name that it finds in the
17420 symbol table, whether or not its arguments are of the correct type. For
17421 overloaded functions that are class member functions, @value{GDBN}
17422 searches for a function whose signature @emph{exactly} matches the
17425 @kindex show overload-resolution
17426 @item show overload-resolution
17427 Show the current setting of overload resolution.
17429 @item @r{Overloaded symbol names}
17430 You can specify a particular definition of an overloaded symbol, using
17431 the same notation that is used to declare such symbols in C@t{++}: type
17432 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
17433 also use the @value{GDBN} command-line word completion facilities to list the
17434 available choices, or to finish the type list for you.
17435 @xref{Completion,, Command Completion}, for details on how to do this.
17437 @item @r{Breakpoints in template functions}
17439 Similar to how overloaded symbols are handled, @value{GDBN} will ignore
17440 template parameter lists when it encounters a symbol which includes a
17441 C@t{++} template. This permits setting breakpoints on families of template functions
17442 or functions whose parameters include template types.
17444 The @kbd{-qualified} flag may be used to override this behavior, causing
17445 @value{GDBN} to search for a specific function or type.
17447 The @value{GDBN} command-line word completion facility also understands
17448 template parameters and may be used to list available choices or finish
17449 template parameter lists for you. @xref{Completion,, Command Completion}, for
17450 details on how to do this.
17452 @item @r{Breakpoints in functions with ABI tags}
17454 The GNU C@t{++} compiler introduced the notion of ABI ``tags'', which
17455 correspond to changes in the ABI of a type, function, or variable that
17456 would not otherwise be reflected in a mangled name. See
17457 @url{https://developers.redhat.com/blog/2015/02/05/gcc5-and-the-c11-abi/}
17460 The ABI tags are visible in C@t{++} demangled names. For example, a
17461 function that returns a std::string:
17464 std::string function(int);
17468 when compiled for the C++11 ABI is marked with the @code{cxx11} ABI
17469 tag, and @value{GDBN} displays the symbol like this:
17472 function[abi:cxx11](int)
17475 You can set a breakpoint on such functions simply as if they had no
17479 (@value{GDBP}) b function(int)
17480 Breakpoint 2 at 0x40060d: file main.cc, line 10.
17481 (@value{GDBP}) info breakpoints
17482 Num Type Disp Enb Address What
17483 1 breakpoint keep y 0x0040060d in function[abi:cxx11](int)
17487 On the rare occasion you need to disambiguate between different ABI
17488 tags, you can do so by simply including the ABI tag in the function
17492 (@value{GDBP}) b ambiguous[abi:other_tag](int)
17496 @node Decimal Floating Point
17497 @subsubsection Decimal Floating Point format
17498 @cindex decimal floating point format
17500 @value{GDBN} can examine, set and perform computations with numbers in
17501 decimal floating point format, which in the C language correspond to the
17502 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
17503 specified by the extension to support decimal floating-point arithmetic.
17505 There are two encodings in use, depending on the architecture: BID (Binary
17506 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
17507 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
17510 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
17511 to manipulate decimal floating point numbers, it is not possible to convert
17512 (using a cast, for example) integers wider than 32-bit to decimal float.
17514 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
17515 point computations, error checking in decimal float operations ignores
17516 underflow, overflow and divide by zero exceptions.
17518 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
17519 to inspect @code{_Decimal128} values stored in floating point registers.
17520 See @ref{PowerPC,,PowerPC} for more details.
17526 @value{GDBN} can be used to debug programs written in D and compiled with
17527 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
17528 specific feature --- dynamic arrays.
17533 @cindex Go (programming language)
17534 @value{GDBN} can be used to debug programs written in Go and compiled with
17535 @file{gccgo} or @file{6g} compilers.
17537 Here is a summary of the Go-specific features and restrictions:
17540 @cindex current Go package
17541 @item The current Go package
17542 The name of the current package does not need to be specified when
17543 specifying global variables and functions.
17545 For example, given the program:
17549 var myglob = "Shall we?"
17555 When stopped inside @code{main} either of these work:
17558 (@value{GDBP}) p myglob
17559 (@value{GDBP}) p main.myglob
17562 @cindex builtin Go types
17563 @item Builtin Go types
17564 The @code{string} type is recognized by @value{GDBN} and is printed
17567 @cindex builtin Go functions
17568 @item Builtin Go functions
17569 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
17570 function and handles it internally.
17572 @cindex restrictions on Go expressions
17573 @item Restrictions on Go expressions
17574 All Go operators are supported except @code{&^}.
17575 The Go @code{_} ``blank identifier'' is not supported.
17576 Automatic dereferencing of pointers is not supported.
17580 @subsection Objective-C
17582 @cindex Objective-C
17583 This section provides information about some commands and command
17584 options that are useful for debugging Objective-C code. See also
17585 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
17586 few more commands specific to Objective-C support.
17589 * Method Names in Commands::
17590 * The Print Command with Objective-C::
17593 @node Method Names in Commands
17594 @subsubsection Method Names in Commands
17596 The following commands have been extended to accept Objective-C method
17597 names as line specifications:
17599 @kindex clear@r{, and Objective-C}
17600 @kindex break@r{, and Objective-C}
17601 @kindex info line@r{, and Objective-C}
17602 @kindex jump@r{, and Objective-C}
17603 @kindex list@r{, and Objective-C}
17607 @item @code{info line}
17612 A fully qualified Objective-C method name is specified as
17615 -[@var{Class} @var{methodName}]
17618 where the minus sign is used to indicate an instance method and a
17619 plus sign (not shown) is used to indicate a class method. The class
17620 name @var{Class} and method name @var{methodName} are enclosed in
17621 brackets, similar to the way messages are specified in Objective-C
17622 source code. For example, to set a breakpoint at the @code{create}
17623 instance method of class @code{Fruit} in the program currently being
17627 break -[Fruit create]
17630 To list ten program lines around the @code{initialize} class method,
17634 list +[NSText initialize]
17637 In the current version of @value{GDBN}, the plus or minus sign is
17638 required. In future versions of @value{GDBN}, the plus or minus
17639 sign will be optional, but you can use it to narrow the search. It
17640 is also possible to specify just a method name:
17646 You must specify the complete method name, including any colons. If
17647 your program's source files contain more than one @code{create} method,
17648 you'll be presented with a numbered list of classes that implement that
17649 method. Indicate your choice by number, or type @samp{0} to exit if
17652 As another example, to clear a breakpoint established at the
17653 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
17656 clear -[NSWindow makeKeyAndOrderFront:]
17659 @node The Print Command with Objective-C
17660 @subsubsection The Print Command With Objective-C
17661 @cindex Objective-C, print objects
17662 @kindex print-object
17663 @kindex po @r{(@code{print-object})}
17665 The print command has also been extended to accept methods. For example:
17668 print -[@var{object} hash]
17671 @cindex print an Objective-C object description
17672 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
17674 will tell @value{GDBN} to send the @code{hash} message to @var{object}
17675 and print the result. Also, an additional command has been added,
17676 @code{print-object} or @code{po} for short, which is meant to print
17677 the description of an object. However, this command may only work
17678 with certain Objective-C libraries that have a particular hook
17679 function, @code{_NSPrintForDebugger}, defined.
17682 @subsection OpenCL C
17685 This section provides information about @value{GDBN}s OpenCL C support.
17688 * OpenCL C Datatypes::
17689 * OpenCL C Expressions::
17690 * OpenCL C Operators::
17693 @node OpenCL C Datatypes
17694 @subsubsection OpenCL C Datatypes
17696 @cindex OpenCL C Datatypes
17697 @value{GDBN} supports the builtin scalar and vector datatypes specified
17698 by OpenCL 1.1. In addition the half- and double-precision floating point
17699 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
17700 extensions are also known to @value{GDBN}.
17702 @node OpenCL C Expressions
17703 @subsubsection OpenCL C Expressions
17705 @cindex OpenCL C Expressions
17706 @value{GDBN} supports accesses to vector components including the access as
17707 lvalue where possible. Since OpenCL C is based on C99 most C expressions
17708 supported by @value{GDBN} can be used as well.
17710 @node OpenCL C Operators
17711 @subsubsection OpenCL C Operators
17713 @cindex OpenCL C Operators
17714 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
17718 @subsection Fortran
17719 @cindex Fortran-specific support in @value{GDBN}
17721 @value{GDBN} can be used to debug programs written in Fortran. Note, that not
17722 all Fortran language features are available yet.
17724 @cindex trailing underscore, in Fortran symbols
17725 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
17726 among them) append an underscore to the names of variables and
17727 functions. When you debug programs compiled by those compilers, you
17728 will need to refer to variables and functions with a trailing
17731 @cindex Fortran Defaults
17732 Fortran symbols are usually case-insensitive, so @value{GDBN} by
17733 default uses case-insensitive matching for Fortran symbols. You can
17734 change that with the @samp{set case-insensitive} command, see
17735 @ref{Symbols}, for the details.
17738 * Fortran Types:: Fortran builtin types
17739 * Fortran Operators:: Fortran operators and expressions
17740 * Fortran Intrinsics:: Fortran intrinsic functions
17741 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
17744 @node Fortran Types
17745 @subsubsection Fortran Types
17747 @cindex Fortran Types
17749 In Fortran the primitive data-types have an associated @code{KIND} type
17750 parameter, written as @samp{@var{type}*@var{kindparam}},
17751 @samp{@var{type}*@var{kindparam}}, or in the @value{GDBN}-only dialect
17752 @samp{@var{type}_@var{kindparam}}. A concrete example would be
17753 @samp{@code{Real*4}}, @samp{@code{Real(kind=4)}}, and @samp{@code{Real_4}}.
17754 The kind of a type can be retrieved by using the intrinsic function
17755 @code{KIND}, see @ref{Fortran Intrinsics}.
17757 Generally, the actual implementation of the @code{KIND} type parameter is
17758 compiler specific. In @value{GDBN} the kind parameter is implemented in
17759 accordance with its use in the @sc{gnu} @command{gfortran} compiler. Here, the
17760 kind parameter for a given @var{type} specifies its size in memory --- a
17761 Fortran @code{Integer*4} or @code{Integer(kind=4)} would be an integer type
17762 occupying 4 bytes of memory. An exception to this rule is the @code{Complex}
17763 type for which the kind of the type does not specify its entire size, but
17764 the size of each of the two @code{Real}'s it is composed of. A
17765 @code{Complex*4} would thus consist of two @code{Real*4}s and occupy 8 bytes
17768 For every type there is also a default kind associated with it, e.g.@
17769 @code{Integer} in @value{GDBN} will internally be an @code{Integer*4} (see the
17770 table below for default types). The default types are the same as in @sc{gnu}
17771 compilers but note, that the @sc{gnu} default types can actually be changed by
17772 compiler flags such as @option{-fdefault-integer-8} and
17773 @option{-fdefault-real-8}.
17775 Not every kind parameter is valid for every type and in @value{GDBN} the
17776 following type kinds are available.
17780 @code{Integer*1}, @code{Integer*2}, @code{Integer*4}, @code{Integer*8}, and
17781 @code{Integer} = @code{Integer*4}.
17784 @code{Logical*1}, @code{Logical*2}, @code{Logical*4}, @code{Logical*8}, and
17785 @code{Logical} = @code{Logical*4}.
17788 @code{Real*4}, @code{Real*8}, @code{Real*16}, and @code{Real} = @code{Real*4}.
17791 @code{Complex*4}, @code{Complex*8}, @code{Complex*16}, and @code{Complex} =
17796 @node Fortran Operators
17797 @subsubsection Fortran Operators and Expressions
17799 @cindex Fortran operators and expressions
17801 Operators must be defined on values of specific types. For instance,
17802 @code{+} is defined on numbers, but not on characters or other non-
17803 arithmetic types. Operators are often defined on groups of types.
17807 The exponentiation operator. It raises the first operand to the power
17811 The range operator. Normally used in the form of array(low:high) to
17812 represent a section of array.
17815 The access component operator. Normally used to access elements in derived
17816 types. Also suitable for unions. As unions aren't part of regular Fortran,
17817 this can only happen when accessing a register that uses a gdbarch-defined
17820 The scope operator. Normally used to access variables in modules or
17821 to set breakpoints on subroutines nested in modules or in other
17822 subroutines (internal subroutines).
17825 @node Fortran Intrinsics
17826 @subsubsection Fortran Intrinsics
17828 @cindex Fortran Intrinsics
17830 Fortran provides a large set of intrinsic procedures. @value{GDBN} implements
17831 an incomplete subset of those procedures and their overloads. Some of these
17832 procedures take an optional @code{KIND} parameter, see @ref{Fortran Types}.
17836 Computes the absolute value of its argument @var{a}. Currently not supported
17837 for @code{Complex} arguments.
17839 @item ALLOCATE(@var{array})
17840 Returns whether @var{array} is allocated or not.
17842 @item ASSOCIATED(@var{pointer} [, @var{target}])
17843 Returns the association status of the pointer @var{pointer} or, if @var{target}
17844 is present, whether @var{pointer} is associated with the target @var{target}.
17846 @item CEILING(@var{a} [, @var{kind}])
17847 Computes the least integer greater than or equal to @var{a}. The optional
17848 parameter @var{kind} specifies the kind of the return type
17849 @code{Integer(@var{kind})}.
17851 @item CMPLX(@var{x} [, @var{y} [, @var{kind}]])
17852 Returns a complex number where @var{x} is converted to the real component. If
17853 @var{y} is present it is converted to the imaginary component. If @var{y} is
17854 not present then the imaginary component is set to @code{0.0} except if @var{x}
17855 itself is of @code{Complex} type. The optional parameter @var{kind} specifies
17856 the kind of the return type @code{Complex(@var{kind})}.
17858 @item FLOOR(@var{a} [, @var{kind}])
17859 Computes the greatest integer less than or equal to @var{a}. The optional
17860 parameter @var{kind} specifies the kind of the return type
17861 @code{Integer(@var{kind})}.
17863 @item KIND(@var{a})
17864 Returns the kind value of the argument @var{a}, see @ref{Fortran Types}.
17866 @item LBOUND(@var{array} [, @var{dim} [, @var{kind}]])
17867 Returns the lower bounds of an @var{array}, or a single lower bound along the
17868 @var{dim} dimension if present. The optional parameter @var{kind} specifies
17869 the kind of the return type @code{Integer(@var{kind})}.
17872 Returns the address of @var{x} as an @code{Integer}.
17874 @item MOD(@var{a}, @var{p})
17875 Computes the remainder of the division of @var{a} by @var{p}.
17877 @item MODULO(@var{a}, @var{p})
17878 Computes the @var{a} modulo @var{p}.
17880 @item RANK(@var{a})
17881 Returns the rank of a scalar or array (scalars have rank @code{0}).
17883 @item SHAPE(@var{a})
17884 Returns the shape of a scalar or array (scalars have shape @samp{()}).
17886 @item SIZE(@var{array}[, @var{dim} [, @var{kind}]])
17887 Returns the extent of @var{array} along a specified dimension @var{dim}, or the
17888 total number of elements in @var{array} if @var{dim} is absent. The optional
17889 parameter @var{kind} specifies the kind of the return type
17890 @code{Integer(@var{kind})}.
17892 @item UBOUND(@var{array} [, @var{dim} [, @var{kind}]])
17893 Returns the upper bounds of an @var{array}, or a single upper bound along the
17894 @var{dim} dimension if present. The optional parameter @var{kind} specifies
17895 the kind of the return type @code{Integer(@var{kind})}.
17899 @node Special Fortran Commands
17900 @subsubsection Special Fortran Commands
17902 @cindex Special Fortran commands
17904 @value{GDBN} has some commands to support Fortran-specific features,
17905 such as displaying common blocks.
17908 @cindex @code{COMMON} blocks, Fortran
17909 @kindex info common
17910 @item info common @r{[}@var{common-name}@r{]}
17911 This command prints the values contained in the Fortran @code{COMMON}
17912 block whose name is @var{common-name}. With no argument, the names of
17913 all @code{COMMON} blocks visible at the current program location are
17915 @cindex arrays slices (Fortran)
17916 @kindex set fortran repack-array-slices
17917 @kindex show fortran repack-array-slices
17918 @item set fortran repack-array-slices [on|off]
17919 @item show fortran repack-array-slices
17920 When taking a slice from an array, a Fortran compiler can choose to
17921 either produce an array descriptor that describes the slice in place,
17922 or it may repack the slice, copying the elements of the slice into a
17923 new region of memory.
17925 When this setting is on, then @value{GDBN} will also repack array
17926 slices in some situations. When this setting is off, then
17927 @value{GDBN} will create array descriptors for slices that reference
17928 the original data in place.
17930 @value{GDBN} will never repack an array slice if the data for the
17931 slice is contiguous within the original array.
17933 @value{GDBN} will always repack string slices if the data for the
17934 slice is non-contiguous within the original string as @value{GDBN}
17935 does not support printing non-contiguous strings.
17937 The default for this setting is @code{off}.
17943 @cindex Pascal support in @value{GDBN}, limitations
17944 Debugging Pascal programs which use sets, subranges, file variables, or
17945 nested functions does not currently work. @value{GDBN} does not support
17946 entering expressions, printing values, or similar features using Pascal
17949 The Pascal-specific command @code{set print pascal_static-members}
17950 controls whether static members of Pascal objects are displayed.
17951 @xref{Print Settings, pascal_static-members}.
17956 @value{GDBN} supports the @url{https://www.rust-lang.org/, Rust
17957 Programming Language}. Type- and value-printing, and expression
17958 parsing, are reasonably complete. However, there are a few
17959 peculiarities and holes to be aware of.
17963 Linespecs (@pxref{Location Specifications}) are never relative to the
17964 current crate. Instead, they act as if there were a global namespace
17965 of crates, somewhat similar to the way @code{extern crate} behaves.
17967 That is, if @value{GDBN} is stopped at a breakpoint in a function in
17968 crate @samp{A}, module @samp{B}, then @code{break B::f} will attempt
17969 to set a breakpoint in a function named @samp{f} in a crate named
17972 As a consequence of this approach, linespecs also cannot refer to
17973 items using @samp{self::} or @samp{super::}.
17976 Because @value{GDBN} implements Rust name-lookup semantics in
17977 expressions, it will sometimes prepend the current crate to a name.
17978 For example, if @value{GDBN} is stopped at a breakpoint in the crate
17979 @samp{K}, then @code{print ::x::y} will try to find the symbol
17982 However, since it is useful to be able to refer to other crates when
17983 debugging, @value{GDBN} provides the @code{extern} extension to
17984 circumvent this. To use the extension, just put @code{extern} before
17985 a path expression to refer to the otherwise unavailable ``global''
17988 In the above example, if you wanted to refer to the symbol @samp{y} in
17989 the crate @samp{x}, you would use @code{print extern x::y}.
17992 The Rust expression evaluator does not support ``statement-like''
17993 expressions such as @code{if} or @code{match}, or lambda expressions.
17996 Tuple expressions are not implemented.
17999 The Rust expression evaluator does not currently implement the
18000 @code{Drop} trait. Objects that may be created by the evaluator will
18001 never be destroyed.
18004 @value{GDBN} does not implement type inference for generics. In order
18005 to call generic functions or otherwise refer to generic items, you
18006 will have to specify the type parameters manually.
18009 @value{GDBN} currently uses the C@t{++} demangler for Rust. In most
18010 cases this does not cause any problems. However, in an expression
18011 context, completing a generic function name will give syntactically
18012 invalid results. This happens because Rust requires the @samp{::}
18013 operator between the function name and its generic arguments. For
18014 example, @value{GDBN} might provide a completion like
18015 @code{crate::f<u32>}, where the parser would require
18016 @code{crate::f::<u32>}.
18019 As of this writing, the Rust compiler (version 1.8) has a few holes in
18020 the debugging information it generates. These holes prevent certain
18021 features from being implemented by @value{GDBN}:
18025 Method calls cannot be made via traits.
18028 Operator overloading is not implemented.
18031 When debugging in a monomorphized function, you cannot use the generic
18035 The type @code{Self} is not available.
18038 @code{use} statements are not available, so some names may not be
18039 available in the crate.
18044 @subsection Modula-2
18046 @cindex Modula-2, @value{GDBN} support
18048 The extensions made to @value{GDBN} to support Modula-2 only support
18049 output from the @sc{gnu} Modula-2 compiler (which is currently being
18050 developed). Other Modula-2 compilers are not currently supported, and
18051 attempting to debug executables produced by them is most likely
18052 to give an error as @value{GDBN} reads in the executable's symbol
18055 @cindex expressions in Modula-2
18057 * M2 Operators:: Built-in operators
18058 * Built-In Func/Proc:: Built-in functions and procedures
18059 * M2 Constants:: Modula-2 constants
18060 * M2 Types:: Modula-2 types
18061 * M2 Defaults:: Default settings for Modula-2
18062 * Deviations:: Deviations from standard Modula-2
18063 * M2 Checks:: Modula-2 type and range checks
18064 * M2 Scope:: The scope operators @code{::} and @code{.}
18065 * GDB/M2:: @value{GDBN} and Modula-2
18069 @subsubsection Operators
18070 @cindex Modula-2 operators
18072 Operators must be defined on values of specific types. For instance,
18073 @code{+} is defined on numbers, but not on structures. Operators are
18074 often defined on groups of types. For the purposes of Modula-2, the
18075 following definitions hold:
18080 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
18084 @emph{Character types} consist of @code{CHAR} and its subranges.
18087 @emph{Floating-point types} consist of @code{REAL}.
18090 @emph{Pointer types} consist of anything declared as @code{POINTER TO
18094 @emph{Scalar types} consist of all of the above.
18097 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
18100 @emph{Boolean types} consist of @code{BOOLEAN}.
18104 The following operators are supported, and appear in order of
18105 increasing precedence:
18109 Function argument or array index separator.
18112 Assignment. The value of @var{var} @code{:=} @var{value} is
18116 Less than, greater than on integral, floating-point, or enumerated
18120 Less than or equal to, greater than or equal to
18121 on integral, floating-point and enumerated types, or set inclusion on
18122 set types. Same precedence as @code{<}.
18124 @item =@r{, }<>@r{, }#
18125 Equality and two ways of expressing inequality, valid on scalar types.
18126 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
18127 available for inequality, since @code{#} conflicts with the script
18131 Set membership. Defined on set types and the types of their members.
18132 Same precedence as @code{<}.
18135 Boolean disjunction. Defined on boolean types.
18138 Boolean conjunction. Defined on boolean types.
18141 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
18144 Addition and subtraction on integral and floating-point types, or union
18145 and difference on set types.
18148 Multiplication on integral and floating-point types, or set intersection
18152 Division on floating-point types, or symmetric set difference on set
18153 types. Same precedence as @code{*}.
18156 Integer division and remainder. Defined on integral types. Same
18157 precedence as @code{*}.
18160 Negative. Defined on @code{INTEGER} and @code{REAL} data.
18163 Pointer dereferencing. Defined on pointer types.
18166 Boolean negation. Defined on boolean types. Same precedence as
18170 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
18171 precedence as @code{^}.
18174 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
18177 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
18181 @value{GDBN} and Modula-2 scope operators.
18185 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
18186 treats the use of the operator @code{IN}, or the use of operators
18187 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
18188 @code{<=}, and @code{>=} on sets as an error.
18192 @node Built-In Func/Proc
18193 @subsubsection Built-in Functions and Procedures
18194 @cindex Modula-2 built-ins
18196 Modula-2 also makes available several built-in procedures and functions.
18197 In describing these, the following metavariables are used:
18202 represents an @code{ARRAY} variable.
18205 represents a @code{CHAR} constant or variable.
18208 represents a variable or constant of integral type.
18211 represents an identifier that belongs to a set. Generally used in the
18212 same function with the metavariable @var{s}. The type of @var{s} should
18213 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
18216 represents a variable or constant of integral or floating-point type.
18219 represents a variable or constant of floating-point type.
18225 represents a variable.
18228 represents a variable or constant of one of many types. See the
18229 explanation of the function for details.
18232 All Modula-2 built-in procedures also return a result, described below.
18236 Returns the absolute value of @var{n}.
18239 If @var{c} is a lower case letter, it returns its upper case
18240 equivalent, otherwise it returns its argument.
18243 Returns the character whose ordinal value is @var{i}.
18246 Decrements the value in the variable @var{v} by one. Returns the new value.
18248 @item DEC(@var{v},@var{i})
18249 Decrements the value in the variable @var{v} by @var{i}. Returns the
18252 @item EXCL(@var{m},@var{s})
18253 Removes the element @var{m} from the set @var{s}. Returns the new
18256 @item FLOAT(@var{i})
18257 Returns the floating point equivalent of the integer @var{i}.
18259 @item HIGH(@var{a})
18260 Returns the index of the last member of @var{a}.
18263 Increments the value in the variable @var{v} by one. Returns the new value.
18265 @item INC(@var{v},@var{i})
18266 Increments the value in the variable @var{v} by @var{i}. Returns the
18269 @item INCL(@var{m},@var{s})
18270 Adds the element @var{m} to the set @var{s} if it is not already
18271 there. Returns the new set.
18274 Returns the maximum value of the type @var{t}.
18277 Returns the minimum value of the type @var{t}.
18280 Returns boolean TRUE if @var{i} is an odd number.
18283 Returns the ordinal value of its argument. For example, the ordinal
18284 value of a character is its @sc{ascii} value (on machines supporting
18285 the @sc{ascii} character set). The argument @var{x} must be of an
18286 ordered type, which include integral, character and enumerated types.
18288 @item SIZE(@var{x})
18289 Returns the size of its argument. The argument @var{x} can be a
18290 variable or a type.
18292 @item TRUNC(@var{r})
18293 Returns the integral part of @var{r}.
18295 @item TSIZE(@var{x})
18296 Returns the size of its argument. The argument @var{x} can be a
18297 variable or a type.
18299 @item VAL(@var{t},@var{i})
18300 Returns the member of the type @var{t} whose ordinal value is @var{i}.
18304 @emph{Warning:} Sets and their operations are not yet supported, so
18305 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
18309 @cindex Modula-2 constants
18311 @subsubsection Constants
18313 @value{GDBN} allows you to express the constants of Modula-2 in the following
18319 Integer constants are simply a sequence of digits. When used in an
18320 expression, a constant is interpreted to be type-compatible with the
18321 rest of the expression. Hexadecimal integers are specified by a
18322 trailing @samp{H}, and octal integers by a trailing @samp{B}.
18325 Floating point constants appear as a sequence of digits, followed by a
18326 decimal point and another sequence of digits. An optional exponent can
18327 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
18328 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
18329 digits of the floating point constant must be valid decimal (base 10)
18333 Character constants consist of a single character enclosed by a pair of
18334 like quotes, either single (@code{'}) or double (@code{"}). They may
18335 also be expressed by their ordinal value (their @sc{ascii} value, usually)
18336 followed by a @samp{C}.
18339 String constants consist of a sequence of characters enclosed by a
18340 pair of like quotes, either single (@code{'}) or double (@code{"}).
18341 Escape sequences in the style of C are also allowed. @xref{C
18342 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
18346 Enumerated constants consist of an enumerated identifier.
18349 Boolean constants consist of the identifiers @code{TRUE} and
18353 Pointer constants consist of integral values only.
18356 Set constants are not yet supported.
18360 @subsubsection Modula-2 Types
18361 @cindex Modula-2 types
18363 Currently @value{GDBN} can print the following data types in Modula-2
18364 syntax: array types, record types, set types, pointer types, procedure
18365 types, enumerated types, subrange types and base types. You can also
18366 print the contents of variables declared using these type.
18367 This section gives a number of simple source code examples together with
18368 sample @value{GDBN} sessions.
18370 The first example contains the following section of code:
18379 and you can request @value{GDBN} to interrogate the type and value of
18380 @code{r} and @code{s}.
18383 (@value{GDBP}) print s
18385 (@value{GDBP}) ptype s
18387 (@value{GDBP}) print r
18389 (@value{GDBP}) ptype r
18394 Likewise if your source code declares @code{s} as:
18398 s: SET ['A'..'Z'] ;
18402 then you may query the type of @code{s} by:
18405 (@value{GDBP}) ptype s
18406 type = SET ['A'..'Z']
18410 Note that at present you cannot interactively manipulate set
18411 expressions using the debugger.
18413 The following example shows how you might declare an array in Modula-2
18414 and how you can interact with @value{GDBN} to print its type and contents:
18418 s: ARRAY [-10..10] OF CHAR ;
18422 (@value{GDBP}) ptype s
18423 ARRAY [-10..10] OF CHAR
18426 Note that the array handling is not yet complete and although the type
18427 is printed correctly, expression handling still assumes that all
18428 arrays have a lower bound of zero and not @code{-10} as in the example
18431 Here are some more type related Modula-2 examples:
18435 colour = (blue, red, yellow, green) ;
18436 t = [blue..yellow] ;
18444 The @value{GDBN} interaction shows how you can query the data type
18445 and value of a variable.
18448 (@value{GDBP}) print s
18450 (@value{GDBP}) ptype t
18451 type = [blue..yellow]
18455 In this example a Modula-2 array is declared and its contents
18456 displayed. Observe that the contents are written in the same way as
18457 their @code{C} counterparts.
18461 s: ARRAY [1..5] OF CARDINAL ;
18467 (@value{GDBP}) print s
18468 $1 = @{1, 0, 0, 0, 0@}
18469 (@value{GDBP}) ptype s
18470 type = ARRAY [1..5] OF CARDINAL
18473 The Modula-2 language interface to @value{GDBN} also understands
18474 pointer types as shown in this example:
18478 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
18485 and you can request that @value{GDBN} describes the type of @code{s}.
18488 (@value{GDBP}) ptype s
18489 type = POINTER TO ARRAY [1..5] OF CARDINAL
18492 @value{GDBN} handles compound types as we can see in this example.
18493 Here we combine array types, record types, pointer types and subrange
18504 myarray = ARRAY myrange OF CARDINAL ;
18505 myrange = [-2..2] ;
18507 s: POINTER TO ARRAY myrange OF foo ;
18511 and you can ask @value{GDBN} to describe the type of @code{s} as shown
18515 (@value{GDBP}) ptype s
18516 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
18519 f3 : ARRAY [-2..2] OF CARDINAL;
18524 @subsubsection Modula-2 Defaults
18525 @cindex Modula-2 defaults
18527 If type and range checking are set automatically by @value{GDBN}, they
18528 both default to @code{on} whenever the working language changes to
18529 Modula-2. This happens regardless of whether you or @value{GDBN}
18530 selected the working language.
18532 If you allow @value{GDBN} to set the language automatically, then entering
18533 code compiled from a file whose name ends with @file{.mod} sets the
18534 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
18535 Infer the Source Language}, for further details.
18538 @subsubsection Deviations from Standard Modula-2
18539 @cindex Modula-2, deviations from
18541 A few changes have been made to make Modula-2 programs easier to debug.
18542 This is done primarily via loosening its type strictness:
18546 Unlike in standard Modula-2, pointer constants can be formed by
18547 integers. This allows you to modify pointer variables during
18548 debugging. (In standard Modula-2, the actual address contained in a
18549 pointer variable is hidden from you; it can only be modified
18550 through direct assignment to another pointer variable or expression that
18551 returned a pointer.)
18554 C escape sequences can be used in strings and characters to represent
18555 non-printable characters. @value{GDBN} prints out strings with these
18556 escape sequences embedded. Single non-printable characters are
18557 printed using the @samp{CHR(@var{nnn})} format.
18560 The assignment operator (@code{:=}) returns the value of its right-hand
18564 All built-in procedures both modify @emph{and} return their argument.
18568 @subsubsection Modula-2 Type and Range Checks
18569 @cindex Modula-2 checks
18572 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
18575 @c FIXME remove warning when type/range checks added
18577 @value{GDBN} considers two Modula-2 variables type equivalent if:
18581 They are of types that have been declared equivalent via a @code{TYPE
18582 @var{t1} = @var{t2}} statement
18585 They have been declared on the same line. (Note: This is true of the
18586 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
18589 As long as type checking is enabled, any attempt to combine variables
18590 whose types are not equivalent is an error.
18592 Range checking is done on all mathematical operations, assignment, array
18593 index bounds, and all built-in functions and procedures.
18596 @subsubsection The Scope Operators @code{::} and @code{.}
18598 @cindex @code{.}, Modula-2 scope operator
18599 @cindex colon, doubled as scope operator
18601 @vindex colon-colon@r{, in Modula-2}
18602 @c Info cannot handle :: but TeX can.
18605 @vindex ::@r{, in Modula-2}
18608 There are a few subtle differences between the Modula-2 scope operator
18609 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
18614 @var{module} . @var{id}
18615 @var{scope} :: @var{id}
18619 where @var{scope} is the name of a module or a procedure,
18620 @var{module} the name of a module, and @var{id} is any declared
18621 identifier within your program, except another module.
18623 Using the @code{::} operator makes @value{GDBN} search the scope
18624 specified by @var{scope} for the identifier @var{id}. If it is not
18625 found in the specified scope, then @value{GDBN} searches all scopes
18626 enclosing the one specified by @var{scope}.
18628 Using the @code{.} operator makes @value{GDBN} search the current scope for
18629 the identifier specified by @var{id} that was imported from the
18630 definition module specified by @var{module}. With this operator, it is
18631 an error if the identifier @var{id} was not imported from definition
18632 module @var{module}, or if @var{id} is not an identifier in
18636 @subsubsection @value{GDBN} and Modula-2
18638 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
18639 Five subcommands of @code{set print} and @code{show print} apply
18640 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
18641 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
18642 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
18643 analogue in Modula-2.
18645 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
18646 with any language, is not useful with Modula-2. Its
18647 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
18648 created in Modula-2 as they can in C or C@t{++}. However, because an
18649 address can be specified by an integral constant, the construct
18650 @samp{@{@var{type}@}@var{adrexp}} is still useful.
18652 @cindex @code{#} in Modula-2
18653 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
18654 interpreted as the beginning of a comment. Use @code{<>} instead.
18660 The extensions made to @value{GDBN} for Ada only support
18661 output from the @sc{gnu} Ada (GNAT) compiler.
18662 Other Ada compilers are not currently supported, and
18663 attempting to debug executables produced by them is most likely
18667 @cindex expressions in Ada
18669 * Ada Mode Intro:: General remarks on the Ada syntax
18670 and semantics supported by Ada mode
18672 * Omissions from Ada:: Restrictions on the Ada expression syntax.
18673 * Additions to Ada:: Extensions of the Ada expression syntax.
18674 * Overloading support for Ada:: Support for expressions involving overloaded
18676 * Stopping Before Main Program:: Debugging the program during elaboration.
18677 * Ada Exceptions:: Ada Exceptions
18678 * Ada Tasks:: Listing and setting breakpoints in tasks.
18679 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
18680 * Ravenscar Profile:: Tasking Support when using the Ravenscar
18682 * Ada Source Character Set:: Character set of Ada source files.
18683 * Ada Glitches:: Known peculiarities of Ada mode.
18686 @node Ada Mode Intro
18687 @subsubsection Introduction
18688 @cindex Ada mode, general
18690 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
18691 syntax, with some extensions.
18692 The philosophy behind the design of this subset is
18696 That @value{GDBN} should provide basic literals and access to operations for
18697 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
18698 leaving more sophisticated computations to subprograms written into the
18699 program (which therefore may be called from @value{GDBN}).
18702 That type safety and strict adherence to Ada language restrictions
18703 are not particularly important to the @value{GDBN} user.
18706 That brevity is important to the @value{GDBN} user.
18709 Thus, for brevity, the debugger acts as if all names declared in
18710 user-written packages are directly visible, even if they are not visible
18711 according to Ada rules, thus making it unnecessary to fully qualify most
18712 names with their packages, regardless of context. Where this causes
18713 ambiguity, @value{GDBN} asks the user's intent.
18715 The debugger will start in Ada mode if it detects an Ada main program.
18716 As for other languages, it will enter Ada mode when stopped in a program that
18717 was translated from an Ada source file.
18719 While in Ada mode, you may use `@t{--}' for comments. This is useful
18720 mostly for documenting command files. The standard @value{GDBN} comment
18721 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
18722 middle (to allow based literals).
18724 @node Omissions from Ada
18725 @subsubsection Omissions from Ada
18726 @cindex Ada, omissions from
18728 Here are the notable omissions from the subset:
18732 Only a subset of the attributes are supported:
18736 @t{'First}, @t{'Last}, and @t{'Length}
18737 on array objects (not on types and subtypes).
18740 @t{'Min} and @t{'Max}.
18743 @t{'Pos} and @t{'Val}.
18749 @t{'Range} on array objects (not subtypes), but only as the right
18750 operand of the membership (@code{in}) operator.
18753 @t{'Access}, @t{'Unchecked_Access}, and
18754 @t{'Unrestricted_Access} (a GNAT extension).
18761 The names in @code{Characters.Latin_1} are not available.
18764 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
18765 equality of representations. They will generally work correctly
18766 for strings and arrays whose elements have integer or enumeration types.
18767 They may not work correctly for arrays whose element
18768 types have user-defined equality, for arrays of real values
18769 (in particular, IEEE-conformant floating point, because of negative
18770 zeroes and NaNs), and for arrays whose elements contain unused bits with
18771 indeterminate values.
18774 The other component-by-component array operations (@code{and}, @code{or},
18775 @code{xor}, @code{not}, and relational tests other than equality)
18776 are not implemented.
18779 @cindex array aggregates (Ada)
18780 @cindex record aggregates (Ada)
18781 @cindex aggregates (Ada)
18782 There is limited support for array and record aggregates. They are
18783 permitted only on the right sides of assignments, as in these examples:
18786 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
18787 (@value{GDBP}) set An_Array := (1, others => 0)
18788 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
18789 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
18790 (@value{GDBP}) set A_Record := (1, "Peter", True);
18791 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
18795 discriminant's value by assigning an aggregate has an
18796 undefined effect if that discriminant is used within the record.
18797 However, you can first modify discriminants by directly assigning to
18798 them (which normally would not be allowed in Ada), and then performing an
18799 aggregate assignment. For example, given a variable @code{A_Rec}
18800 declared to have a type such as:
18803 type Rec (Len : Small_Integer := 0) is record
18805 Vals : IntArray (1 .. Len);
18809 you can assign a value with a different size of @code{Vals} with two
18813 (@value{GDBP}) set A_Rec.Len := 4
18814 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
18817 As this example also illustrates, @value{GDBN} is very loose about the usual
18818 rules concerning aggregates. You may leave out some of the
18819 components of an array or record aggregate (such as the @code{Len}
18820 component in the assignment to @code{A_Rec} above); they will retain their
18821 original values upon assignment. You may freely use dynamic values as
18822 indices in component associations. You may even use overlapping or
18823 redundant component associations, although which component values are
18824 assigned in such cases is not defined.
18827 Calls to dispatching subprograms are not implemented.
18830 The overloading algorithm is much more limited (i.e., less selective)
18831 than that of real Ada. It makes only limited use of the context in
18832 which a subexpression appears to resolve its meaning, and it is much
18833 looser in its rules for allowing type matches. As a result, some
18834 function calls will be ambiguous, and the user will be asked to choose
18835 the proper resolution.
18838 The @code{new} operator is not implemented.
18841 Entry calls are not implemented.
18844 Aside from printing, arithmetic operations on the native VAX floating-point
18845 formats are not supported.
18848 It is not possible to slice a packed array.
18851 The names @code{True} and @code{False}, when not part of a qualified name,
18852 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
18854 Should your program
18855 redefine these names in a package or procedure (at best a dubious practice),
18856 you will have to use fully qualified names to access their new definitions.
18859 Based real literals are not implemented.
18862 @node Additions to Ada
18863 @subsubsection Additions to Ada
18864 @cindex Ada, deviations from
18866 As it does for other languages, @value{GDBN} makes certain generic
18867 extensions to Ada (@pxref{Expressions}):
18871 If the expression @var{E} is a variable residing in memory (typically
18872 a local variable or array element) and @var{N} is a positive integer,
18873 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
18874 @var{N}-1 adjacent variables following it in memory as an array. In
18875 Ada, this operator is generally not necessary, since its prime use is
18876 in displaying parts of an array, and slicing will usually do this in
18877 Ada. However, there are occasional uses when debugging programs in
18878 which certain debugging information has been optimized away.
18881 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
18882 appears in function or file @var{B}.'' When @var{B} is a file name,
18883 you must typically surround it in single quotes.
18886 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
18887 @var{type} that appears at address @var{addr}.''
18890 A name starting with @samp{$} is a convenience variable
18891 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
18894 In addition, @value{GDBN} provides a few other shortcuts and outright
18895 additions specific to Ada:
18899 The assignment statement is allowed as an expression, returning
18900 its right-hand operand as its value. Thus, you may enter
18903 (@value{GDBP}) set x := y + 3
18904 (@value{GDBP}) print A(tmp := y + 1)
18908 The semicolon is allowed as an ``operator,'' returning as its value
18909 the value of its right-hand operand.
18910 This allows, for example,
18911 complex conditional breaks:
18914 (@value{GDBP}) break f
18915 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
18919 An extension to based literals can be used to specify the exact byte
18920 contents of a floating-point literal. After the base, you can use
18921 from zero to two @samp{l} characters, followed by an @samp{f}. The
18922 number of @samp{l} characters controls the width of the resulting real
18923 constant: zero means @code{Float} is used, one means
18924 @code{Long_Float}, and two means @code{Long_Long_Float}.
18927 (@value{GDBP}) print 16f#41b80000#
18932 Rather than use catenation and symbolic character names to introduce special
18933 characters into strings, one may instead use a special bracket notation,
18934 which is also used to print strings. A sequence of characters of the form
18935 @samp{["@var{XX}"]} within a string or character literal denotes the
18936 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
18937 sequence of characters @samp{["""]} also denotes a single quotation mark
18938 in strings. For example,
18940 "One line.["0a"]Next line.["0a"]"
18943 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
18947 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
18948 @t{'Max} is optional (and is ignored in any case). For example, it is valid
18952 (@value{GDBP}) print 'max(x, y)
18956 When printing arrays, @value{GDBN} uses positional notation when the
18957 array has a lower bound of 1, and uses a modified named notation otherwise.
18958 For example, a one-dimensional array of three integers with a lower bound
18959 of 3 might print as
18966 That is, in contrast to valid Ada, only the first component has a @code{=>}
18970 You may abbreviate attributes in expressions with any unique,
18971 multi-character subsequence of
18972 their names (an exact match gets preference).
18973 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
18974 in place of @t{a'length}.
18977 @cindex quoting Ada internal identifiers
18978 Since Ada is case-insensitive, the debugger normally maps identifiers you type
18979 to lower case. The GNAT compiler uses upper-case characters for
18980 some of its internal identifiers, which are normally of no interest to users.
18981 For the rare occasions when you actually have to look at them,
18982 enclose them in angle brackets to avoid the lower-case mapping.
18985 (@value{GDBP}) print <JMPBUF_SAVE>[0]
18989 Printing an object of class-wide type or dereferencing an
18990 access-to-class-wide value will display all the components of the object's
18991 specific type (as indicated by its run-time tag). Likewise, component
18992 selection on such a value will operate on the specific type of the
18997 @node Overloading support for Ada
18998 @subsubsection Overloading support for Ada
18999 @cindex overloading, Ada
19001 The debugger supports limited overloading. Given a subprogram call in which
19002 the function symbol has multiple definitions, it will use the number of
19003 actual parameters and some information about their types to attempt to narrow
19004 the set of definitions. It also makes very limited use of context, preferring
19005 procedures to functions in the context of the @code{call} command, and
19006 functions to procedures elsewhere.
19008 If, after narrowing, the set of matching definitions still contains more than
19009 one definition, @value{GDBN} will display a menu to query which one it should
19013 (@value{GDBP}) print f(1)
19014 Multiple matches for f
19016 [1] foo.f (integer) return boolean at foo.adb:23
19017 [2] foo.f (foo.new_integer) return boolean at foo.adb:28
19021 In this case, just select one menu entry either to cancel expression evaluation
19022 (type @kbd{0} and press @key{RET}) or to continue evaluation with a specific
19023 instance (type the corresponding number and press @key{RET}).
19025 Here are a couple of commands to customize @value{GDBN}'s behavior in this
19030 @kindex set ada print-signatures
19031 @item set ada print-signatures
19032 Control whether parameter types and return types are displayed in overloads
19033 selection menus. It is @code{on} by default.
19034 @xref{Overloading support for Ada}.
19036 @kindex show ada print-signatures
19037 @item show ada print-signatures
19038 Show the current setting for displaying parameter types and return types in
19039 overloads selection menu.
19040 @xref{Overloading support for Ada}.
19044 @node Stopping Before Main Program
19045 @subsubsection Stopping at the Very Beginning
19047 @cindex breakpointing Ada elaboration code
19048 It is sometimes necessary to debug the program during elaboration, and
19049 before reaching the main procedure.
19050 As defined in the Ada Reference
19051 Manual, the elaboration code is invoked from a procedure called
19052 @code{adainit}. To run your program up to the beginning of
19053 elaboration, simply use the following two commands:
19054 @code{tbreak adainit} and @code{run}.
19056 @node Ada Exceptions
19057 @subsubsection Ada Exceptions
19059 A command is provided to list all Ada exceptions:
19062 @kindex info exceptions
19063 @item info exceptions
19064 @itemx info exceptions @var{regexp}
19065 The @code{info exceptions} command allows you to list all Ada exceptions
19066 defined within the program being debugged, as well as their addresses.
19067 With a regular expression, @var{regexp}, as argument, only those exceptions
19068 whose names match @var{regexp} are listed.
19071 Below is a small example, showing how the command can be used, first
19072 without argument, and next with a regular expression passed as an
19076 (@value{GDBP}) info exceptions
19077 All defined Ada exceptions:
19078 constraint_error: 0x613da0
19079 program_error: 0x613d20
19080 storage_error: 0x613ce0
19081 tasking_error: 0x613ca0
19082 const.aint_global_e: 0x613b00
19083 (@value{GDBP}) info exceptions const.aint
19084 All Ada exceptions matching regular expression "const.aint":
19085 constraint_error: 0x613da0
19086 const.aint_global_e: 0x613b00
19089 It is also possible to ask @value{GDBN} to stop your program's execution
19090 when an exception is raised. For more details, see @ref{Set Catchpoints}.
19093 @subsubsection Extensions for Ada Tasks
19094 @cindex Ada, tasking
19096 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
19097 @value{GDBN} provides the following task-related commands:
19102 This command shows a list of current Ada tasks, as in the following example:
19109 (@value{GDBP}) info tasks
19110 ID TID P-ID Pri State Name
19111 1 8088000 0 15 Child Activation Wait main_task
19112 2 80a4000 1 15 Accept Statement b
19113 3 809a800 1 15 Child Activation Wait a
19114 * 4 80ae800 3 15 Runnable c
19119 In this listing, the asterisk before the last task indicates it to be the
19120 task currently being inspected.
19124 Represents @value{GDBN}'s internal task number.
19130 The parent's task ID (@value{GDBN}'s internal task number).
19133 The base priority of the task.
19136 Current state of the task.
19140 The task has been created but has not been activated. It cannot be
19144 The task is not blocked for any reason known to Ada. (It may be waiting
19145 for a mutex, though.) It is conceptually "executing" in normal mode.
19148 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
19149 that were waiting on terminate alternatives have been awakened and have
19150 terminated themselves.
19152 @item Child Activation Wait
19153 The task is waiting for created tasks to complete activation.
19155 @item Accept or Select Term
19156 The task is waiting on an accept or selective wait statement.
19158 @item Waiting on entry call
19159 The task is waiting on an entry call.
19161 @item Async Select Wait
19162 The task is waiting to start the abortable part of an asynchronous
19166 The task is waiting on a select statement with only a delay
19169 @item Child Termination Wait
19170 The task is sleeping having completed a master within itself, and is
19171 waiting for the tasks dependent on that master to become terminated or
19172 waiting on a terminate Phase.
19174 @item Wait Child in Term Alt
19175 The task is sleeping waiting for tasks on terminate alternatives to
19176 finish terminating.
19178 @item Asynchronous Hold
19179 The task has been held by @code{Ada.Asynchronous_Task_Control.Hold_Task}.
19182 The task has been created and is being made runnable.
19184 @item Selective Wait
19185 The task is waiting in a selective wait statement.
19187 @item Accepting RV with @var{taskno}
19188 The task is accepting a rendez-vous with the task @var{taskno}.
19190 @item Waiting on RV with @var{taskno}
19191 The task is waiting for a rendez-vous with the task @var{taskno}.
19195 Name of the task in the program.
19199 @kindex info task @var{taskno}
19200 @item info task @var{taskno}
19201 This command shows detailed information on the specified task, as in
19202 the following example:
19207 (@value{GDBP}) info tasks
19208 ID TID P-ID Pri State Name
19209 1 8077880 0 15 Child Activation Wait main_task
19210 * 2 807c468 1 15 Runnable task_1
19211 (@value{GDBP}) info task 2
19212 Ada Task: 0x807c468
19216 Parent: 1 ("main_task")
19222 @kindex task@r{ (Ada)}
19223 @cindex current Ada task ID
19224 This command prints the ID and name of the current task.
19230 (@value{GDBP}) info tasks
19231 ID TID P-ID Pri State Name
19232 1 8077870 0 15 Child Activation Wait main_task
19233 * 2 807c458 1 15 Runnable some_task
19234 (@value{GDBP}) task
19235 [Current task is 2 "some_task"]
19238 @item task @var{taskno}
19239 @cindex Ada task switching
19240 This command is like the @code{thread @var{thread-id}}
19241 command (@pxref{Threads}). It switches the context of debugging
19242 from the current task to the given task.
19248 (@value{GDBP}) info tasks
19249 ID TID P-ID Pri State Name
19250 1 8077870 0 15 Child Activation Wait main_task
19251 * 2 807c458 1 15 Runnable some_task
19252 (@value{GDBP}) task 1
19253 [Switching to task 1 "main_task"]
19254 #0 0x8067726 in pthread_cond_wait ()
19256 #0 0x8067726 in pthread_cond_wait ()
19257 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
19258 #2 0x805cb63 in system.task_primitives.operations.sleep ()
19259 #3 0x806153e in system.tasking.stages.activate_tasks ()
19260 #4 0x804aacc in un () at un.adb:5
19263 @item task apply [@var{task-id-list} | all] [@var{flag}]@dots{} @var{command}
19264 The @code{task apply} command is the Ada tasking analogue of
19265 @code{thread apply} (@pxref{Threads}). It allows you to apply the
19266 named @var{command} to one or more tasks. Specify the tasks that you
19267 want affected using a list of task IDs, or specify @code{all} to apply
19270 The @var{flag} arguments control what output to produce and how to
19271 handle errors raised when applying @var{command} to a task.
19272 @var{flag} must start with a @code{-} directly followed by one letter
19273 in @code{qcs}. If several flags are provided, they must be given
19274 individually, such as @code{-c -q}.
19276 By default, @value{GDBN} displays some task information before the
19277 output produced by @var{command}, and an error raised during the
19278 execution of a @var{command} will abort @code{task apply}. The
19279 following flags can be used to fine-tune this behavior:
19283 The flag @code{-c}, which stands for @samp{continue}, causes any
19284 errors in @var{command} to be displayed, and the execution of
19285 @code{task apply} then continues.
19287 The flag @code{-s}, which stands for @samp{silent}, causes any errors
19288 or empty output produced by a @var{command} to be silently ignored.
19289 That is, the execution continues, but the task information and errors
19292 The flag @code{-q} (@samp{quiet}) disables printing the task
19296 Flags @code{-c} and @code{-s} cannot be used together.
19298 @item break @var{locspec} task @var{taskno}
19299 @itemx break @var{locspec} task @var{taskno} if @dots{}
19300 @cindex breakpoints and tasks, in Ada
19301 @cindex task breakpoints, in Ada
19302 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
19303 These commands are like the @code{break @dots{} thread @dots{}}
19304 command (@pxref{Thread Stops}). @xref{Location Specifications}, for
19305 the various forms of @var{locspec}.
19307 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
19308 to specify that you only want @value{GDBN} to stop the program when a
19309 particular Ada task reaches this breakpoint. The @var{taskno} is one of the
19310 numeric task identifiers assigned by @value{GDBN}, shown in the first
19311 column of the @samp{info tasks} display.
19313 If you do not specify @samp{task @var{taskno}} when you set a
19314 breakpoint, the breakpoint applies to @emph{all} tasks of your
19317 You can use the @code{task} qualifier on conditional breakpoints as
19318 well; in this case, place @samp{task @var{taskno}} before the
19319 breakpoint condition (before the @code{if}).
19327 (@value{GDBP}) info tasks
19328 ID TID P-ID Pri State Name
19329 1 140022020 0 15 Child Activation Wait main_task
19330 2 140045060 1 15 Accept/Select Wait t2
19331 3 140044840 1 15 Runnable t1
19332 * 4 140056040 1 15 Runnable t3
19333 (@value{GDBP}) b 15 task 2
19334 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
19335 (@value{GDBP}) cont
19340 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
19342 (@value{GDBP}) info tasks
19343 ID TID P-ID Pri State Name
19344 1 140022020 0 15 Child Activation Wait main_task
19345 * 2 140045060 1 15 Runnable t2
19346 3 140044840 1 15 Runnable t1
19347 4 140056040 1 15 Delay Sleep t3
19351 @node Ada Tasks and Core Files
19352 @subsubsection Tasking Support when Debugging Core Files
19353 @cindex Ada tasking and core file debugging
19355 When inspecting a core file, as opposed to debugging a live program,
19356 tasking support may be limited or even unavailable, depending on
19357 the platform being used.
19358 For instance, on x86-linux, the list of tasks is available, but task
19359 switching is not supported.
19361 On certain platforms, the debugger needs to perform some
19362 memory writes in order to provide Ada tasking support. When inspecting
19363 a core file, this means that the core file must be opened with read-write
19364 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
19365 Under these circumstances, you should make a backup copy of the core
19366 file before inspecting it with @value{GDBN}.
19368 @node Ravenscar Profile
19369 @subsubsection Tasking Support when using the Ravenscar Profile
19370 @cindex Ravenscar Profile
19372 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
19373 specifically designed for systems with safety-critical real-time
19377 @kindex set ravenscar task-switching on
19378 @cindex task switching with program using Ravenscar Profile
19379 @item set ravenscar task-switching on
19380 Allows task switching when debugging a program that uses the Ravenscar
19381 Profile. This is the default.
19383 @kindex set ravenscar task-switching off
19384 @item set ravenscar task-switching off
19385 Turn off task switching when debugging a program that uses the Ravenscar
19386 Profile. This is mostly intended to disable the code that adds support
19387 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
19388 the Ravenscar runtime is preventing @value{GDBN} from working properly.
19389 To be effective, this command should be run before the program is started.
19391 @kindex show ravenscar task-switching
19392 @item show ravenscar task-switching
19393 Show whether it is possible to switch from task to task in a program
19394 using the Ravenscar Profile.
19398 @cindex Ravenscar thread
19399 When Ravenscar task-switching is enabled, Ravenscar tasks are
19400 announced by @value{GDBN} as if they were threads:
19404 [New Ravenscar Thread 0x2b8f0]
19407 Both Ravenscar tasks and the underlying CPU threads will show up in
19408 the output of @code{info threads}:
19413 1 Thread 1 (CPU#0 [running]) simple () at simple.adb:10
19414 2 Thread 2 (CPU#1 [running]) 0x0000000000003d34 in __gnat_initialize_cpu_devices ()
19415 3 Thread 3 (CPU#2 [running]) 0x0000000000003d28 in __gnat_initialize_cpu_devices ()
19416 4 Thread 4 (CPU#3 [halted ]) 0x000000000000c6ec in system.task_primitives.operations.idle ()
19417 * 5 Ravenscar Thread 0x2b8f0 simple () at simple.adb:10
19418 6 Ravenscar Thread 0x2f150 0x000000000000c6ec in system.task_primitives.operations.idle ()
19421 One known limitation of the Ravenscar support in @value{GDBN} is that
19422 it isn't currently possible to single-step through the runtime
19423 initialization sequence. If you need to debug this code, you should
19424 use @code{set ravenscar task-switching off}.
19426 @node Ada Source Character Set
19427 @subsubsection Ada Source Character Set
19428 @cindex Ada, source character set
19430 The GNAT compiler supports a number of character sets for source
19431 files. @xref{Character Set Control, , Character Set Control,
19432 gnat_ugn}. @value{GDBN} includes support for this as well.
19435 @item set ada source-charset @var{charset}
19436 @kindex set ada source-charset
19437 Set the source character set for Ada. The character set must be
19438 supported by GNAT. Because this setting affects the decoding of
19439 symbols coming from the debug information in your program, the setting
19440 should be set as early as possible. The default is @code{ISO-8859-1},
19441 because that is also GNAT's default.
19443 @item show ada source-charset
19444 @kindex show ada source-charset
19445 Show the current source character set for Ada.
19449 @subsubsection Known Peculiarities of Ada Mode
19450 @cindex Ada, problems
19452 Besides the omissions listed previously (@pxref{Omissions from Ada}),
19453 we know of several problems with and limitations of Ada mode in
19455 some of which will be fixed with planned future releases of the debugger
19456 and the GNU Ada compiler.
19460 Static constants that the compiler chooses not to materialize as objects in
19461 storage are invisible to the debugger.
19464 Named parameter associations in function argument lists are ignored (the
19465 argument lists are treated as positional).
19468 Many useful library packages are currently invisible to the debugger.
19471 Fixed-point arithmetic, conversions, input, and output is carried out using
19472 floating-point arithmetic, and may give results that only approximate those on
19476 The GNAT compiler never generates the prefix @code{Standard} for any of
19477 the standard symbols defined by the Ada language. @value{GDBN} knows about
19478 this: it will strip the prefix from names when you use it, and will never
19479 look for a name you have so qualified among local symbols, nor match against
19480 symbols in other packages or subprograms. If you have
19481 defined entities anywhere in your program other than parameters and
19482 local variables whose simple names match names in @code{Standard},
19483 GNAT's lack of qualification here can cause confusion. When this happens,
19484 you can usually resolve the confusion
19485 by qualifying the problematic names with package
19486 @code{Standard} explicitly.
19489 Older versions of the compiler sometimes generate erroneous debugging
19490 information, resulting in the debugger incorrectly printing the value
19491 of affected entities. In some cases, the debugger is able to work
19492 around an issue automatically. In other cases, the debugger is able
19493 to work around the issue, but the work-around has to be specifically
19496 @kindex set ada trust-PAD-over-XVS
19497 @kindex show ada trust-PAD-over-XVS
19500 @item set ada trust-PAD-over-XVS on
19501 Configure GDB to strictly follow the GNAT encoding when computing the
19502 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
19503 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
19504 a complete description of the encoding used by the GNAT compiler).
19505 This is the default.
19507 @item set ada trust-PAD-over-XVS off
19508 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
19509 sometimes prints the wrong value for certain entities, changing @code{ada
19510 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
19511 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
19512 @code{off}, but this incurs a slight performance penalty, so it is
19513 recommended to leave this setting to @code{on} unless necessary.
19517 @cindex GNAT descriptive types
19518 @cindex GNAT encoding
19519 Internally, the debugger also relies on the compiler following a number
19520 of conventions known as the @samp{GNAT Encoding}, all documented in
19521 @file{gcc/ada/exp_dbug.ads} in the GCC sources. This encoding describes
19522 how the debugging information should be generated for certain types.
19523 In particular, this convention makes use of @dfn{descriptive types},
19524 which are artificial types generated purely to help the debugger.
19526 These encodings were defined at a time when the debugging information
19527 format used was not powerful enough to describe some of the more complex
19528 types available in Ada. Since DWARF allows us to express nearly all
19529 Ada features, the long-term goal is to slowly replace these descriptive
19530 types by their pure DWARF equivalent. To facilitate that transition,
19531 a new maintenance option is available to force the debugger to ignore
19532 those descriptive types. It allows the user to quickly evaluate how
19533 well @value{GDBN} works without them.
19537 @kindex maint ada set ignore-descriptive-types
19538 @item maintenance ada set ignore-descriptive-types [on|off]
19539 Control whether the debugger should ignore descriptive types.
19540 The default is not to ignore descriptives types (@code{off}).
19542 @kindex maint ada show ignore-descriptive-types
19543 @item maintenance ada show ignore-descriptive-types
19544 Show if descriptive types are ignored by @value{GDBN}.
19548 @node Unsupported Languages
19549 @section Unsupported Languages
19551 @cindex unsupported languages
19552 @cindex minimal language
19553 In addition to the other fully-supported programming languages,
19554 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
19555 It does not represent a real programming language, but provides a set
19556 of capabilities close to what the C or assembly languages provide.
19557 This should allow most simple operations to be performed while debugging
19558 an application that uses a language currently not supported by @value{GDBN}.
19560 If the language is set to @code{auto}, @value{GDBN} will automatically
19561 select this language if the current frame corresponds to an unsupported
19565 @chapter Examining the Symbol Table
19567 The commands described in this chapter allow you to inquire about the
19568 symbols (names of variables, functions and types) defined in your
19569 program. This information is inherent in the text of your program and
19570 does not change as your program executes. @value{GDBN} finds it in your
19571 program's symbol table, in the file indicated when you started @value{GDBN}
19572 (@pxref{File Options, ,Choosing Files}), or by one of the
19573 file-management commands (@pxref{Files, ,Commands to Specify Files}).
19575 @cindex symbol names
19576 @cindex names of symbols
19577 @cindex quoting names
19578 @anchor{quoting names}
19579 Occasionally, you may need to refer to symbols that contain unusual
19580 characters, which @value{GDBN} ordinarily treats as word delimiters. The
19581 most frequent case is in referring to static variables in other
19582 source files (@pxref{Variables,,Program Variables}). File names
19583 are recorded in object files as debugging symbols, but @value{GDBN} would
19584 ordinarily parse a typical file name, like @file{foo.c}, as the three words
19585 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
19586 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
19593 looks up the value of @code{x} in the scope of the file @file{foo.c}.
19596 @cindex case-insensitive symbol names
19597 @cindex case sensitivity in symbol names
19598 @kindex set case-sensitive
19599 @item set case-sensitive on
19600 @itemx set case-sensitive off
19601 @itemx set case-sensitive auto
19602 Normally, when @value{GDBN} looks up symbols, it matches their names
19603 with case sensitivity determined by the current source language.
19604 Occasionally, you may wish to control that. The command @code{set
19605 case-sensitive} lets you do that by specifying @code{on} for
19606 case-sensitive matches or @code{off} for case-insensitive ones. If
19607 you specify @code{auto}, case sensitivity is reset to the default
19608 suitable for the source language. The default is case-sensitive
19609 matches for all languages except for Fortran, for which the default is
19610 case-insensitive matches.
19612 @kindex show case-sensitive
19613 @item show case-sensitive
19614 This command shows the current setting of case sensitivity for symbols
19617 @kindex set print type methods
19618 @item set print type methods
19619 @itemx set print type methods on
19620 @itemx set print type methods off
19621 Normally, when @value{GDBN} prints a class, it displays any methods
19622 declared in that class. You can control this behavior either by
19623 passing the appropriate flag to @code{ptype}, or using @command{set
19624 print type methods}. Specifying @code{on} will cause @value{GDBN} to
19625 display the methods; this is the default. Specifying @code{off} will
19626 cause @value{GDBN} to omit the methods.
19628 @kindex show print type methods
19629 @item show print type methods
19630 This command shows the current setting of method display when printing
19633 @kindex set print type nested-type-limit
19634 @item set print type nested-type-limit @var{limit}
19635 @itemx set print type nested-type-limit unlimited
19636 Set the limit of displayed nested types that the type printer will
19637 show. A @var{limit} of @code{unlimited} or @code{-1} will show all
19638 nested definitions. By default, the type printer will not show any nested
19639 types defined in classes.
19641 @kindex show print type nested-type-limit
19642 @item show print type nested-type-limit
19643 This command shows the current display limit of nested types when
19646 @kindex set print type typedefs
19647 @item set print type typedefs
19648 @itemx set print type typedefs on
19649 @itemx set print type typedefs off
19651 Normally, when @value{GDBN} prints a class, it displays any typedefs
19652 defined in that class. You can control this behavior either by
19653 passing the appropriate flag to @code{ptype}, or using @command{set
19654 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
19655 display the typedef definitions; this is the default. Specifying
19656 @code{off} will cause @value{GDBN} to omit the typedef definitions.
19657 Note that this controls whether the typedef definition itself is
19658 printed, not whether typedef names are substituted when printing other
19661 @kindex show print type typedefs
19662 @item show print type typedefs
19663 This command shows the current setting of typedef display when
19666 @kindex set print type hex
19667 @item set print type hex
19668 @itemx set print type hex on
19669 @itemx set print type hex off
19671 When @value{GDBN} prints sizes and offsets of struct members, it can use
19672 either the decimal or hexadecimal notation. You can select one or the
19673 other either by passing the appropriate flag to @code{ptype}, or by using
19674 the @command{set print type hex} command.
19676 @kindex show print type hex
19677 @item show print type hex
19678 This command shows whether the sizes and offsets of struct members are
19679 printed in decimal or hexadecimal notation.
19681 @kindex info address
19682 @cindex address of a symbol
19683 @item info address @var{symbol}
19684 Describe where the data for @var{symbol} is stored. For a register
19685 variable, this says which register it is kept in. For a non-register
19686 local variable, this prints the stack-frame offset at which the variable
19689 Note the contrast with @samp{print &@var{symbol}}, which does not work
19690 at all for a register variable, and for a stack local variable prints
19691 the exact address of the current instantiation of the variable.
19693 @kindex info symbol
19694 @cindex symbol from address
19695 @cindex closest symbol and offset for an address
19696 @item info symbol @var{addr}
19697 Print the name of a symbol which is stored at the address @var{addr}.
19698 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
19699 nearest symbol and an offset from it:
19702 (@value{GDBP}) info symbol 0x54320
19703 _initialize_vx + 396 in section .text
19707 This is the opposite of the @code{info address} command. You can use
19708 it to find out the name of a variable or a function given its address.
19710 For dynamically linked executables, the name of executable or shared
19711 library containing the symbol is also printed:
19714 (@value{GDBP}) info symbol 0x400225
19715 _start + 5 in section .text of /tmp/a.out
19716 (@value{GDBP}) info symbol 0x2aaaac2811cf
19717 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
19722 @item demangle @r{[}-l @var{language}@r{]} @r{[}@var{--}@r{]} @var{name}
19723 Demangle @var{name}.
19724 If @var{language} is provided it is the name of the language to demangle
19725 @var{name} in. Otherwise @var{name} is demangled in the current language.
19727 The @samp{--} option specifies the end of options,
19728 and is useful when @var{name} begins with a dash.
19730 The parameter @code{demangle-style} specifies how to interpret the kind
19731 of mangling used. @xref{Print Settings}.
19734 @item whatis[/@var{flags}] [@var{arg}]
19735 Print the data type of @var{arg}, which can be either an expression
19736 or a name of a data type. With no argument, print the data type of
19737 @code{$}, the last value in the value history.
19739 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
19740 is not actually evaluated, and any side-effecting operations (such as
19741 assignments or function calls) inside it do not take place.
19743 If @var{arg} is a variable or an expression, @code{whatis} prints its
19744 literal type as it is used in the source code. If the type was
19745 defined using a @code{typedef}, @code{whatis} will @emph{not} print
19746 the data type underlying the @code{typedef}. If the type of the
19747 variable or the expression is a compound data type, such as
19748 @code{struct} or @code{class}, @code{whatis} never prints their
19749 fields or methods. It just prints the @code{struct}/@code{class}
19750 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
19751 such a compound data type, use @code{ptype}.
19753 If @var{arg} is a type name that was defined using @code{typedef},
19754 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
19755 Unrolling means that @code{whatis} will show the underlying type used
19756 in the @code{typedef} declaration of @var{arg}. However, if that
19757 underlying type is also a @code{typedef}, @code{whatis} will not
19760 For C code, the type names may also have the form @samp{class
19761 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
19762 @var{union-tag}} or @samp{enum @var{enum-tag}}.
19764 @var{flags} can be used to modify how the type is displayed.
19765 Available flags are:
19769 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
19770 parameters and typedefs defined in a class when printing the class'
19771 members. The @code{/r} flag disables this.
19774 Do not print methods defined in the class.
19777 Print methods defined in the class. This is the default, but the flag
19778 exists in case you change the default with @command{set print type methods}.
19781 Do not print typedefs defined in the class. Note that this controls
19782 whether the typedef definition itself is printed, not whether typedef
19783 names are substituted when printing other types.
19786 Print typedefs defined in the class. This is the default, but the flag
19787 exists in case you change the default with @command{set print type typedefs}.
19790 Print the offsets and sizes of fields in a struct, similar to what the
19791 @command{pahole} tool does. This option implies the @code{/tm} flags.
19794 Use hexadecimal notation when printing offsets and sizes of fields in a
19798 Use decimal notation when printing offsets and sizes of fields in a
19801 For example, given the following declarations:
19837 Issuing a @kbd{ptype /o struct tuv} command would print:
19840 (@value{GDBP}) ptype /o struct tuv
19841 /* offset | size */ type = struct tuv @{
19842 /* 0 | 4 */ int a1;
19843 /* XXX 4-byte hole */
19844 /* 8 | 8 */ char *a2;
19845 /* 16 | 4 */ int a3;
19847 /* total size (bytes): 24 */
19851 Notice the format of the first column of comments. There, you can
19852 find two parts separated by the @samp{|} character: the @emph{offset},
19853 which indicates where the field is located inside the struct, in
19854 bytes, and the @emph{size} of the field. Another interesting line is
19855 the marker of a @emph{hole} in the struct, indicating that it may be
19856 possible to pack the struct and make it use less space by reorganizing
19859 It is also possible to print offsets inside an union:
19862 (@value{GDBP}) ptype /o union qwe
19863 /* offset | size */ type = union qwe @{
19864 /* 24 */ struct tuv @{
19865 /* 0 | 4 */ int a1;
19866 /* XXX 4-byte hole */
19867 /* 8 | 8 */ char *a2;
19868 /* 16 | 4 */ int a3;
19870 /* total size (bytes): 24 */
19872 /* 40 */ struct xyz @{
19873 /* 0 | 4 */ int f1;
19874 /* 4 | 1 */ char f2;
19875 /* XXX 3-byte hole */
19876 /* 8 | 8 */ void *f3;
19877 /* 16 | 24 */ struct tuv @{
19878 /* 16 | 4 */ int a1;
19879 /* XXX 4-byte hole */
19880 /* 24 | 8 */ char *a2;
19881 /* 32 | 4 */ int a3;
19883 /* total size (bytes): 24 */
19886 /* total size (bytes): 40 */
19889 /* total size (bytes): 40 */
19893 In this case, since @code{struct tuv} and @code{struct xyz} occupy the
19894 same space (because we are dealing with an union), the offset is not
19895 printed for them. However, you can still examine the offset of each
19896 of these structures' fields.
19898 Another useful scenario is printing the offsets of a struct containing
19902 (@value{GDBP}) ptype /o struct tyu
19903 /* offset | size */ type = struct tyu @{
19904 /* 0:31 | 4 */ int a1 : 1;
19905 /* 0:28 | 4 */ int a2 : 3;
19906 /* 0: 5 | 4 */ int a3 : 23;
19907 /* 3: 3 | 1 */ signed char a4 : 2;
19908 /* XXX 3-bit hole */
19909 /* XXX 4-byte hole */
19910 /* 8 | 8 */ int64_t a5;
19911 /* 16: 0 | 4 */ int a6 : 5;
19912 /* 16: 5 | 8 */ int64_t a7 : 3;
19913 /* XXX 7-byte padding */
19915 /* total size (bytes): 24 */
19919 Note how the offset information is now extended to also include the
19920 first bit of the bitfield.
19924 @item ptype[/@var{flags}] [@var{arg}]
19925 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
19926 detailed description of the type, instead of just the name of the type.
19927 @xref{Expressions, ,Expressions}.
19929 Contrary to @code{whatis}, @code{ptype} always unrolls any
19930 @code{typedef}s in its argument declaration, whether the argument is
19931 a variable, expression, or a data type. This means that @code{ptype}
19932 of a variable or an expression will not print literally its type as
19933 present in the source code---use @code{whatis} for that. @code{typedef}s at
19934 the pointer or reference targets are also unrolled. Only @code{typedef}s of
19935 fields, methods and inner @code{class typedef}s of @code{struct}s,
19936 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
19938 For example, for this variable declaration:
19941 typedef double real_t;
19942 struct complex @{ real_t real; double imag; @};
19943 typedef struct complex complex_t;
19945 real_t *real_pointer_var;
19949 the two commands give this output:
19953 (@value{GDBP}) whatis var
19955 (@value{GDBP}) ptype var
19956 type = struct complex @{
19960 (@value{GDBP}) whatis complex_t
19961 type = struct complex
19962 (@value{GDBP}) whatis struct complex
19963 type = struct complex
19964 (@value{GDBP}) ptype struct complex
19965 type = struct complex @{
19969 (@value{GDBP}) whatis real_pointer_var
19971 (@value{GDBP}) ptype real_pointer_var
19977 As with @code{whatis}, using @code{ptype} without an argument refers to
19978 the type of @code{$}, the last value in the value history.
19980 @cindex incomplete type
19981 Sometimes, programs use opaque data types or incomplete specifications
19982 of complex data structure. If the debug information included in the
19983 program does not allow @value{GDBN} to display a full declaration of
19984 the data type, it will say @samp{<incomplete type>}. For example,
19985 given these declarations:
19989 struct foo *fooptr;
19993 but no definition for @code{struct foo} itself, @value{GDBN} will say:
19996 (@value{GDBP}) ptype foo
19997 $1 = <incomplete type>
20001 ``Incomplete type'' is C terminology for data types that are not
20002 completely specified.
20004 @cindex unknown type
20005 Othertimes, information about a variable's type is completely absent
20006 from the debug information included in the program. This most often
20007 happens when the program or library where the variable is defined
20008 includes no debug information at all. @value{GDBN} knows the variable
20009 exists from inspecting the linker/loader symbol table (e.g., the ELF
20010 dynamic symbol table), but such symbols do not contain type
20011 information. Inspecting the type of a (global) variable for which
20012 @value{GDBN} has no type information shows:
20015 (@value{GDBP}) ptype var
20016 type = <data variable, no debug info>
20019 @xref{Variables, no debug info variables}, for how to print the values
20023 @item info types [-q] [@var{regexp}]
20024 Print a brief description of all types whose names match the regular
20025 expression @var{regexp} (or all types in your program, if you supply
20026 no argument). Each complete typename is matched as though it were a
20027 complete line; thus, @samp{i type value} gives information on all
20028 types in your program whose names include the string @code{value}, but
20029 @samp{i type ^value$} gives information only on types whose complete
20030 name is @code{value}.
20032 In programs using different languages, @value{GDBN} chooses the syntax
20033 to print the type description according to the
20034 @samp{set language} value: using @samp{set language auto}
20035 (see @ref{Automatically, ,Set Language Automatically}) means to use the
20036 language of the type, other values mean to use
20037 the manually specified language (see @ref{Manually, ,Set Language Manually}).
20039 This command differs from @code{ptype} in two ways: first, like
20040 @code{whatis}, it does not print a detailed description; second, it
20041 lists all source files and line numbers where a type is defined.
20043 The output from @samp{into types} is proceeded with a header line
20044 describing what types are being listed. The optional flag @samp{-q},
20045 which stands for @samp{quiet}, disables printing this header
20048 @kindex info type-printers
20049 @item info type-printers
20050 Versions of @value{GDBN} that ship with Python scripting enabled may
20051 have ``type printers'' available. When using @command{ptype} or
20052 @command{whatis}, these printers are consulted when the name of a type
20053 is needed. @xref{Type Printing API}, for more information on writing
20056 @code{info type-printers} displays all the available type printers.
20058 @kindex enable type-printer
20059 @kindex disable type-printer
20060 @item enable type-printer @var{name}@dots{}
20061 @item disable type-printer @var{name}@dots{}
20062 These commands can be used to enable or disable type printers.
20065 @cindex local variables
20066 @item info scope @var{locspec}
20067 List all the variables local to the lexical scope of the code location
20068 that results from resolving @var{locspec}. @xref{Location
20069 Specifications}, for details about supported forms of @var{locspec}.
20073 (@value{GDBP}) @b{info scope command_line_handler}
20074 Scope for command_line_handler:
20075 Symbol rl is an argument at stack/frame offset 8, length 4.
20076 Symbol linebuffer is in static storage at address 0x150a18, length 4.
20077 Symbol linelength is in static storage at address 0x150a1c, length 4.
20078 Symbol p is a local variable in register $esi, length 4.
20079 Symbol p1 is a local variable in register $ebx, length 4.
20080 Symbol nline is a local variable in register $edx, length 4.
20081 Symbol repeat is a local variable at frame offset -8, length 4.
20085 This command is especially useful for determining what data to collect
20086 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
20089 @kindex info source
20091 Show information about the current source file---that is, the source file for
20092 the function containing the current point of execution:
20095 the name of the source file, and the directory containing it,
20097 the directory it was compiled in,
20099 its length, in lines,
20101 which programming language it is written in,
20103 if the debug information provides it, the program that compiled the file
20104 (which may include, e.g., the compiler version and command line arguments),
20106 whether the executable includes debugging information for that file, and
20107 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
20109 whether the debugging information includes information about
20110 preprocessor macros.
20114 @kindex info sources
20115 @item info sources @r{[}-dirname | -basename@r{]} @r{[}--@r{]} @r{[}@var{regexp}@r{]}
20118 With no options @samp{info sources} prints the names of all source
20119 files in your program for which there is debugging information. The
20120 source files are presented based on a list of object files
20121 (executables and libraries) currently loaded into @value{GDBN}. For
20122 each object file all of the associated source files are listed.
20124 Each source file will only be printed once for each object file, but a
20125 single source file can be repeated in the output if it is part of
20126 multiple object files.
20128 If the optional @var{regexp} is provided, then only source files that
20129 match the regular expression will be printed. The matching is
20130 case-sensitive, except on operating systems that have case-insensitive
20131 filesystem (e.g., MS-Windows). @samp{--} can be used before
20132 @var{regexp} to prevent @value{GDBN} interpreting @var{regexp} as a
20133 command option (e.g. if @var{regexp} starts with @samp{-}).
20135 By default, the @var{regexp} is used to match anywhere in the
20136 filename. If @code{-dirname}, only files having a dirname matching
20137 @var{regexp} are shown. If @code{-basename}, only files having a
20138 basename matching @var{regexp} are shown.
20140 It is possible that an object file may be printed in the list with no
20141 associated source files. This can happen when either no source files
20142 match @var{regexp}, or, the object file was compiled without debug
20143 information and so @value{GDBN} is unable to find any source file
20146 @kindex info functions
20147 @item info functions [-q] [-n]
20148 Print the names and data types of all defined functions.
20149 Similarly to @samp{info types}, this command groups its output by source
20150 files and annotates each function definition with its source line
20153 In programs using different languages, @value{GDBN} chooses the syntax
20154 to print the function name and type according to the
20155 @samp{set language} value: using @samp{set language auto}
20156 (see @ref{Automatically, ,Set Language Automatically}) means to use the
20157 language of the function, other values mean to use
20158 the manually specified language (see @ref{Manually, ,Set Language Manually}).
20160 The @samp{-n} flag excludes @dfn{non-debugging symbols} from the
20161 results. A non-debugging symbol is a symbol that comes from the
20162 executable's symbol table, not from the debug information (for
20163 example, DWARF) associated with the executable.
20165 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
20166 printing header information and messages explaining why no functions
20169 @item info functions [-q] [-n] [-t @var{type_regexp}] [@var{regexp}]
20170 Like @samp{info functions}, but only print the names and data types
20171 of the functions selected with the provided regexp(s).
20173 If @var{regexp} is provided, print only the functions whose names
20174 match the regular expression @var{regexp}.
20175 Thus, @samp{info fun step} finds all functions whose
20176 names include @code{step}; @samp{info fun ^step} finds those whose names
20177 start with @code{step}. If a function name contains characters that
20178 conflict with the regular expression language (e.g.@:
20179 @samp{operator*()}), they may be quoted with a backslash.
20181 If @var{type_regexp} is provided, print only the functions whose
20182 types, as printed by the @code{whatis} command, match
20183 the regular expression @var{type_regexp}.
20184 If @var{type_regexp} contains space(s), it should be enclosed in
20185 quote characters. If needed, use backslash to escape the meaning
20186 of special characters or quotes.
20187 Thus, @samp{info fun -t '^int ('} finds the functions that return
20188 an integer; @samp{info fun -t '(.*int.*'} finds the functions that
20189 have an argument type containing int; @samp{info fun -t '^int (' ^step}
20190 finds the functions whose names start with @code{step} and that return
20193 If both @var{regexp} and @var{type_regexp} are provided, a function
20194 is printed only if its name matches @var{regexp} and its type matches
20198 @kindex info variables
20199 @item info variables [-q] [-n]
20200 Print the names and data types of all variables that are defined
20201 outside of functions (i.e.@: excluding local variables).
20202 The printed variables are grouped by source files and annotated with
20203 their respective source line numbers.
20205 In programs using different languages, @value{GDBN} chooses the syntax
20206 to print the variable name and type according to the
20207 @samp{set language} value: using @samp{set language auto}
20208 (see @ref{Automatically, ,Set Language Automatically}) means to use the
20209 language of the variable, other values mean to use
20210 the manually specified language (see @ref{Manually, ,Set Language Manually}).
20212 The @samp{-n} flag excludes non-debugging symbols from the results.
20214 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
20215 printing header information and messages explaining why no variables
20218 @item info variables [-q] [-n] [-t @var{type_regexp}] [@var{regexp}]
20219 Like @kbd{info variables}, but only print the variables selected
20220 with the provided regexp(s).
20222 If @var{regexp} is provided, print only the variables whose names
20223 match the regular expression @var{regexp}.
20225 If @var{type_regexp} is provided, print only the variables whose
20226 types, as printed by the @code{whatis} command, match
20227 the regular expression @var{type_regexp}.
20228 If @var{type_regexp} contains space(s), it should be enclosed in
20229 quote characters. If needed, use backslash to escape the meaning
20230 of special characters or quotes.
20232 If both @var{regexp} and @var{type_regexp} are provided, an argument
20233 is printed only if its name matches @var{regexp} and its type matches
20236 @kindex info modules
20238 @item info modules @r{[}-q@r{]} @r{[}@var{regexp}@r{]}
20239 List all Fortran modules in the program, or all modules matching the
20240 optional regular expression @var{regexp}.
20242 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
20243 printing header information and messages explaining why no modules
20246 @kindex info module
20247 @cindex Fortran modules, information about
20248 @cindex functions and variables by Fortran module
20249 @cindex module functions and variables
20250 @item info module functions @r{[}-q@r{]} @r{[}-m @var{module-regexp}@r{]} @r{[}-t @var{type-regexp}@r{]} @r{[}@var{regexp}@r{]}
20251 @itemx info module variables @r{[}-q@r{]} @r{[}-m @var{module-regexp}@r{]} @r{[}-t @var{type-regexp}@r{]} @r{[}@var{regexp}@r{]}
20252 List all functions or variables within all Fortran modules. The set
20253 of functions or variables listed can be limited by providing some or
20254 all of the optional regular expressions. If @var{module-regexp} is
20255 provided, then only Fortran modules matching @var{module-regexp} will
20256 be searched. Only functions or variables whose type matches the
20257 optional regular expression @var{type-regexp} will be listed. And
20258 only functions or variables whose name matches the optional regular
20259 expression @var{regexp} will be listed.
20261 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
20262 printing header information and messages explaining why no functions
20263 or variables have been printed.
20267 Print the name of the starting function of the program. This serves
20268 primarily Fortran programs, which have a user-supplied name for the
20271 @kindex info classes
20272 @cindex Objective-C, classes and selectors
20274 @itemx info classes @var{regexp}
20275 Display all Objective-C classes in your program, or
20276 (with the @var{regexp} argument) all those matching a particular regular
20279 @kindex info selectors
20280 @item info selectors
20281 @itemx info selectors @var{regexp}
20282 Display all Objective-C selectors in your program, or
20283 (with the @var{regexp} argument) all those matching a particular regular
20287 This was never implemented.
20288 @kindex info methods
20290 @itemx info methods @var{regexp}
20291 The @code{info methods} command permits the user to examine all defined
20292 methods within C@t{++} program, or (with the @var{regexp} argument) a
20293 specific set of methods found in the various C@t{++} classes. Many
20294 C@t{++} classes provide a large number of methods. Thus, the output
20295 from the @code{ptype} command can be overwhelming and hard to use. The
20296 @code{info-methods} command filters the methods, printing only those
20297 which match the regular-expression @var{regexp}.
20300 @cindex opaque data types
20301 @kindex set opaque-type-resolution
20302 @item set opaque-type-resolution on
20303 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
20304 declared as a pointer to a @code{struct}, @code{class}, or
20305 @code{union}---for example, @code{struct MyType *}---that is used in one
20306 source file although the full declaration of @code{struct MyType} is in
20307 another source file. The default is on.
20309 A change in the setting of this subcommand will not take effect until
20310 the next time symbols for a file are loaded.
20312 @item set opaque-type-resolution off
20313 Tell @value{GDBN} not to resolve opaque types. In this case, the type
20314 is printed as follows:
20316 @{<no data fields>@}
20319 @kindex show opaque-type-resolution
20320 @item show opaque-type-resolution
20321 Show whether opaque types are resolved or not.
20323 @kindex set print symbol-loading
20324 @cindex print messages when symbols are loaded
20325 @item set print symbol-loading
20326 @itemx set print symbol-loading full
20327 @itemx set print symbol-loading brief
20328 @itemx set print symbol-loading off
20329 The @code{set print symbol-loading} command allows you to control the
20330 printing of messages when @value{GDBN} loads symbol information.
20331 By default a message is printed for the executable and one for each
20332 shared library, and normally this is what you want. However, when
20333 debugging apps with large numbers of shared libraries these messages
20335 When set to @code{brief} a message is printed for each executable,
20336 and when @value{GDBN} loads a collection of shared libraries at once
20337 it will only print one message regardless of the number of shared
20338 libraries. When set to @code{off} no messages are printed.
20340 @kindex show print symbol-loading
20341 @item show print symbol-loading
20342 Show whether messages will be printed when a @value{GDBN} command
20343 entered from the keyboard causes symbol information to be loaded.
20345 @kindex maint print symbols
20346 @cindex symbol dump
20347 @kindex maint print psymbols
20348 @cindex partial symbol dump
20349 @kindex maint print msymbols
20350 @cindex minimal symbol dump
20351 @item maint print symbols @r{[}-pc @var{address}@r{]} @r{[}@var{filename}@r{]}
20352 @itemx maint print symbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
20353 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-pc @var{address}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
20354 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
20355 @itemx maint print msymbols @r{[}-objfile @var{objfile}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
20356 Write a dump of debugging symbol data into the file @var{filename} or
20357 the terminal if @var{filename} is unspecified.
20358 If @code{-objfile @var{objfile}} is specified, only dump symbols for
20360 If @code{-pc @var{address}} is specified, only dump symbols for the file
20361 with code at that address. Note that @var{address} may be a symbol like
20363 If @code{-source @var{source}} is specified, only dump symbols for that
20366 These commands are used to debug the @value{GDBN} symbol-reading code.
20367 These commands do not modify internal @value{GDBN} state, therefore
20368 @samp{maint print symbols} will only print symbols for already expanded symbol
20370 You can use the command @code{info sources} to find out which files these are.
20371 If you use @samp{maint print psymbols} instead, the dump shows information
20372 about symbols that @value{GDBN} only knows partially---that is, symbols
20373 defined in files that @value{GDBN} has skimmed, but not yet read completely.
20374 Finally, @samp{maint print msymbols} just dumps ``minimal symbols'', e.g.,
20377 @xref{Files, ,Commands to Specify Files}, for a discussion of how
20378 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
20380 @kindex maint info symtabs
20381 @kindex maint info psymtabs
20382 @cindex listing @value{GDBN}'s internal symbol tables
20383 @cindex symbol tables, listing @value{GDBN}'s internal
20384 @cindex full symbol tables, listing @value{GDBN}'s internal
20385 @cindex partial symbol tables, listing @value{GDBN}'s internal
20386 @item maint info symtabs @r{[} @var{regexp} @r{]}
20387 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
20389 List the @code{struct symtab} or @code{struct partial_symtab}
20390 structures whose names match @var{regexp}. If @var{regexp} is not
20391 given, list them all. The output includes expressions which you can
20392 copy into a @value{GDBN} debugging this one to examine a particular
20393 structure in more detail. For example:
20396 (@value{GDBP}) maint info psymtabs dwarf2read
20397 @{ objfile /home/gnu/build/gdb/gdb
20398 ((struct objfile *) 0x82e69d0)
20399 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
20400 ((struct partial_symtab *) 0x8474b10)
20403 text addresses 0x814d3c8 -- 0x8158074
20404 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
20405 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
20406 dependencies (none)
20409 (@value{GDBP}) maint info symtabs
20413 We see that there is one partial symbol table whose filename contains
20414 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
20415 and we see that @value{GDBN} has not read in any symtabs yet at all.
20416 If we set a breakpoint on a function, that will cause @value{GDBN} to
20417 read the symtab for the compilation unit containing that function:
20420 (@value{GDBP}) break dwarf2_psymtab_to_symtab
20421 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
20423 (@value{GDBP}) maint info symtabs
20424 @{ objfile /home/gnu/build/gdb/gdb
20425 ((struct objfile *) 0x82e69d0)
20426 @{ symtab /home/gnu/src/gdb/dwarf2read.c
20427 ((struct symtab *) 0x86c1f38)
20430 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
20431 linetable ((struct linetable *) 0x8370fa0)
20432 debugformat DWARF 2
20438 @kindex maint info line-table
20439 @cindex listing @value{GDBN}'s internal line tables
20440 @cindex line tables, listing @value{GDBN}'s internal
20441 @item maint info line-table @r{[} @var{regexp} @r{]}
20443 List the @code{struct linetable} from all @code{struct symtab}
20444 instances whose name matches @var{regexp}. If @var{regexp} is not
20445 given, list the @code{struct linetable} from all @code{struct symtab}.
20449 (@value{GDBP}) maint info line-table
20450 objfile: /home/gnu/build/a.out ((struct objfile *) 0x6120000e0d40)
20451 compunit_symtab: simple.cpp ((struct compunit_symtab *) 0x6210000ff450)
20452 symtab: /home/gnu/src/simple.cpp ((struct symtab *) 0x6210000ff4d0)
20453 linetable: ((struct linetable *) 0x62100012b760):
20454 INDEX LINE ADDRESS IS-STMT PROLOGUE-END
20455 0 3 0x0000000000401110 Y
20456 1 4 0x0000000000401114 Y Y
20457 2 9 0x0000000000401120 Y
20458 3 10 0x0000000000401124 Y Y
20459 4 10 0x0000000000401129
20460 5 15 0x0000000000401130 Y
20461 6 16 0x0000000000401134 Y Y
20462 7 16 0x0000000000401139
20463 8 21 0x0000000000401140 Y
20464 9 22 0x000000000040114f Y Y
20465 10 22 0x0000000000401154
20466 11 END 0x000000000040115a Y
20469 The @samp{IS-STMT} column indicates if the address is a recommended breakpoint
20470 location to represent a line or a statement. The @samp{PROLOGUE-END} column
20471 indicates that a given address is an adequate place to set a breakpoint at the
20472 first instruction following a function prologue.
20474 @kindex set always-read-ctf [on|off]
20475 @kindex show always-read-ctf
20476 @cindex always-read-ctf
20477 @cindex CTF info, when to read
20478 @item set always-read-ctf [on|off]
20479 @itemx show always-read-ctf
20481 When off, CTF debug info is only read if DWARF debug info is not
20482 present. When on, CTF debug info is read regardless of whether DWARF
20483 debug info is present. The default value is off.
20485 @kindex maint set symbol-cache-size
20486 @cindex symbol cache size
20487 @item maint set symbol-cache-size @var{size}
20488 Set the size of the symbol cache to @var{size}.
20489 The default size is intended to be good enough for debugging
20490 most applications. This option exists to allow for experimenting
20491 with different sizes.
20493 @kindex maint show symbol-cache-size
20494 @item maint show symbol-cache-size
20495 Show the size of the symbol cache.
20497 @kindex maint print symbol-cache
20498 @cindex symbol cache, printing its contents
20499 @item maint print symbol-cache
20500 Print the contents of the symbol cache.
20501 This is useful when debugging symbol cache issues.
20503 @kindex maint print symbol-cache-statistics
20504 @cindex symbol cache, printing usage statistics
20505 @item maint print symbol-cache-statistics
20506 Print symbol cache usage statistics.
20507 This helps determine how well the cache is being utilized.
20509 @kindex maint flush symbol-cache
20510 @kindex maint flush-symbol-cache
20511 @cindex symbol cache, flushing
20512 @item maint flush symbol-cache
20513 @itemx maint flush-symbol-cache
20514 Flush the contents of the symbol cache, all entries are removed. This
20515 command is useful when debugging the symbol cache. It is also useful
20516 when collecting performance data. The command @code{maint
20517 flush-symbol-cache} is deprecated in favor of @code{maint flush
20520 @kindex maint set ignore-prologue-end-flag
20521 @cindex prologue-end
20522 @item maint set ignore-prologue-end-flag [on|off]
20523 Enable or disable the use of the @samp{PROLOGUE-END} flag from the line-table.
20524 When @samp{off} (the default), @value{GDBN} uses the @samp{PROLOGUE-END} flag
20525 to place breakpoints past the end of a function prologue. When @samp{on},
20526 @value{GDBN} ignores the flag and relies on prologue analyzers to skip function
20529 @kindex maint show ignore-prologue-end-flag
20530 @item maint show ignore-prologue-end-flag
20531 Show whether @value{GDBN} will ignore the @samp{PROLOGUE-END} flag.
20536 @chapter Altering Execution
20538 Once you think you have found an error in your program, you might want to
20539 find out for certain whether correcting the apparent error would lead to
20540 correct results in the rest of the run. You can find the answer by
20541 experiment, using the @value{GDBN} features for altering execution of the
20544 For example, you can store new values into variables or memory
20545 locations, give your program a signal, restart it at a different
20546 address, or even return prematurely from a function.
20549 * Assignment:: Assignment to variables
20550 * Jumping:: Continuing at a different address
20551 * Signaling:: Giving your program a signal
20552 * Returning:: Returning from a function
20553 * Calling:: Calling your program's functions
20554 * Patching:: Patching your program
20555 * Compiling and Injecting Code:: Compiling and injecting code in @value{GDBN}
20559 @section Assignment to Variables
20562 @cindex setting variables
20563 To alter the value of a variable, evaluate an assignment expression.
20564 @xref{Expressions, ,Expressions}. For example,
20571 stores the value 4 into the variable @code{x}, and then prints the
20572 value of the assignment expression (which is 4).
20573 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
20574 information on operators in supported languages.
20576 @kindex set variable
20577 @cindex variables, setting
20578 If you are not interested in seeing the value of the assignment, use the
20579 @code{set} command instead of the @code{print} command. @code{set} is
20580 really the same as @code{print} except that the expression's value is
20581 not printed and is not put in the value history (@pxref{Value History,
20582 ,Value History}). The expression is evaluated only for its effects.
20584 If the beginning of the argument string of the @code{set} command
20585 appears identical to a @code{set} subcommand, use the @code{set
20586 variable} command instead of just @code{set}. This command is identical
20587 to @code{set} except for its lack of subcommands. For example, if your
20588 program has a variable @code{width}, you get an error if you try to set
20589 a new value with just @samp{set width=13}, because @value{GDBN} has the
20590 command @code{set width}:
20593 (@value{GDBP}) whatis width
20595 (@value{GDBP}) p width
20597 (@value{GDBP}) set width=47
20598 Invalid syntax in expression.
20602 The invalid expression, of course, is @samp{=47}. In
20603 order to actually set the program's variable @code{width}, use
20606 (@value{GDBP}) set var width=47
20609 Because the @code{set} command has many subcommands that can conflict
20610 with the names of program variables, it is a good idea to use the
20611 @code{set variable} command instead of just @code{set}. For example, if
20612 your program has a variable @code{g}, you run into problems if you try
20613 to set a new value with just @samp{set g=4}, because @value{GDBN} has
20614 the command @code{set gnutarget}, abbreviated @code{set g}:
20618 (@value{GDBP}) whatis g
20622 (@value{GDBP}) set g=4
20626 The program being debugged has been started already.
20627 Start it from the beginning? (y or n) y
20628 Starting program: /home/smith/cc_progs/a.out
20629 "/home/smith/cc_progs/a.out": can't open to read symbols:
20630 Invalid bfd target.
20631 (@value{GDBP}) show g
20632 The current BFD target is "=4".
20637 The program variable @code{g} did not change, and you silently set the
20638 @code{gnutarget} to an invalid value. In order to set the variable
20642 (@value{GDBP}) set var g=4
20645 @value{GDBN} allows more implicit conversions in assignments than C; you can
20646 freely store an integer value into a pointer variable or vice versa,
20647 and you can convert any structure to any other structure that is the
20648 same length or shorter.
20649 @comment FIXME: how do structs align/pad in these conversions?
20650 @comment /doc@cygnus.com 18dec1990
20652 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
20653 construct to generate a value of specified type at a specified address
20654 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
20655 to memory location @code{0x83040} as an integer (which implies a certain size
20656 and representation in memory), and
20659 set @{int@}0x83040 = 4
20663 stores the value 4 into that memory location.
20666 @section Continuing at a Different Address
20668 Ordinarily, when you continue your program, you do so at the place where
20669 it stopped, with the @code{continue} command. You can instead continue at
20670 an address of your own choosing, with the following commands:
20674 @kindex j @r{(@code{jump})}
20675 @item jump @var{locspec}
20676 @itemx j @var{locspec}
20677 Resume execution at the address of the code location that results from
20678 resolving @var{locspec}.
20679 @xref{Location Specifications}, for a description of the different
20680 forms of @var{locspec}. If @var{locspec} resolves to more than one address,
20681 those outside the current compilation unit are ignored. If considering just
20682 the addresses in the current compilation unit still doesn't yield a unique
20683 address, the command aborts before jumping.
20684 Execution stops again immediately if there is a breakpoint there. It
20685 is common practice to use the @code{tbreak} command in conjunction
20686 with @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
20688 The @code{jump} command does not change the current stack frame, or
20689 the stack pointer, or the contents of any memory location or any
20690 register other than the program counter. If @var{locspec} resolves to
20691 an address in a different function from the one currently executing, the
20692 results may be bizarre if the two functions expect different patterns
20693 of arguments or of local variables. For this reason, the @code{jump}
20694 command requests confirmation if the jump address is not in the
20695 function currently executing. However, even bizarre results are
20696 predictable if you are well acquainted with the machine-language code
20700 On many systems, you can get much the same effect as the @code{jump}
20701 command by storing a new value into the register @code{$pc}. The
20702 difference is that this does not start your program running; it only
20703 changes the address of where it @emph{will} run when you continue. For
20711 makes the next @code{continue} command or stepping command execute at
20712 address @code{0x485}, rather than at the address where your program stopped.
20713 @xref{Continuing and Stepping, ,Continuing and Stepping}.
20715 However, writing directly to @code{$pc} will only change the value of
20716 the program-counter register, while using @code{jump} will ensure that
20717 any additional auxiliary state is also updated. For example, on
20718 SPARC, @code{jump} will update both @code{$pc} and @code{$npc}
20719 registers prior to resuming execution. When using the approach of
20720 writing directly to @code{$pc} it is your job to also update the
20721 @code{$npc} register.
20723 The most common occasion to use the @code{jump} command is to back
20724 up---perhaps with more breakpoints set---over a portion of a program
20725 that has already executed, in order to examine its execution in more
20730 @section Giving your Program a Signal
20731 @cindex deliver a signal to a program
20735 @item signal @var{signal}
20736 Resume execution where your program is stopped, but immediately give it the
20737 signal @var{signal}. The @var{signal} can be the name or the number of a
20738 signal. For example, on many systems @code{signal 2} and @code{signal
20739 SIGINT} are both ways of sending an interrupt signal.
20741 Alternatively, if @var{signal} is zero, continue execution without
20742 giving a signal. This is useful when your program stopped on account of
20743 a signal and would ordinarily see the signal when resumed with the
20744 @code{continue} command; @samp{signal 0} causes it to resume without a
20747 @emph{Note:} When resuming a multi-threaded program, @var{signal} is
20748 delivered to the currently selected thread, not the thread that last
20749 reported a stop. This includes the situation where a thread was
20750 stopped due to a signal. So if you want to continue execution
20751 suppressing the signal that stopped a thread, you should select that
20752 same thread before issuing the @samp{signal 0} command. If you issue
20753 the @samp{signal 0} command with another thread as the selected one,
20754 @value{GDBN} detects that and asks for confirmation.
20756 Invoking the @code{signal} command is not the same as invoking the
20757 @code{kill} utility from the shell. Sending a signal with @code{kill}
20758 causes @value{GDBN} to decide what to do with the signal depending on
20759 the signal handling tables (@pxref{Signals}). The @code{signal} command
20760 passes the signal directly to your program.
20762 @code{signal} does not repeat when you press @key{RET} a second time
20763 after executing the command.
20765 @kindex queue-signal
20766 @item queue-signal @var{signal}
20767 Queue @var{signal} to be delivered immediately to the current thread
20768 when execution of the thread resumes. The @var{signal} can be the name or
20769 the number of a signal. For example, on many systems @code{signal 2} and
20770 @code{signal SIGINT} are both ways of sending an interrupt signal.
20771 The handling of the signal must be set to pass the signal to the program,
20772 otherwise @value{GDBN} will report an error.
20773 You can control the handling of signals from @value{GDBN} with the
20774 @code{handle} command (@pxref{Signals}).
20776 Alternatively, if @var{signal} is zero, any currently queued signal
20777 for the current thread is discarded and when execution resumes no signal
20778 will be delivered. This is useful when your program stopped on account
20779 of a signal and would ordinarily see the signal when resumed with the
20780 @code{continue} command.
20782 This command differs from the @code{signal} command in that the signal
20783 is just queued, execution is not resumed. And @code{queue-signal} cannot
20784 be used to pass a signal whose handling state has been set to @code{nopass}
20789 @xref{stepping into signal handlers}, for information on how stepping
20790 commands behave when the thread has a signal queued.
20793 @section Returning from a Function
20796 @cindex returning from a function
20799 @itemx return @var{expression}
20800 You can cancel execution of a function call with the @code{return}
20801 command. If you give an
20802 @var{expression} argument, its value is used as the function's return
20806 When you use @code{return}, @value{GDBN} discards the selected stack frame
20807 (and all frames within it). You can think of this as making the
20808 discarded frame return prematurely. If you wish to specify a value to
20809 be returned, give that value as the argument to @code{return}.
20811 This pops the selected stack frame (@pxref{Selection, ,Selecting a
20812 Frame}), and any other frames inside of it, leaving its caller as the
20813 innermost remaining frame. That frame becomes selected. The
20814 specified value is stored in the registers used for returning values
20817 The @code{return} command does not resume execution; it leaves the
20818 program stopped in the state that would exist if the function had just
20819 returned. In contrast, the @code{finish} command (@pxref{Continuing
20820 and Stepping, ,Continuing and Stepping}) resumes execution until the
20821 selected stack frame returns naturally.
20823 @value{GDBN} needs to know how the @var{expression} argument should be set for
20824 the inferior. The concrete registers assignment depends on the OS ABI and the
20825 type being returned by the selected stack frame. For example it is common for
20826 OS ABI to return floating point values in FPU registers while integer values in
20827 CPU registers. Still some ABIs return even floating point values in CPU
20828 registers. Larger integer widths (such as @code{long long int}) also have
20829 specific placement rules. @value{GDBN} already knows the OS ABI from its
20830 current target so it needs to find out also the type being returned to make the
20831 assignment into the right register(s).
20833 Normally, the selected stack frame has debug info. @value{GDBN} will always
20834 use the debug info instead of the implicit type of @var{expression} when the
20835 debug info is available. For example, if you type @kbd{return -1}, and the
20836 function in the current stack frame is declared to return a @code{long long
20837 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
20838 into a @code{long long int}:
20841 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
20843 (@value{GDBP}) return -1
20844 Make func return now? (y or n) y
20845 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
20846 43 printf ("result=%lld\n", func ());
20850 However, if the selected stack frame does not have a debug info, e.g., if the
20851 function was compiled without debug info, @value{GDBN} has to find out the type
20852 to return from user. Specifying a different type by mistake may set the value
20853 in different inferior registers than the caller code expects. For example,
20854 typing @kbd{return -1} with its implicit type @code{int} would set only a part
20855 of a @code{long long int} result for a debug info less function (on 32-bit
20856 architectures). Therefore the user is required to specify the return type by
20857 an appropriate cast explicitly:
20860 Breakpoint 2, 0x0040050b in func ()
20861 (@value{GDBP}) return -1
20862 Return value type not available for selected stack frame.
20863 Please use an explicit cast of the value to return.
20864 (@value{GDBP}) return (long long int) -1
20865 Make selected stack frame return now? (y or n) y
20866 #0 0x00400526 in main ()
20871 @section Calling Program Functions
20874 @cindex calling functions
20875 @cindex inferior functions, calling
20876 @item print @var{expr}
20877 Evaluate the expression @var{expr} and display the resulting value.
20878 The expression may include calls to functions in the program being
20882 @item call @var{expr}
20883 Evaluate the expression @var{expr} without displaying @code{void}
20886 You can use this variant of the @code{print} command if you want to
20887 execute a function from your program that does not return anything
20888 (a.k.a.@: @dfn{a void function}), but without cluttering the output
20889 with @code{void} returned values that @value{GDBN} will otherwise
20890 print. If the result is not void, it is printed and saved in the
20894 It is possible for the function you call via the @code{print} or
20895 @code{call} command to generate a signal (e.g., if there's a bug in
20896 the function, or if you passed it incorrect arguments). What happens
20897 in that case is controlled by the @code{set unwindonsignal} command.
20899 Similarly, with a C@t{++} program it is possible for the function you
20900 call via the @code{print} or @code{call} command to generate an
20901 exception that is not handled due to the constraints of the dummy
20902 frame. In this case, any exception that is raised in the frame, but has
20903 an out-of-frame exception handler will not be found. GDB builds a
20904 dummy-frame for the inferior function call, and the unwinder cannot
20905 seek for exception handlers outside of this dummy-frame. What happens
20906 in that case is controlled by the
20907 @code{set unwind-on-terminating-exception} command.
20909 @anchor{stack unwind settings}
20911 @item set unwindonsignal
20912 @kindex set unwindonsignal
20913 @cindex unwind stack in called functions
20914 @cindex call dummy stack unwinding
20915 Set unwinding of the stack if a signal is received while in a function
20916 that @value{GDBN} called in the program being debugged. If set to on,
20917 @value{GDBN} unwinds the stack it created for the call and restores
20918 the context to what it was before the call. If set to off (the
20919 default), @value{GDBN} stops in the frame where the signal was
20922 @item show unwindonsignal
20923 @kindex show unwindonsignal
20924 Show the current setting of stack unwinding in the functions called by
20927 @item set unwind-on-terminating-exception
20928 @kindex set unwind-on-terminating-exception
20929 @cindex unwind stack in called functions with unhandled exceptions
20930 @cindex call dummy stack unwinding on unhandled exception.
20931 Set unwinding of the stack if a C@t{++} exception is raised, but left
20932 unhandled while in a function that @value{GDBN} called in the program being
20933 debugged. If set to on (the default), @value{GDBN} unwinds the stack
20934 it created for the call and restores the context to what it was before
20935 the call. If set to off, @value{GDBN} the exception is delivered to
20936 the default C@t{++} exception handler and the inferior terminated.
20938 @item show unwind-on-terminating-exception
20939 @kindex show unwind-on-terminating-exception
20940 Show the current setting of stack unwinding in the functions called by
20943 @item set may-call-functions
20944 @kindex set may-call-functions
20945 @cindex disabling calling functions in the program
20946 @cindex calling functions in the program, disabling
20947 Set permission to call functions in the program.
20948 This controls whether @value{GDBN} will attempt to call functions in
20949 the program, such as with expressions in the @code{print} command. It
20950 defaults to @code{on}.
20952 To call a function in the program, @value{GDBN} has to temporarily
20953 modify the state of the inferior. This has potentially undesired side
20954 effects. Also, having @value{GDBN} call nested functions is likely to
20955 be erroneous and may even crash the program being debugged. You can
20956 avoid such hazards by forbidding @value{GDBN} from calling functions
20957 in the program being debugged. If calling functions in the program
20958 is forbidden, GDB will throw an error when a command (such as printing
20959 an expression) starts a function call in the program.
20961 @item show may-call-functions
20962 @kindex show may-call-functions
20963 Show permission to call functions in the program.
20967 When calling a function within a program, it is possible that the
20968 program could enter a state from which the called function may never
20969 return. If this happens then it is possible to interrupt the function
20970 call by typing the interrupt character (often @kbd{Ctrl-c}).
20972 If a called function is interrupted for any reason, including hitting
20973 a breakpoint, or triggering a watchpoint, and the stack is not unwound
20974 due to @code{set unwind-on-terminating-exception on} or @code{set
20975 unwindonsignal on} (@pxref{stack unwind settings}),
20976 then the dummy-frame, created by @value{GDBN} to facilitate the call
20977 to the program function, will be visible in the backtrace, for example
20978 frame @code{#3} in the following backtrace:
20981 (@value{GDBP}) backtrace
20982 #0 0x00007ffff7b3d1e7 in nanosleep () from /lib64/libc.so.6
20983 #1 0x00007ffff7b3d11e in sleep () from /lib64/libc.so.6
20984 #2 0x000000000040113f in deadlock () at test.cc:13
20985 #3 <function called from gdb>
20986 #4 breakpt () at test.cc:20
20987 #5 0x0000000000401151 in main () at test.cc:25
20990 At this point it is possible to examine the state of the inferior just
20991 like any other stop.
20993 Depending on why the function was interrupted then it may be possible
20994 to resume the inferior (using commands like @code{continue},
20995 @code{step}, etc). In this case, when the inferior finally returns to
20996 the dummy-frame, @value{GDBN} will once again halt the inferior.
20998 @subsection Calling functions with no debug info
21000 @cindex no debug info functions
21001 Sometimes, a function you wish to call is missing debug information.
21002 In such case, @value{GDBN} does not know the type of the function,
21003 including the types of the function's parameters. To avoid calling
21004 the inferior function incorrectly, which could result in the called
21005 function functioning erroneously and even crash, @value{GDBN} refuses
21006 to call the function unless you tell it the type of the function.
21008 For prototyped (i.e.@: ANSI/ISO style) functions, there are two ways
21009 to do that. The simplest is to cast the call to the function's
21010 declared return type. For example:
21013 (@value{GDBP}) p getenv ("PATH")
21014 'getenv' has unknown return type; cast the call to its declared return type
21015 (@value{GDBP}) p (char *) getenv ("PATH")
21016 $1 = 0x7fffffffe7ba "/usr/local/bin:/"...
21019 Casting the return type of a no-debug function is equivalent to
21020 casting the function to a pointer to a prototyped function that has a
21021 prototype that matches the types of the passed-in arguments, and
21022 calling that. I.e., the call above is equivalent to:
21025 (@value{GDBP}) p ((char * (*) (const char *)) getenv) ("PATH")
21029 and given this prototyped C or C++ function with float parameters:
21032 float multiply (float v1, float v2) @{ return v1 * v2; @}
21036 these calls are equivalent:
21039 (@value{GDBP}) p (float) multiply (2.0f, 3.0f)
21040 (@value{GDBP}) p ((float (*) (float, float)) multiply) (2.0f, 3.0f)
21043 If the function you wish to call is declared as unprototyped (i.e.@:
21044 old K&R style), you must use the cast-to-function-pointer syntax, so
21045 that @value{GDBN} knows that it needs to apply default argument
21046 promotions (promote float arguments to double). @xref{ABI, float
21047 promotion}. For example, given this unprototyped C function with
21048 float parameters, and no debug info:
21052 multiply_noproto (v1, v2)
21060 you call it like this:
21063 (@value{GDBP}) p ((float (*) ()) multiply_noproto) (2.0f, 3.0f)
21067 @section Patching Programs
21069 @cindex patching binaries
21070 @cindex writing into executables
21071 @cindex writing into corefiles
21073 By default, @value{GDBN} opens the file containing your program's
21074 executable code (or the corefile) read-only. This prevents accidental
21075 alterations to machine code; but it also prevents you from intentionally
21076 patching your program's binary.
21078 If you'd like to be able to patch the binary, you can specify that
21079 explicitly with the @code{set write} command. For example, you might
21080 want to turn on internal debugging flags, or even to make emergency
21086 @itemx set write off
21087 If you specify @samp{set write on}, @value{GDBN} opens executable and
21088 core files for both reading and writing; if you specify @kbd{set write
21089 off} (the default), @value{GDBN} opens them read-only.
21091 If you have already loaded a file, you must load it again (using the
21092 @code{exec-file} or @code{core-file} command) after changing @code{set
21093 write}, for your new setting to take effect.
21097 Display whether executable files and core files are opened for writing
21098 as well as reading.
21101 @node Compiling and Injecting Code
21102 @section Compiling and injecting code in @value{GDBN}
21103 @cindex injecting code
21104 @cindex writing into executables
21105 @cindex compiling code
21107 @value{GDBN} supports on-demand compilation and code injection into
21108 programs running under @value{GDBN}. GCC 5.0 or higher built with
21109 @file{libcc1.so} must be installed for this functionality to be enabled.
21110 This functionality is implemented with the following commands.
21113 @kindex compile code
21114 @item compile code @var{source-code}
21115 @itemx compile code -raw @var{--} @var{source-code}
21116 Compile @var{source-code} with the compiler language found as the current
21117 language in @value{GDBN} (@pxref{Languages}). If compilation and
21118 injection is not supported with the current language specified in
21119 @value{GDBN}, or the compiler does not support this feature, an error
21120 message will be printed. If @var{source-code} compiles and links
21121 successfully, @value{GDBN} will load the object-code emitted,
21122 and execute it within the context of the currently selected inferior.
21123 It is important to note that the compiled code is executed immediately.
21124 After execution, the compiled code is removed from @value{GDBN} and any
21125 new types or variables you have defined will be deleted.
21127 The command allows you to specify @var{source-code} in two ways.
21128 The simplest method is to provide a single line of code to the command.
21132 compile code printf ("hello world\n");
21135 If you specify options on the command line as well as source code, they
21136 may conflict. The @samp{--} delimiter can be used to separate options
21137 from actual source code. E.g.:
21140 compile code -r -- printf ("hello world\n");
21143 Alternatively you can enter source code as multiple lines of text. To
21144 enter this mode, invoke the @samp{compile code} command without any text
21145 following the command. This will start the multiple-line editor and
21146 allow you to type as many lines of source code as required. When you
21147 have completed typing, enter @samp{end} on its own line to exit the
21152 >printf ("hello\n");
21153 >printf ("world\n");
21157 Specifying @samp{-raw}, prohibits @value{GDBN} from wrapping the
21158 provided @var{source-code} in a callable scope. In this case, you must
21159 specify the entry point of the code by defining a function named
21160 @code{_gdb_expr_}. The @samp{-raw} code cannot access variables of the
21161 inferior. Using @samp{-raw} option may be needed for example when
21162 @var{source-code} requires @samp{#include} lines which may conflict with
21163 inferior symbols otherwise.
21165 @kindex compile file
21166 @item compile file @var{filename}
21167 @itemx compile file -raw @var{filename}
21168 Like @code{compile code}, but take the source code from @var{filename}.
21171 compile file /home/user/example.c
21176 @item compile print [[@var{options}] --] @var{expr}
21177 @itemx compile print [[@var{options}] --] /@var{f} @var{expr}
21178 Compile and execute @var{expr} with the compiler language found as the
21179 current language in @value{GDBN} (@pxref{Languages}). By default the
21180 value of @var{expr} is printed in a format appropriate to its data type;
21181 you can choose a different format by specifying @samp{/@var{f}}, where
21182 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
21183 Formats}. The @code{compile print} command accepts the same options
21184 as the @code{print} command; see @ref{print options}.
21186 @item compile print [[@var{options}] --]
21187 @itemx compile print [[@var{options}] --] /@var{f}
21188 @cindex reprint the last value
21189 Alternatively you can enter the expression (source code producing it) as
21190 multiple lines of text. To enter this mode, invoke the @samp{compile print}
21191 command without any text following the command. This will start the
21192 multiple-line editor.
21196 The process of compiling and injecting the code can be inspected using:
21199 @anchor{set debug compile}
21200 @item set debug compile
21201 @cindex compile command debugging info
21202 Turns on or off display of @value{GDBN} process of compiling and
21203 injecting the code. The default is off.
21205 @item show debug compile
21206 Displays the current state of displaying @value{GDBN} process of
21207 compiling and injecting the code.
21209 @anchor{set debug compile-cplus-types}
21210 @item set debug compile-cplus-types
21211 @cindex compile C@t{++} type conversion
21212 Turns on or off the display of C@t{++} type conversion debugging information.
21213 The default is off.
21215 @item show debug compile-cplus-types
21216 Displays the current state of displaying debugging information for
21217 C@t{++} type conversion.
21220 @subsection Compilation options for the @code{compile} command
21222 @value{GDBN} needs to specify the right compilation options for the code
21223 to be injected, in part to make its ABI compatible with the inferior
21224 and in part to make the injected code compatible with @value{GDBN}'s
21228 The options used, in increasing precedence:
21231 @item target architecture and OS options (@code{gdbarch})
21232 These options depend on target processor type and target operating
21233 system, usually they specify at least 32-bit (@code{-m32}) or 64-bit
21234 (@code{-m64}) compilation option.
21236 @item compilation options recorded in the target
21237 @value{NGCC} (since version 4.7) stores the options used for compilation
21238 into @code{DW_AT_producer} part of DWARF debugging information according
21239 to the @value{NGCC} option @code{-grecord-gcc-switches}. One has to
21240 explicitly specify @code{-g} during inferior compilation otherwise
21241 @value{NGCC} produces no DWARF. This feature is only relevant for
21242 platforms where @code{-g} produces DWARF by default, otherwise one may
21243 try to enforce DWARF by using @code{-gdwarf-4}.
21245 @item compilation options set by @code{set compile-args}
21249 You can override compilation options using the following command:
21252 @item set compile-args
21253 @cindex compile command options override
21254 Set compilation options used for compiling and injecting code with the
21255 @code{compile} commands. These options override any conflicting ones
21256 from the target architecture and/or options stored during inferior
21259 @item show compile-args
21260 Displays the current state of compilation options override.
21261 This does not show all the options actually used during compilation,
21262 use @ref{set debug compile} for that.
21265 @subsection Caveats when using the @code{compile} command
21267 There are a few caveats to keep in mind when using the @code{compile}
21268 command. As the caveats are different per language, the table below
21269 highlights specific issues on a per language basis.
21272 @item C code examples and caveats
21273 When the language in @value{GDBN} is set to @samp{C}, the compiler will
21274 attempt to compile the source code with a @samp{C} compiler. The source
21275 code provided to the @code{compile} command will have much the same
21276 access to variables and types as it normally would if it were part of
21277 the program currently being debugged in @value{GDBN}.
21279 Below is a sample program that forms the basis of the examples that
21280 follow. This program has been compiled and loaded into @value{GDBN},
21281 much like any other normal debugging session.
21284 void function1 (void)
21287 printf ("function 1\n");
21290 void function2 (void)
21305 For the purposes of the examples in this section, the program above has
21306 been compiled, loaded into @value{GDBN}, stopped at the function
21307 @code{main}, and @value{GDBN} is awaiting input from the user.
21309 To access variables and types for any program in @value{GDBN}, the
21310 program must be compiled and packaged with debug information. The
21311 @code{compile} command is not an exception to this rule. Without debug
21312 information, you can still use the @code{compile} command, but you will
21313 be very limited in what variables and types you can access.
21315 So with that in mind, the example above has been compiled with debug
21316 information enabled. The @code{compile} command will have access to
21317 all variables and types (except those that may have been optimized
21318 out). Currently, as @value{GDBN} has stopped the program in the
21319 @code{main} function, the @code{compile} command would have access to
21320 the variable @code{k}. You could invoke the @code{compile} command
21321 and type some source code to set the value of @code{k}. You can also
21322 read it, or do anything with that variable you would normally do in
21323 @code{C}. Be aware that changes to inferior variables in the
21324 @code{compile} command are persistent. In the following example:
21327 compile code k = 3;
21331 the variable @code{k} is now 3. It will retain that value until
21332 something else in the example program changes it, or another
21333 @code{compile} command changes it.
21335 Normal scope and access rules apply to source code compiled and
21336 injected by the @code{compile} command. In the example, the variables
21337 @code{j} and @code{k} are not accessible yet, because the program is
21338 currently stopped in the @code{main} function, where these variables
21339 are not in scope. Therefore, the following command
21342 compile code j = 3;
21346 will result in a compilation error message.
21348 Once the program is continued, execution will bring these variables in
21349 scope, and they will become accessible; then the code you specify via
21350 the @code{compile} command will be able to access them.
21352 You can create variables and types with the @code{compile} command as
21353 part of your source code. Variables and types that are created as part
21354 of the @code{compile} command are not visible to the rest of the program for
21355 the duration of its run. This example is valid:
21358 compile code int ff = 5; printf ("ff is %d\n", ff);
21361 However, if you were to type the following into @value{GDBN} after that
21362 command has completed:
21365 compile code printf ("ff is %d\n'', ff);
21369 a compiler error would be raised as the variable @code{ff} no longer
21370 exists. Object code generated and injected by the @code{compile}
21371 command is removed when its execution ends. Caution is advised
21372 when assigning to program variables values of variables created by the
21373 code submitted to the @code{compile} command. This example is valid:
21376 compile code int ff = 5; k = ff;
21379 The value of the variable @code{ff} is assigned to @code{k}. The variable
21380 @code{k} does not require the existence of @code{ff} to maintain the value
21381 it has been assigned. However, pointers require particular care in
21382 assignment. If the source code compiled with the @code{compile} command
21383 changed the address of a pointer in the example program, perhaps to a
21384 variable created in the @code{compile} command, that pointer would point
21385 to an invalid location when the command exits. The following example
21386 would likely cause issues with your debugged program:
21389 compile code int ff = 5; p = &ff;
21392 In this example, @code{p} would point to @code{ff} when the
21393 @code{compile} command is executing the source code provided to it.
21394 However, as variables in the (example) program persist with their
21395 assigned values, the variable @code{p} would point to an invalid
21396 location when the command exists. A general rule should be followed
21397 in that you should either assign @code{NULL} to any assigned pointers,
21398 or restore a valid location to the pointer before the command exits.
21400 Similar caution must be exercised with any structs, unions, and typedefs
21401 defined in @code{compile} command. Types defined in the @code{compile}
21402 command will no longer be available in the next @code{compile} command.
21403 Therefore, if you cast a variable to a type defined in the
21404 @code{compile} command, care must be taken to ensure that any future
21405 need to resolve the type can be achieved.
21408 (@value{GDBP}) compile code static struct a @{ int a; @} v = @{ 42 @}; argv = &v;
21409 (@value{GDBP}) compile code printf ("%d\n", ((struct a *) argv)->a);
21410 gdb command line:1:36: error: dereferencing pointer to incomplete type ‘struct a’
21411 Compilation failed.
21412 (@value{GDBP}) compile code struct a @{ int a; @}; printf ("%d\n", ((struct a *) argv)->a);
21416 Variables that have been optimized away by the compiler are not
21417 accessible to the code submitted to the @code{compile} command.
21418 Access to those variables will generate a compiler error which @value{GDBN}
21419 will print to the console.
21422 @subsection Compiler search for the @code{compile} command
21424 @value{GDBN} needs to find @value{NGCC} for the inferior being debugged
21425 which may not be obvious for remote targets of different architecture
21426 than where @value{GDBN} is running. Environment variable @env{PATH} on
21427 @value{GDBN} host is searched for @value{NGCC} binary matching the
21428 target architecture and operating system. This search can be overriden
21429 by @code{set compile-gcc} @value{GDBN} command below. @env{PATH} is
21430 taken from shell that executed @value{GDBN}, it is not the value set by
21431 @value{GDBN} command @code{set environment}). @xref{Environment}.
21434 Specifically @env{PATH} is searched for binaries matching regular expression
21435 @code{@var{arch}(-[^-]*)?-@var{os}-gcc} according to the inferior target being
21436 debugged. @var{arch} is processor name --- multiarch is supported, so for
21437 example both @code{i386} and @code{x86_64} targets look for pattern
21438 @code{(x86_64|i.86)} and both @code{s390} and @code{s390x} targets look
21439 for pattern @code{s390x?}. @var{os} is currently supported only for
21440 pattern @code{linux(-gnu)?}.
21442 On Posix hosts the compiler driver @value{GDBN} needs to find also
21443 shared library @file{libcc1.so} from the compiler. It is searched in
21444 default shared library search path (overridable with usual environment
21445 variable @env{LD_LIBRARY_PATH}), unrelated to @env{PATH} or @code{set
21446 compile-gcc} settings. Contrary to it @file{libcc1plugin.so} is found
21447 according to the installation of the found compiler --- as possibly
21448 specified by the @code{set compile-gcc} command.
21451 @item set compile-gcc
21452 @cindex compile command driver filename override
21453 Set compilation command used for compiling and injecting code with the
21454 @code{compile} commands. If this option is not set (it is set to
21455 an empty string), the search described above will occur --- that is the
21458 @item show compile-gcc
21459 Displays the current compile command @value{NGCC} driver filename.
21460 If set, it is the main command @command{gcc}, found usually for example
21461 under name @file{x86_64-linux-gnu-gcc}.
21465 @chapter @value{GDBN} Files
21467 @value{GDBN} needs to know the file name of the program to be debugged,
21468 both in order to read its symbol table and in order to start your
21469 program. To debug a core dump of a previous run, you must also tell
21470 @value{GDBN} the name of the core dump file.
21473 * Files:: Commands to specify files
21474 * File Caching:: Information about @value{GDBN}'s file caching
21475 * Separate Debug Files:: Debugging information in separate files
21476 * MiniDebugInfo:: Debugging information in a special section
21477 * Index Files:: Index files speed up GDB
21478 * Symbol Errors:: Errors reading symbol files
21479 * Data Files:: GDB data files
21483 @section Commands to Specify Files
21485 @cindex symbol table
21486 @cindex core dump file
21488 You may want to specify executable and core dump file names. The usual
21489 way to do this is at start-up time, using the arguments to
21490 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
21491 Out of @value{GDBN}}).
21493 Occasionally it is necessary to change to a different file during a
21494 @value{GDBN} session. Or you may run @value{GDBN} and forget to
21495 specify a file you want to use. Or you are debugging a remote target
21496 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
21497 Program}). In these situations the @value{GDBN} commands to specify
21498 new files are useful.
21501 @cindex executable file
21503 @item file @var{filename}
21504 Use @var{filename} as the program to be debugged. It is read for its
21505 symbols and for the contents of pure memory. It is also the program
21506 executed when you use the @code{run} command. If you do not specify a
21507 directory and the file is not found in the @value{GDBN} working directory,
21508 @value{GDBN} uses the environment variable @env{PATH} as a list of
21509 directories to search, just as the shell does when looking for a program
21510 to run. You can change the value of this variable, for both @value{GDBN}
21511 and your program, using the @code{path} command.
21513 @cindex unlinked object files
21514 @cindex patching object files
21515 You can load unlinked object @file{.o} files into @value{GDBN} using
21516 the @code{file} command. You will not be able to ``run'' an object
21517 file, but you can disassemble functions and inspect variables. Also,
21518 if the underlying BFD functionality supports it, you could use
21519 @kbd{gdb -write} to patch object files using this technique. Note
21520 that @value{GDBN} can neither interpret nor modify relocations in this
21521 case, so branches and some initialized variables will appear to go to
21522 the wrong place. But this feature is still handy from time to time.
21525 @code{file} with no argument makes @value{GDBN} discard any information it
21526 has on both executable file and the symbol table.
21529 @item exec-file @r{[} @var{filename} @r{]}
21530 Specify that the program to be run (but not the symbol table) is found
21531 in @var{filename}. @value{GDBN} searches the environment variable @env{PATH}
21532 if necessary to locate your program. Omitting @var{filename} means to
21533 discard information on the executable file.
21535 @kindex symbol-file
21536 @item symbol-file @r{[} @var{filename} @r{[} -o @var{offset} @r{]]}
21537 Read symbol table information from file @var{filename}. @env{PATH} is
21538 searched when necessary. Use the @code{file} command to get both symbol
21539 table and program to run from the same file.
21541 If an optional @var{offset} is specified, it is added to the start
21542 address of each section in the symbol file. This is useful if the
21543 program is relocated at runtime, such as the Linux kernel with kASLR
21546 @code{symbol-file} with no argument clears out @value{GDBN} information on your
21547 program's symbol table.
21549 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
21550 some breakpoints and auto-display expressions. This is because they may
21551 contain pointers to the internal data recording symbols and data types,
21552 which are part of the old symbol table data being discarded inside
21555 @code{symbol-file} does not repeat if you press @key{RET} again after
21558 When @value{GDBN} is configured for a particular environment, it
21559 understands debugging information in whatever format is the standard
21560 generated for that environment; you may use either a @sc{gnu} compiler, or
21561 other compilers that adhere to the local conventions.
21562 Best results are usually obtained from @sc{gnu} compilers; for example,
21563 using @code{@value{NGCC}} you can generate debugging information for
21566 For most kinds of object files, with the exception of old SVR3 systems
21567 using COFF, the @code{symbol-file} command does not normally read the
21568 symbol table in full right away. Instead, it scans the symbol table
21569 quickly to find which source files and which symbols are present. The
21570 details are read later, one source file at a time, as they are needed.
21572 The purpose of this two-stage reading strategy is to make @value{GDBN}
21573 start up faster. For the most part, it is invisible except for
21574 occasional pauses while the symbol table details for a particular source
21575 file are being read. (The @code{set verbose} command can turn these
21576 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
21577 Warnings and Messages}.)
21579 We have not implemented the two-stage strategy for COFF yet. When the
21580 symbol table is stored in COFF format, @code{symbol-file} reads the
21581 symbol table data in full right away. Note that ``stabs-in-COFF''
21582 still does the two-stage strategy, since the debug info is actually
21586 @cindex reading symbols immediately
21587 @cindex symbols, reading immediately
21588 @item symbol-file @r{[} -readnow @r{]} @var{filename}
21589 @itemx file @r{[} -readnow @r{]} @var{filename}
21590 You can override the @value{GDBN} two-stage strategy for reading symbol
21591 tables by using the @samp{-readnow} option with any of the commands that
21592 load symbol table information, if you want to be sure @value{GDBN} has the
21593 entire symbol table available.
21595 @cindex @code{-readnever}, option for symbol-file command
21596 @cindex never read symbols
21597 @cindex symbols, never read
21598 @item symbol-file @r{[} -readnever @r{]} @var{filename}
21599 @itemx file @r{[} -readnever @r{]} @var{filename}
21600 You can instruct @value{GDBN} to never read the symbolic information
21601 contained in @var{filename} by using the @samp{-readnever} option.
21602 @xref{--readnever}.
21604 @c FIXME: for now no mention of directories, since this seems to be in
21605 @c flux. 13mar1992 status is that in theory GDB would look either in
21606 @c current dir or in same dir as myprog; but issues like competing
21607 @c GDB's, or clutter in system dirs, mean that in practice right now
21608 @c only current dir is used. FFish says maybe a special GDB hierarchy
21609 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
21613 @item core-file @r{[}@var{filename}@r{]}
21615 Specify the whereabouts of a core dump file to be used as the ``contents
21616 of memory''. Traditionally, core files contain only some parts of the
21617 address space of the process that generated them; @value{GDBN} can access the
21618 executable file itself for other parts.
21620 @code{core-file} with no argument specifies that no core file is
21623 Note that the core file is ignored when your program is actually running
21624 under @value{GDBN}. So, if you have been running your program and you
21625 wish to debug a core file instead, you must kill the subprocess in which
21626 the program is running. To do this, use the @code{kill} command
21627 (@pxref{Kill Process, ,Killing the Child Process}).
21629 @kindex add-symbol-file
21630 @cindex dynamic linking
21631 @item add-symbol-file @var{filename} @r{[} -readnow @r{|} -readnever @r{]} @r{[} -o @var{offset} @r{]} @r{[} @var{textaddress} @r{]} @r{[} -s @var{section} @var{address} @dots{} @r{]}
21632 The @code{add-symbol-file} command reads additional symbol table
21633 information from the file @var{filename}. You would use this command
21634 when @var{filename} has been dynamically loaded (by some other means)
21635 into the program that is running. The @var{textaddress} parameter gives
21636 the memory address at which the file's text section has been loaded.
21637 You can additionally specify the base address of other sections using
21638 an arbitrary number of @samp{-s @var{section} @var{address}} pairs.
21639 If a section is omitted, @value{GDBN} will use its default addresses
21640 as found in @var{filename}. Any @var{address} or @var{textaddress}
21641 can be given as an expression.
21643 If an optional @var{offset} is specified, it is added to the start
21644 address of each section, except those for which the address was
21645 specified explicitly.
21647 The symbol table of the file @var{filename} is added to the symbol table
21648 originally read with the @code{symbol-file} command. You can use the
21649 @code{add-symbol-file} command any number of times; the new symbol data
21650 thus read is kept in addition to the old.
21652 Changes can be reverted using the command @code{remove-symbol-file}.
21654 @cindex relocatable object files, reading symbols from
21655 @cindex object files, relocatable, reading symbols from
21656 @cindex reading symbols from relocatable object files
21657 @cindex symbols, reading from relocatable object files
21658 @cindex @file{.o} files, reading symbols from
21659 Although @var{filename} is typically a shared library file, an
21660 executable file, or some other object file which has been fully
21661 relocated for loading into a process, you can also load symbolic
21662 information from relocatable @file{.o} files, as long as:
21666 the file's symbolic information refers only to linker symbols defined in
21667 that file, not to symbols defined by other object files,
21669 every section the file's symbolic information refers to has actually
21670 been loaded into the inferior, as it appears in the file, and
21672 you can determine the address at which every section was loaded, and
21673 provide these to the @code{add-symbol-file} command.
21677 Some embedded operating systems, like Sun Chorus and VxWorks, can load
21678 relocatable files into an already running program; such systems
21679 typically make the requirements above easy to meet. However, it's
21680 important to recognize that many native systems use complex link
21681 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
21682 assembly, for example) that make the requirements difficult to meet. In
21683 general, one cannot assume that using @code{add-symbol-file} to read a
21684 relocatable object file's symbolic information will have the same effect
21685 as linking the relocatable object file into the program in the normal
21688 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
21690 @kindex remove-symbol-file
21691 @item remove-symbol-file @var{filename}
21692 @item remove-symbol-file -a @var{address}
21693 Remove a symbol file added via the @code{add-symbol-file} command. The
21694 file to remove can be identified by its @var{filename} or by an @var{address}
21695 that lies within the boundaries of this symbol file in memory. Example:
21698 (@value{GDBP}) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
21699 add symbol table from file "/home/user/gdb/mylib.so" at
21700 .text_addr = 0x7ffff7ff9480
21702 Reading symbols from /home/user/gdb/mylib.so...
21703 (@value{GDBP}) remove-symbol-file -a 0x7ffff7ff9480
21704 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
21709 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
21711 @kindex add-symbol-file-from-memory
21712 @cindex @code{syscall DSO}
21713 @cindex load symbols from memory
21714 @item add-symbol-file-from-memory @var{address}
21715 Load symbols from the given @var{address} in a dynamically loaded
21716 object file whose image is mapped directly into the inferior's memory.
21717 For example, the Linux kernel maps a @code{syscall DSO} into each
21718 process's address space; this DSO provides kernel-specific code for
21719 some system calls. The argument can be any expression whose
21720 evaluation yields the address of the file's shared object file header.
21721 For this command to work, you must have used @code{symbol-file} or
21722 @code{exec-file} commands in advance.
21725 @item section @var{section} @var{addr}
21726 The @code{section} command changes the base address of the named
21727 @var{section} of the exec file to @var{addr}. This can be used if the
21728 exec file does not contain section addresses, (such as in the
21729 @code{a.out} format), or when the addresses specified in the file
21730 itself are wrong. Each section must be changed separately. The
21731 @code{info files} command, described below, lists all the sections and
21735 @kindex info target
21738 @code{info files} and @code{info target} are synonymous; both print the
21739 current target (@pxref{Targets, ,Specifying a Debugging Target}),
21740 including the names of the executable and core dump files currently in
21741 use by @value{GDBN}, and the files from which symbols were loaded. The
21742 command @code{help target} lists all possible targets rather than
21745 @kindex maint info sections
21746 @item maint info sections @r{[}-all-objects@r{]} @r{[}@var{filter-list}@r{]}
21747 Another command that can give you extra information about program sections
21748 is @code{maint info sections}. In addition to the section information
21749 displayed by @code{info files}, this command displays the flags and file
21750 offset of each section in the executable and core dump files.
21752 When @samp{-all-objects} is passed then sections from all loaded object
21753 files, including shared libraries, are printed.
21755 The optional @var{filter-list} is a space separated list of filter
21756 keywords. Sections that match any one of the filter criteria will be
21757 printed. There are two types of filter:
21760 @item @var{section-name}
21761 Display information about any section named @var{section-name}.
21762 @item @var{section-flag}
21763 Display information for any section with @var{section-flag}. The
21764 section flags that @value{GDBN} currently knows about are:
21767 Section will have space allocated in the process when loaded.
21768 Set for all sections except those containing debug information.
21770 Section will be loaded from the file into the child process memory.
21771 Set for pre-initialized code and data, clear for @code{.bss} sections.
21773 Section needs to be relocated before loading.
21775 Section cannot be modified by the child process.
21777 Section contains executable code only.
21779 Section contains data only (no executable code).
21781 Section will reside in ROM.
21783 Section contains data for constructor/destructor lists.
21785 Section is not empty.
21787 An instruction to the linker to not output the section.
21788 @item COFF_SHARED_LIBRARY
21789 A notification to the linker that the section contains
21790 COFF shared library information.
21792 Section contains common symbols.
21796 @kindex maint info target-sections
21797 @item maint info target-sections
21798 This command prints @value{GDBN}'s internal section table. For each
21799 target @value{GDBN} maintains a table containing the allocatable
21800 sections from all currently mapped objects, along with information
21801 about where the section is mapped.
21803 @kindex set trust-readonly-sections
21804 @cindex read-only sections
21805 @item set trust-readonly-sections on
21806 Tell @value{GDBN} that readonly sections in your object file
21807 really are read-only (i.e.@: that their contents will not change).
21808 In that case, @value{GDBN} can fetch values from these sections
21809 out of the object file, rather than from the target program.
21810 For some targets (notably embedded ones), this can be a significant
21811 enhancement to debugging performance.
21813 The default is off.
21815 @item set trust-readonly-sections off
21816 Tell @value{GDBN} not to trust readonly sections. This means that
21817 the contents of the section might change while the program is running,
21818 and must therefore be fetched from the target when needed.
21820 @item show trust-readonly-sections
21821 Show the current setting of trusting readonly sections.
21824 All file-specifying commands allow both absolute and relative file names
21825 as arguments. @value{GDBN} always converts the file name to an absolute file
21826 name and remembers it that way.
21828 @cindex shared libraries
21829 @anchor{Shared Libraries}
21830 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, SunOS,
21831 Darwin/Mach-O, SVr4, IBM RS/6000 AIX, QNX Neutrino, FDPIC (FR-V), and
21832 DSBT (TIC6X) shared libraries.
21834 On MS-Windows @value{GDBN} must be linked with the Expat library to support
21835 shared libraries. @xref{Expat}.
21837 @value{GDBN} automatically loads symbol definitions from shared libraries
21838 when you use the @code{run} command, or when you examine a core file.
21839 (Before you issue the @code{run} command, @value{GDBN} does not understand
21840 references to a function in a shared library, however---unless you are
21841 debugging a core file).
21843 @c FIXME: some @value{GDBN} release may permit some refs to undef
21844 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
21845 @c FIXME...lib; check this from time to time when updating manual
21847 There are times, however, when you may wish to not automatically load
21848 symbol definitions from shared libraries, such as when they are
21849 particularly large or there are many of them.
21851 To control the automatic loading of shared library symbols, use the
21855 @kindex set auto-solib-add
21856 @item set auto-solib-add @var{mode}
21857 If @var{mode} is @code{on}, symbols from all shared object libraries
21858 will be loaded automatically when the inferior begins execution, you
21859 attach to an independently started inferior, or when the dynamic linker
21860 informs @value{GDBN} that a new library has been loaded. If @var{mode}
21861 is @code{off}, symbols must be loaded manually, using the
21862 @code{sharedlibrary} command. The default value is @code{on}.
21864 @cindex memory used for symbol tables
21865 If your program uses lots of shared libraries with debug info that
21866 takes large amounts of memory, you can decrease the @value{GDBN}
21867 memory footprint by preventing it from automatically loading the
21868 symbols from shared libraries. To that end, type @kbd{set
21869 auto-solib-add off} before running the inferior, then load each
21870 library whose debug symbols you do need with @kbd{sharedlibrary
21871 @var{regexp}}, where @var{regexp} is a regular expression that matches
21872 the libraries whose symbols you want to be loaded.
21874 @kindex show auto-solib-add
21875 @item show auto-solib-add
21876 Display the current autoloading mode.
21879 @cindex load shared library
21880 To explicitly load shared library symbols, use the @code{sharedlibrary}
21884 @kindex info sharedlibrary
21886 @item info share @var{regex}
21887 @itemx info sharedlibrary @var{regex}
21888 Print the names of the shared libraries which are currently loaded
21889 that match @var{regex}. If @var{regex} is omitted then print
21890 all shared libraries that are loaded.
21893 @item info dll @var{regex}
21894 This is an alias of @code{info sharedlibrary}.
21896 @kindex sharedlibrary
21898 @item sharedlibrary @var{regex}
21899 @itemx share @var{regex}
21900 Load shared object library symbols for files matching a
21901 Unix regular expression.
21902 As with files loaded automatically, it only loads shared libraries
21903 required by your program for a core file or after typing @code{run}. If
21904 @var{regex} is omitted all shared libraries required by your program are
21907 @item nosharedlibrary
21908 @kindex nosharedlibrary
21909 @cindex unload symbols from shared libraries
21910 Unload all shared object library symbols. This discards all symbols
21911 that have been loaded from all shared libraries. Symbols from shared
21912 libraries that were loaded by explicit user requests are not
21916 Sometimes you may wish that @value{GDBN} stops and gives you control
21917 when any of shared library events happen. The best way to do this is
21918 to use @code{catch load} and @code{catch unload} (@pxref{Set
21921 @value{GDBN} also supports the @code{set stop-on-solib-events}
21922 command for this. This command exists for historical reasons. It is
21923 less useful than setting a catchpoint, because it does not allow for
21924 conditions or commands as a catchpoint does.
21927 @item set stop-on-solib-events
21928 @kindex set stop-on-solib-events
21929 This command controls whether @value{GDBN} should give you control
21930 when the dynamic linker notifies it about some shared library event.
21931 The most common event of interest is loading or unloading of a new
21934 @item show stop-on-solib-events
21935 @kindex show stop-on-solib-events
21936 Show whether @value{GDBN} stops and gives you control when shared
21937 library events happen.
21940 Shared libraries are also supported in many cross or remote debugging
21941 configurations. @value{GDBN} needs to have access to the target's libraries;
21942 this can be accomplished either by providing copies of the libraries
21943 on the host system, or by asking @value{GDBN} to automatically retrieve the
21944 libraries from the target. If copies of the target libraries are
21945 provided, they need to be the same as the target libraries, although the
21946 copies on the target can be stripped as long as the copies on the host are
21949 @cindex where to look for shared libraries
21950 For remote debugging, you need to tell @value{GDBN} where the target
21951 libraries are, so that it can load the correct copies---otherwise, it
21952 may try to load the host's libraries. @value{GDBN} has two variables
21953 to specify the search directories for target libraries.
21956 @cindex prefix for executable and shared library file names
21957 @cindex system root, alternate
21958 @kindex set solib-absolute-prefix
21959 @kindex set sysroot
21960 @item set sysroot @var{path}
21961 Use @var{path} as the system root for the program being debugged. Any
21962 absolute shared library paths will be prefixed with @var{path}; many
21963 runtime loaders store the absolute paths to the shared library in the
21964 target program's memory. When starting processes remotely, and when
21965 attaching to already-running processes (local or remote), their
21966 executable filenames will be prefixed with @var{path} if reported to
21967 @value{GDBN} as absolute by the operating system. If you use
21968 @code{set sysroot} to find executables and shared libraries, they need
21969 to be laid out in the same way that they are on the target, with
21970 e.g.@: a @file{/bin}, @file{/lib} and @file{/usr/lib} hierarchy under
21973 If @var{path} starts with the sequence @file{target:} and the target
21974 system is remote then @value{GDBN} will retrieve the target binaries
21975 from the remote system. This is only supported when using a remote
21976 target that supports the @code{remote get} command (@pxref{File
21977 Transfer,,Sending files to a remote system}). The part of @var{path}
21978 following the initial @file{target:} (if present) is used as system
21979 root prefix on the remote file system. If @var{path} starts with the
21980 sequence @file{remote:} this is converted to the sequence
21981 @file{target:} by @code{set sysroot}@footnote{Historically the
21982 functionality to retrieve binaries from the remote system was
21983 provided by prefixing @var{path} with @file{remote:}}. If you want
21984 to specify a local system root using a directory that happens to be
21985 named @file{target:} or @file{remote:}, you need to use some
21986 equivalent variant of the name like @file{./target:}.
21988 For targets with an MS-DOS based filesystem, such as MS-Windows,
21989 @value{GDBN} tries prefixing a few variants of the target
21990 absolute file name with @var{path}. But first, on Unix hosts,
21991 @value{GDBN} converts all backslash directory separators into forward
21992 slashes, because the backslash is not a directory separator on Unix:
21995 c:\foo\bar.dll @result{} c:/foo/bar.dll
21998 Then, @value{GDBN} attempts prefixing the target file name with
21999 @var{path}, and looks for the resulting file name in the host file
22003 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
22006 If that does not find the binary, @value{GDBN} tries removing
22007 the @samp{:} character from the drive spec, both for convenience, and,
22008 for the case of the host file system not supporting file names with
22012 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
22015 This makes it possible to have a system root that mirrors a target
22016 with more than one drive. E.g., you may want to setup your local
22017 copies of the target system shared libraries like so (note @samp{c} vs
22021 @file{/path/to/sysroot/c/sys/bin/foo.dll}
22022 @file{/path/to/sysroot/c/sys/bin/bar.dll}
22023 @file{/path/to/sysroot/z/sys/bin/bar.dll}
22027 and point the system root at @file{/path/to/sysroot}, so that
22028 @value{GDBN} can find the correct copies of both
22029 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
22031 If that still does not find the binary, @value{GDBN} tries
22032 removing the whole drive spec from the target file name:
22035 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
22038 This last lookup makes it possible to not care about the drive name,
22039 if you don't want or need to.
22041 The @code{set solib-absolute-prefix} command is an alias for @code{set
22044 @cindex default system root
22045 @cindex @samp{--with-sysroot}
22046 You can set the default system root by using the configure-time
22047 @samp{--with-sysroot} option. If the system root is inside
22048 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
22049 @samp{--exec-prefix}), then the default system root will be updated
22050 automatically if the installed @value{GDBN} is moved to a new
22053 @kindex show sysroot
22055 Display the current executable and shared library prefix.
22057 @kindex set solib-search-path
22058 @item set solib-search-path @var{path}
22059 If this variable is set, @var{path} is a colon-separated list of
22060 directories to search for shared libraries. @samp{solib-search-path}
22061 is used after @samp{sysroot} fails to locate the library, or if the
22062 path to the library is relative instead of absolute. If you want to
22063 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
22064 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
22065 finding your host's libraries. @samp{sysroot} is preferred; setting
22066 it to a nonexistent directory may interfere with automatic loading
22067 of shared library symbols.
22069 @kindex show solib-search-path
22070 @item show solib-search-path
22071 Display the current shared library search path.
22073 @cindex DOS file-name semantics of file names.
22074 @kindex set target-file-system-kind (unix|dos-based|auto)
22075 @kindex show target-file-system-kind
22076 @item set target-file-system-kind @var{kind}
22077 Set assumed file system kind for target reported file names.
22079 Shared library file names as reported by the target system may not
22080 make sense as is on the system @value{GDBN} is running on. For
22081 example, when remote debugging a target that has MS-DOS based file
22082 system semantics, from a Unix host, the target may be reporting to
22083 @value{GDBN} a list of loaded shared libraries with file names such as
22084 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
22085 drive letters, so the @samp{c:\} prefix is not normally understood as
22086 indicating an absolute file name, and neither is the backslash
22087 normally considered a directory separator character. In that case,
22088 the native file system would interpret this whole absolute file name
22089 as a relative file name with no directory components. This would make
22090 it impossible to point @value{GDBN} at a copy of the remote target's
22091 shared libraries on the host using @code{set sysroot}, and impractical
22092 with @code{set solib-search-path}. Setting
22093 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
22094 to interpret such file names similarly to how the target would, and to
22095 map them to file names valid on @value{GDBN}'s native file system
22096 semantics. The value of @var{kind} can be @code{"auto"}, in addition
22097 to one of the supported file system kinds. In that case, @value{GDBN}
22098 tries to determine the appropriate file system variant based on the
22099 current target's operating system (@pxref{ABI, ,Configuring the
22100 Current ABI}). The supported file system settings are:
22104 Instruct @value{GDBN} to assume the target file system is of Unix
22105 kind. Only file names starting the forward slash (@samp{/}) character
22106 are considered absolute, and the directory separator character is also
22110 Instruct @value{GDBN} to assume the target file system is DOS based.
22111 File names starting with either a forward slash, or a drive letter
22112 followed by a colon (e.g., @samp{c:}), are considered absolute, and
22113 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
22114 considered directory separators.
22117 Instruct @value{GDBN} to use the file system kind associated with the
22118 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
22119 This is the default.
22123 @cindex file name canonicalization
22124 @cindex base name differences
22125 When processing file names provided by the user, @value{GDBN}
22126 frequently needs to compare them to the file names recorded in the
22127 program's debug info. Normally, @value{GDBN} compares just the
22128 @dfn{base names} of the files as strings, which is reasonably fast
22129 even for very large programs. (The base name of a file is the last
22130 portion of its name, after stripping all the leading directories.)
22131 This shortcut in comparison is based upon the assumption that files
22132 cannot have more than one base name. This is usually true, but
22133 references to files that use symlinks or similar filesystem
22134 facilities violate that assumption. If your program records files
22135 using such facilities, or if you provide file names to @value{GDBN}
22136 using symlinks etc., you can set @code{basenames-may-differ} to
22137 @code{true} to instruct @value{GDBN} to completely canonicalize each
22138 pair of file names it needs to compare. This will make file-name
22139 comparisons accurate, but at a price of a significant slowdown.
22142 @item set basenames-may-differ
22143 @kindex set basenames-may-differ
22144 Set whether a source file may have multiple base names.
22146 @item show basenames-may-differ
22147 @kindex show basenames-may-differ
22148 Show whether a source file may have multiple base names.
22152 @section File Caching
22153 @cindex caching of opened files
22154 @cindex caching of bfd objects
22156 To speed up file loading, and reduce memory usage, @value{GDBN} will
22157 reuse the @code{bfd} objects used to track open files. @xref{Top, ,
22158 BFD, bfd, The Binary File Descriptor Library}. The following commands
22159 allow visibility and control of the caching behavior.
22162 @kindex maint info bfds
22163 @item maint info bfds
22164 This prints information about each @code{bfd} object that is known to
22167 @kindex maint set bfd-sharing
22168 @kindex maint show bfd-sharing
22169 @kindex bfd caching
22170 @item maint set bfd-sharing
22171 @item maint show bfd-sharing
22172 Control whether @code{bfd} objects can be shared. When sharing is
22173 enabled @value{GDBN} reuses already open @code{bfd} objects rather
22174 than reopening the same file. Turning sharing off does not cause
22175 already shared @code{bfd} objects to be unshared, but all future files
22176 that are opened will create a new @code{bfd} object. Similarly,
22177 re-enabling sharing does not cause multiple existing @code{bfd}
22178 objects to be collapsed into a single shared @code{bfd} object.
22180 @kindex set debug bfd-cache @var{level}
22181 @kindex bfd caching
22182 @item set debug bfd-cache @var{level}
22183 Turns on debugging of the bfd cache, setting the level to @var{level}.
22185 @kindex show debug bfd-cache
22186 @kindex bfd caching
22187 @item show debug bfd-cache
22188 Show the current debugging level of the bfd cache.
22191 @node Separate Debug Files
22192 @section Debugging Information in Separate Files
22193 @cindex separate debugging information files
22194 @cindex debugging information in separate files
22195 @cindex @file{.debug} subdirectories
22196 @cindex debugging information directory, global
22197 @cindex global debugging information directories
22198 @cindex build ID, and separate debugging files
22199 @cindex @file{.build-id} directory
22201 @value{GDBN} allows you to put a program's debugging information in a
22202 file separate from the executable itself, in a way that allows
22203 @value{GDBN} to find and load the debugging information automatically.
22204 Since debugging information can be very large---sometimes larger
22205 than the executable code itself---some systems distribute debugging
22206 information for their executables in separate files, which users can
22207 install only when they need to debug a problem.
22209 @value{GDBN} supports two ways of specifying the separate debug info
22214 The executable contains a @dfn{debug link} that specifies the name of
22215 the separate debug info file. The separate debug file's name is
22216 usually @file{@var{executable}.debug}, where @var{executable} is the
22217 name of the corresponding executable file without leading directories
22218 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
22219 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
22220 checksum for the debug file, which @value{GDBN} uses to validate that
22221 the executable and the debug file came from the same build.
22225 The executable contains a @dfn{build ID}, a unique bit string that is
22226 also present in the corresponding debug info file. (This is supported
22227 only on some operating systems, when using the ELF or PE file formats
22228 for binary files and the @sc{gnu} Binutils.) For more details about
22229 this feature, see the description of the @option{--build-id}
22230 command-line option in @ref{Options, , Command Line Options, ld,
22231 The GNU Linker}. The debug info file's name is not specified
22232 explicitly by the build ID, but can be computed from the build ID, see
22236 Depending on the way the debug info file is specified, @value{GDBN}
22237 uses two different methods of looking for the debug file:
22241 For the ``debug link'' method, @value{GDBN} looks up the named file in
22242 the directory of the executable file, then in a subdirectory of that
22243 directory named @file{.debug}, and finally under each one of the
22244 global debug directories, in a subdirectory whose name is identical to
22245 the leading directories of the executable's absolute file name. (On
22246 MS-Windows/MS-DOS, the drive letter of the executable's leading
22247 directories is converted to a one-letter subdirectory, i.e.@:
22248 @file{d:/usr/bin/} is converted to @file{/d/usr/bin/}, because Windows
22249 filesystems disallow colons in file names.)
22252 For the ``build ID'' method, @value{GDBN} looks in the
22253 @file{.build-id} subdirectory of each one of the global debug directories for
22254 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
22255 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
22256 are the rest of the bit string. (Real build ID strings are 32 or more
22257 hex characters, not 10.) @value{GDBN} can automatically query
22258 @code{debuginfod} servers using build IDs in order to download separate debug
22259 files that cannot be found locally. For more information see @ref{Debuginfod}.
22262 So, for example, suppose you ask @value{GDBN} to debug
22263 @file{/usr/bin/ls}, which has a debug link that specifies the
22264 file @file{ls.debug}, and a build ID whose value in hex is
22265 @code{abcdef1234}. If the list of the global debug directories includes
22266 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
22267 debug information files, in the indicated order:
22271 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
22273 @file{/usr/bin/ls.debug}
22275 @file{/usr/bin/.debug/ls.debug}
22277 @file{/usr/lib/debug/usr/bin/ls.debug}.
22280 If the debug file still has not been found and @code{debuginfod}
22281 (@pxref{Debuginfod}) is enabled, @value{GDBN} will attempt to download the
22282 file from @code{debuginfod} servers.
22284 @anchor{debug-file-directory}
22285 Global debugging info directories default to what is set by @value{GDBN}
22286 configure option @option{--with-separate-debug-dir} and augmented by the
22287 colon-separated list of directories provided via @value{GDBN} configure
22288 option @option{--additional-debug-dirs}. During @value{GDBN} run you can
22289 also set the global debugging info directories, and view the list
22290 @value{GDBN} is currently using.
22294 @kindex set debug-file-directory
22295 @item set debug-file-directory @var{directories}
22296 Set the directories which @value{GDBN} searches for separate debugging
22297 information files to @var{directory}. Multiple path components can be set
22298 concatenating them by a path separator.
22300 @kindex show debug-file-directory
22301 @item show debug-file-directory
22302 Show the directories @value{GDBN} searches for separate debugging
22307 @cindex @code{.gnu_debuglink} sections
22308 @cindex debug link sections
22309 A debug link is a special section of the executable file named
22310 @code{.gnu_debuglink}. The section must contain:
22314 A filename, with any leading directory components removed, followed by
22317 zero to three bytes of padding, as needed to reach the next four-byte
22318 boundary within the section, and
22320 a four-byte CRC checksum, stored in the same endianness used for the
22321 executable file itself. The checksum is computed on the debugging
22322 information file's full contents by the function given below, passing
22323 zero as the @var{crc} argument.
22326 Any executable file format can carry a debug link, as long as it can
22327 contain a section named @code{.gnu_debuglink} with the contents
22330 @cindex @code{.note.gnu.build-id} sections
22331 @cindex build ID sections
22332 The build ID is a special section in the executable file (and in other
22333 ELF binary files that @value{GDBN} may consider). This section is
22334 often named @code{.note.gnu.build-id}, but that name is not mandatory.
22335 It contains unique identification for the built files---the ID remains
22336 the same across multiple builds of the same build tree. The default
22337 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
22338 content for the build ID string. The same section with an identical
22339 value is present in the original built binary with symbols, in its
22340 stripped variant, and in the separate debugging information file.
22342 The debugging information file itself should be an ordinary
22343 executable, containing a full set of linker symbols, sections, and
22344 debugging information. The sections of the debugging information file
22345 should have the same names, addresses, and sizes as the original file,
22346 but they need not contain any data---much like a @code{.bss} section
22347 in an ordinary executable.
22349 The @sc{gnu} binary utilities (Binutils) package includes the
22350 @samp{objcopy} utility that can produce
22351 the separated executable / debugging information file pairs using the
22352 following commands:
22355 @kbd{objcopy --only-keep-debug foo foo.debug}
22360 These commands remove the debugging
22361 information from the executable file @file{foo} and place it in the file
22362 @file{foo.debug}. You can use the first, second or both methods to link the
22367 The debug link method needs the following additional command to also leave
22368 behind a debug link in @file{foo}:
22371 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
22374 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
22375 a version of the @code{strip} command such that the command @kbd{strip foo -f
22376 foo.debug} has the same functionality as the two @code{objcopy} commands and
22377 the @code{ln -s} command above, together.
22380 Build ID gets embedded into the main executable using @code{ld --build-id} or
22381 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
22382 compatibility fixes for debug files separation are present in @sc{gnu} binary
22383 utilities (Binutils) package since version 2.18.
22388 @cindex CRC algorithm definition
22389 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
22390 IEEE 802.3 using the polynomial:
22392 @c TexInfo requires naked braces for multi-digit exponents for Tex
22393 @c output, but this causes HTML output to barf. HTML has to be set using
22394 @c raw commands. So we end up having to specify this equation in 2
22399 <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>
22400 + <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
22406 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
22407 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
22411 The function is computed byte at a time, taking the least
22412 significant bit of each byte first. The initial pattern
22413 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
22414 the final result is inverted to ensure trailing zeros also affect the
22417 @emph{Note:} This is the same CRC polynomial as used in handling the
22418 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{qCRC packet}).
22419 However in the case of the Remote Serial Protocol, the CRC is computed
22420 @emph{most} significant bit first, and the result is not inverted, so
22421 trailing zeros have no effect on the CRC value.
22423 To complete the description, we show below the code of the function
22424 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
22425 initially supplied @code{crc} argument means that an initial call to
22426 this function passing in zero will start computing the CRC using
22429 @kindex gnu_debuglink_crc32
22432 gnu_debuglink_crc32 (unsigned long crc,
22433 unsigned char *buf, size_t len)
22435 static const unsigned long crc32_table[256] =
22437 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
22438 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
22439 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
22440 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
22441 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
22442 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
22443 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
22444 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
22445 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
22446 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
22447 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
22448 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
22449 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
22450 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
22451 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
22452 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
22453 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
22454 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
22455 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
22456 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
22457 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
22458 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
22459 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
22460 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
22461 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
22462 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
22463 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
22464 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
22465 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
22466 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
22467 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
22468 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
22469 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
22470 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
22471 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
22472 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
22473 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
22474 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
22475 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
22476 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
22477 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
22478 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
22479 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
22480 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
22481 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
22482 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
22483 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
22484 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
22485 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
22486 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
22487 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
22490 unsigned char *end;
22492 crc = ~crc & 0xffffffff;
22493 for (end = buf + len; buf < end; ++buf)
22494 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
22495 return ~crc & 0xffffffff;
22500 This computation does not apply to the ``build ID'' method.
22502 @node MiniDebugInfo
22503 @section Debugging information in a special section
22504 @cindex separate debug sections
22505 @cindex @samp{.gnu_debugdata} section
22507 Some systems ship pre-built executables and libraries that have a
22508 special @samp{.gnu_debugdata} section. This feature is called
22509 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
22510 is used to supply extra symbols for backtraces.
22512 The intent of this section is to provide extra minimal debugging
22513 information for use in simple backtraces. It is not intended to be a
22514 replacement for full separate debugging information (@pxref{Separate
22515 Debug Files}). The example below shows the intended use; however,
22516 @value{GDBN} does not currently put restrictions on what sort of
22517 debugging information might be included in the section.
22519 @value{GDBN} has support for this extension. If the section exists,
22520 then it is used provided that no other source of debugging information
22521 can be found, and that @value{GDBN} was configured with LZMA support.
22523 This section can be easily created using @command{objcopy} and other
22524 standard utilities:
22527 # Extract the dynamic symbols from the main binary, there is no need
22528 # to also have these in the normal symbol table.
22529 nm -D @var{binary} --format=posix --defined-only \
22530 | awk '@{ print $1 @}' | sort > dynsyms
22532 # Extract all the text (i.e. function) symbols from the debuginfo.
22533 # (Note that we actually also accept "D" symbols, for the benefit
22534 # of platforms like PowerPC64 that use function descriptors.)
22535 nm @var{binary} --format=posix --defined-only \
22536 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
22539 # Keep all the function symbols not already in the dynamic symbol
22541 comm -13 dynsyms funcsyms > keep_symbols
22543 # Separate full debug info into debug binary.
22544 objcopy --only-keep-debug @var{binary} debug
22546 # Copy the full debuginfo, keeping only a minimal set of symbols and
22547 # removing some unnecessary sections.
22548 objcopy -S --remove-section .gdb_index --remove-section .comment \
22549 --keep-symbols=keep_symbols debug mini_debuginfo
22551 # Drop the full debug info from the original binary.
22552 strip --strip-all -R .comment @var{binary}
22554 # Inject the compressed data into the .gnu_debugdata section of the
22557 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
22561 @section Index Files Speed Up @value{GDBN}
22562 @cindex index files
22563 @cindex @samp{.gdb_index} section
22565 When @value{GDBN} finds a symbol file, it scans the symbols in the
22566 file in order to construct an internal symbol table. This lets most
22567 @value{GDBN} operations work quickly---at the cost of a delay early
22568 on. For large programs, this delay can be quite lengthy, so
22569 @value{GDBN} provides a way to build an index, which speeds up
22572 For convenience, @value{GDBN} comes with a program,
22573 @command{gdb-add-index}, which can be used to add the index to a
22574 symbol file. It takes the symbol file as its only argument:
22577 $ gdb-add-index symfile
22580 @xref{gdb-add-index}.
22582 It is also possible to do the work manually. Here is what
22583 @command{gdb-add-index} does behind the curtains.
22585 The index is stored as a section in the symbol file. @value{GDBN} can
22586 write the index to a file, then you can put it into the symbol file
22587 using @command{objcopy}.
22589 To create an index file, use the @code{save gdb-index} command:
22592 @item save gdb-index [-dwarf-5] @var{directory}
22593 @kindex save gdb-index
22594 Create index files for all symbol files currently known by
22595 @value{GDBN}. For each known @var{symbol-file}, this command by
22596 default creates it produces a single file
22597 @file{@var{symbol-file}.gdb-index}. If you invoke this command with
22598 the @option{-dwarf-5} option, it produces 2 files:
22599 @file{@var{symbol-file}.debug_names} and
22600 @file{@var{symbol-file}.debug_str}. The files are created in the
22601 given @var{directory}.
22604 Once you have created an index file you can merge it into your symbol
22605 file, here named @file{symfile}, using @command{objcopy}:
22608 $ objcopy --add-section .gdb_index=symfile.gdb-index \
22609 --set-section-flags .gdb_index=readonly symfile symfile
22612 Or for @code{-dwarf-5}:
22615 $ objcopy --dump-section .debug_str=symfile.debug_str.new symfile
22616 $ cat symfile.debug_str >>symfile.debug_str.new
22617 $ objcopy --add-section .debug_names=symfile.gdb-index \
22618 --set-section-flags .debug_names=readonly \
22619 --update-section .debug_str=symfile.debug_str.new symfile symfile
22622 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
22623 sections that have been deprecated. Usually they are deprecated because
22624 they are missing a new feature or have performance issues.
22625 To tell @value{GDBN} to use a deprecated index section anyway
22626 specify @code{set use-deprecated-index-sections on}.
22627 The default is @code{off}.
22628 This can speed up startup, but may result in some functionality being lost.
22629 @xref{Index Section Format}.
22631 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
22632 must be done before gdb reads the file. The following will not work:
22635 $ gdb -ex "set use-deprecated-index-sections on" <program>
22638 Instead you must do, for example,
22641 $ gdb -iex "set use-deprecated-index-sections on" <program>
22644 Indices only work when using DWARF debugging information, not stabs.
22646 @subsection Automatic symbol index cache
22648 @cindex automatic symbol index cache
22649 It is possible for @value{GDBN} to automatically save a copy of this index in a
22650 cache on disk and retrieve it from there when loading the same binary in the
22651 future. This feature can be turned on with @kbd{set index-cache enabled on}.
22652 The following commands can be used to tweak the behavior of the index cache.
22656 @kindex set index-cache
22657 @item set index-cache enabled on
22658 @itemx set index-cache enabled off
22659 Enable or disable the use of the symbol index cache.
22661 @item set index-cache directory @var{directory}
22662 @kindex show index-cache
22663 @itemx show index-cache directory
22664 Set/show the directory where index files will be saved.
22666 The default value for this directory depends on the host platform. On
22667 most systems, the index is cached in the @file{gdb} subdirectory of
22668 the directory pointed to by the @env{XDG_CACHE_HOME} environment
22669 variable, if it is defined, else in the @file{.cache/gdb} subdirectory
22670 of your home directory. However, on some systems, the default may
22671 differ according to local convention.
22673 There is no limit on the disk space used by index cache. It is perfectly safe
22674 to delete the content of that directory to free up disk space.
22676 @item show index-cache stats
22677 Print the number of cache hits and misses since the launch of @value{GDBN}.
22681 @node Symbol Errors
22682 @section Errors Reading Symbol Files
22684 While reading a symbol file, @value{GDBN} occasionally encounters problems,
22685 such as symbol types it does not recognize, or known bugs in compiler
22686 output. By default, @value{GDBN} does not notify you of such problems, since
22687 they are relatively common and primarily of interest to people
22688 debugging compilers. If you are interested in seeing information
22689 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
22690 only one message about each such type of problem, no matter how many
22691 times the problem occurs; or you can ask @value{GDBN} to print more messages,
22692 to see how many times the problems occur, with the @code{set
22693 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
22696 The messages currently printed, and their meanings, include:
22699 @item inner block not inside outer block in @var{symbol}
22701 The symbol information shows where symbol scopes begin and end
22702 (such as at the start of a function or a block of statements). This
22703 error indicates that an inner scope block is not fully contained
22704 in its outer scope blocks.
22706 @value{GDBN} circumvents the problem by treating the inner block as if it had
22707 the same scope as the outer block. In the error message, @var{symbol}
22708 may be shown as ``@code{(don't know)}'' if the outer block is not a
22711 @item block at @var{address} out of order
22713 The symbol information for symbol scope blocks should occur in
22714 order of increasing addresses. This error indicates that it does not
22717 @value{GDBN} does not circumvent this problem, and has trouble
22718 locating symbols in the source file whose symbols it is reading. (You
22719 can often determine what source file is affected by specifying
22720 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
22723 @item bad block start address patched
22725 The symbol information for a symbol scope block has a start address
22726 smaller than the address of the preceding source line. This is known
22727 to occur in the SunOS 4.1.1 (and earlier) C compiler.
22729 @value{GDBN} circumvents the problem by treating the symbol scope block as
22730 starting on the previous source line.
22732 @item bad string table offset in symbol @var{n}
22735 Symbol number @var{n} contains a pointer into the string table which is
22736 larger than the size of the string table.
22738 @value{GDBN} circumvents the problem by considering the symbol to have the
22739 name @code{foo}, which may cause other problems if many symbols end up
22742 @item unknown symbol type @code{0x@var{nn}}
22744 The symbol information contains new data types that @value{GDBN} does
22745 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
22746 uncomprehended information, in hexadecimal.
22748 @value{GDBN} circumvents the error by ignoring this symbol information.
22749 This usually allows you to debug your program, though certain symbols
22750 are not accessible. If you encounter such a problem and feel like
22751 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
22752 on @code{complain}, then go up to the function @code{read_dbx_symtab}
22753 and examine @code{*bufp} to see the symbol.
22755 @item stub type has NULL name
22757 @value{GDBN} could not find the full definition for a struct or class.
22759 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
22760 The symbol information for a C@t{++} member function is missing some
22761 information that recent versions of the compiler should have output for
22764 @item info mismatch between compiler and debugger
22766 @value{GDBN} could not parse a type specification output by the compiler.
22771 @section GDB Data Files
22773 @cindex prefix for data files
22774 @value{GDBN} will sometimes read an auxiliary data file. These files
22775 are kept in a directory known as the @dfn{data directory}.
22777 You can set the data directory's name, and view the name @value{GDBN}
22778 is currently using.
22781 @kindex set data-directory
22782 @item set data-directory @var{directory}
22783 Set the directory which @value{GDBN} searches for auxiliary data files
22784 to @var{directory}.
22786 @kindex show data-directory
22787 @item show data-directory
22788 Show the directory @value{GDBN} searches for auxiliary data files.
22791 @cindex default data directory
22792 @cindex @samp{--with-gdb-datadir}
22793 You can set the default data directory by using the configure-time
22794 @samp{--with-gdb-datadir} option. If the data directory is inside
22795 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
22796 @samp{--exec-prefix}), then the default data directory will be updated
22797 automatically if the installed @value{GDBN} is moved to a new
22800 The data directory may also be specified with the
22801 @code{--data-directory} command line option.
22802 @xref{Mode Options}.
22805 @chapter Specifying a Debugging Target
22807 @cindex debugging target
22808 A @dfn{target} is the execution environment occupied by your program.
22810 Often, @value{GDBN} runs in the same host environment as your program;
22811 in that case, the debugging target is specified as a side effect when
22812 you use the @code{file} or @code{core} commands. When you need more
22813 flexibility---for example, running @value{GDBN} on a physically separate
22814 host, or controlling a standalone system over a serial port or a
22815 realtime system over a TCP/IP connection---you can use the @code{target}
22816 command to specify one of the target types configured for @value{GDBN}
22817 (@pxref{Target Commands, ,Commands for Managing Targets}).
22819 @cindex target architecture
22820 It is possible to build @value{GDBN} for several different @dfn{target
22821 architectures}. When @value{GDBN} is built like that, you can choose
22822 one of the available architectures with the @kbd{set architecture}
22826 @kindex set architecture
22827 @kindex show architecture
22828 @item set architecture @var{arch}
22829 This command sets the current target architecture to @var{arch}. The
22830 value of @var{arch} can be @code{"auto"}, in addition to one of the
22831 supported architectures.
22833 @item show architecture
22834 Show the current target architecture.
22836 @item set processor
22838 @kindex set processor
22839 @kindex show processor
22840 These are alias commands for, respectively, @code{set architecture}
22841 and @code{show architecture}.
22845 * Active Targets:: Active targets
22846 * Target Commands:: Commands for managing targets
22847 * Byte Order:: Choosing target byte order
22850 @node Active Targets
22851 @section Active Targets
22853 @cindex stacking targets
22854 @cindex active targets
22855 @cindex multiple targets
22857 There are multiple classes of targets such as: processes, executable files or
22858 recording sessions. Core files belong to the process class, making core file
22859 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
22860 on multiple active targets, one in each class. This allows you to (for
22861 example) start a process and inspect its activity, while still having access to
22862 the executable file after the process finishes. Or if you start process
22863 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
22864 presented a virtual layer of the recording target, while the process target
22865 remains stopped at the chronologically last point of the process execution.
22867 Use the @code{core-file} and @code{exec-file} commands to select a new core
22868 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
22869 specify as a target a process that is already running, use the @code{attach}
22870 command (@pxref{Attach, ,Debugging an Already-running Process}).
22872 @node Target Commands
22873 @section Commands for Managing Targets
22876 @item target @var{type} @var{parameters}
22877 Connects the @value{GDBN} host environment to a target machine or
22878 process. A target is typically a protocol for talking to debugging
22879 facilities. You use the argument @var{type} to specify the type or
22880 protocol of the target machine.
22882 Further @var{parameters} are interpreted by the target protocol, but
22883 typically include things like device names or host names to connect
22884 with, process numbers, and baud rates.
22886 The @code{target} command does not repeat if you press @key{RET} again
22887 after executing the command.
22889 @kindex help target
22891 Displays the names of all targets available. To display targets
22892 currently selected, use either @code{info target} or @code{info files}
22893 (@pxref{Files, ,Commands to Specify Files}).
22895 @item help target @var{name}
22896 Describe a particular target, including any parameters necessary to
22899 @kindex set gnutarget
22900 @item set gnutarget @var{args}
22901 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
22902 knows whether it is reading an @dfn{executable},
22903 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
22904 with the @code{set gnutarget} command. Unlike most @code{target} commands,
22905 with @code{gnutarget} the @code{target} refers to a program, not a machine.
22908 @emph{Warning:} To specify a file format with @code{set gnutarget},
22909 you must know the actual BFD name.
22913 @xref{Files, , Commands to Specify Files}.
22915 @kindex show gnutarget
22916 @item show gnutarget
22917 Use the @code{show gnutarget} command to display what file format
22918 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
22919 @value{GDBN} will determine the file format for each file automatically,
22920 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
22923 @cindex common targets
22924 Here are some common targets (available, or not, depending on the GDB
22929 @item target exec @var{program}
22930 @cindex executable file target
22931 An executable file. @samp{target exec @var{program}} is the same as
22932 @samp{exec-file @var{program}}.
22934 @item target core @var{filename}
22935 @cindex core dump file target
22936 A core dump file. @samp{target core @var{filename}} is the same as
22937 @samp{core-file @var{filename}}.
22939 @item target remote @var{medium}
22940 @cindex remote target
22941 A remote system connected to @value{GDBN} via a serial line or network
22942 connection. This command tells @value{GDBN} to use its own remote
22943 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
22945 For example, if you have a board connected to @file{/dev/ttya} on the
22946 machine running @value{GDBN}, you could say:
22949 target remote /dev/ttya
22952 @code{target remote} supports the @code{load} command. This is only
22953 useful if you have some other way of getting the stub to the target
22954 system, and you can put it somewhere in memory where it won't get
22955 clobbered by the download.
22957 @item target sim @r{[}@var{simargs}@r{]} @dots{}
22958 @cindex built-in simulator target
22959 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
22967 works; however, you cannot assume that a specific memory map, device
22968 drivers, or even basic I/O is available, although some simulators do
22969 provide these. For info about any processor-specific simulator details,
22970 see the appropriate section in @ref{Embedded Processors, ,Embedded
22973 @item target native
22974 @cindex native target
22975 Setup for local/native process debugging. Useful to make the
22976 @code{run} command spawn native processes (likewise @code{attach},
22977 etc.@:) even when @code{set auto-connect-native-target} is @code{off}
22978 (@pxref{set auto-connect-native-target}).
22982 Different targets are available on different configurations of @value{GDBN};
22983 your configuration may have more or fewer targets.
22985 Many remote targets require you to download the executable's code once
22986 you've successfully established a connection. You may wish to control
22987 various aspects of this process.
22992 @kindex set hash@r{, for remote monitors}
22993 @cindex hash mark while downloading
22994 This command controls whether a hash mark @samp{#} is displayed while
22995 downloading a file to the remote monitor. If on, a hash mark is
22996 displayed after each S-record is successfully downloaded to the
23000 @kindex show hash@r{, for remote monitors}
23001 Show the current status of displaying the hash mark.
23003 @item set debug monitor
23004 @kindex set debug monitor
23005 @cindex display remote monitor communications
23006 Enable or disable display of communications messages between
23007 @value{GDBN} and the remote monitor.
23009 @item show debug monitor
23010 @kindex show debug monitor
23011 Show the current status of displaying communications between
23012 @value{GDBN} and the remote monitor.
23017 @kindex load @var{filename} @var{offset}
23018 @item load @var{filename} @var{offset}
23020 Depending on what remote debugging facilities are configured into
23021 @value{GDBN}, the @code{load} command may be available. Where it exists, it
23022 is meant to make @var{filename} (an executable) available for debugging
23023 on the remote system---by downloading, or dynamic linking, for example.
23024 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
23025 the @code{add-symbol-file} command.
23027 If your @value{GDBN} does not have a @code{load} command, attempting to
23028 execute it gets the error message ``@code{You can't do that when your
23029 target is @dots{}}''
23031 The file is loaded at whatever address is specified in the executable.
23032 For some object file formats, you can specify the load address when you
23033 link the program; for other formats, like a.out, the object file format
23034 specifies a fixed address.
23035 @c FIXME! This would be a good place for an xref to the GNU linker doc.
23037 It is also possible to tell @value{GDBN} to load the executable file at a
23038 specific offset described by the optional argument @var{offset}. When
23039 @var{offset} is provided, @var{filename} must also be provided.
23041 Depending on the remote side capabilities, @value{GDBN} may be able to
23042 load programs into flash memory.
23044 @code{load} does not repeat if you press @key{RET} again after using it.
23049 @kindex flash-erase
23051 @anchor{flash-erase}
23053 Erases all known flash memory regions on the target.
23058 @section Choosing Target Byte Order
23060 @cindex choosing target byte order
23061 @cindex target byte order
23063 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
23064 offer the ability to run either big-endian or little-endian byte
23065 orders. Usually the executable or symbol will include a bit to
23066 designate the endian-ness, and you will not need to worry about
23067 which to use. However, you may still find it useful to adjust
23068 @value{GDBN}'s idea of processor endian-ness manually.
23072 @item set endian big
23073 Instruct @value{GDBN} to assume the target is big-endian.
23075 @item set endian little
23076 Instruct @value{GDBN} to assume the target is little-endian.
23078 @item set endian auto
23079 Instruct @value{GDBN} to use the byte order associated with the
23083 Display @value{GDBN}'s current idea of the target byte order.
23087 If the @code{set endian auto} mode is in effect and no executable has
23088 been selected, then the endianness used is the last one chosen either
23089 by one of the @code{set endian big} and @code{set endian little}
23090 commands or by inferring from the last executable used. If no
23091 endianness has been previously chosen, then the default for this mode
23092 is inferred from the target @value{GDBN} has been built for, and is
23093 @code{little} if the name of the target CPU has an @code{el} suffix
23094 and @code{big} otherwise.
23096 Note that these commands merely adjust interpretation of symbolic
23097 data on the host, and that they have absolutely no effect on the
23101 @node Remote Debugging
23102 @chapter Debugging Remote Programs
23103 @cindex remote debugging
23105 If you are trying to debug a program running on a machine that cannot run
23106 @value{GDBN} in the usual way, it is often useful to use remote debugging.
23107 For example, you might use remote debugging on an operating system kernel,
23108 or on a small system which does not have a general purpose operating system
23109 powerful enough to run a full-featured debugger.
23111 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
23112 to make this work with particular debugging targets. In addition,
23113 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
23114 but not specific to any particular target system) which you can use if you
23115 write the remote stubs---the code that runs on the remote system to
23116 communicate with @value{GDBN}.
23118 Other remote targets may be available in your
23119 configuration of @value{GDBN}; use @code{help target} to list them.
23122 * Connecting:: Connecting to a remote target
23123 * File Transfer:: Sending files to a remote system
23124 * Server:: Using the gdbserver program
23125 * Remote Configuration:: Remote configuration
23126 * Remote Stub:: Implementing a remote stub
23130 @section Connecting to a Remote Target
23131 @cindex remote debugging, connecting
23132 @cindex @code{gdbserver}, connecting
23133 @cindex remote debugging, types of connections
23134 @cindex @code{gdbserver}, types of connections
23135 @cindex @code{gdbserver}, @code{target remote} mode
23136 @cindex @code{gdbserver}, @code{target extended-remote} mode
23138 This section describes how to connect to a remote target, including the
23139 types of connections and their differences, how to set up executable and
23140 symbol files on the host and target, and the commands used for
23141 connecting to and disconnecting from the remote target.
23143 @subsection Types of Remote Connections
23145 @value{GDBN} supports two types of remote connections, @code{target remote}
23146 mode and @code{target extended-remote} mode. Note that many remote targets
23147 support only @code{target remote} mode. There are several major
23148 differences between the two types of connections, enumerated here:
23152 @cindex remote debugging, detach and program exit
23153 @item Result of detach or program exit
23154 @strong{With target remote mode:} When the debugged program exits or you
23155 detach from it, @value{GDBN} disconnects from the target. When using
23156 @code{gdbserver}, @code{gdbserver} will exit.
23158 @strong{With target extended-remote mode:} When the debugged program exits or
23159 you detach from it, @value{GDBN} remains connected to the target, even
23160 though no program is running. You can rerun the program, attach to a
23161 running program, or use @code{monitor} commands specific to the target.
23163 When using @code{gdbserver} in this case, it does not exit unless it was
23164 invoked using the @option{--once} option. If the @option{--once} option
23165 was not used, you can ask @code{gdbserver} to exit using the
23166 @code{monitor exit} command (@pxref{Monitor Commands for gdbserver}).
23168 @item Specifying the program to debug
23169 For both connection types you use the @code{file} command to specify the
23170 program on the host system. If you are using @code{gdbserver} there are
23171 some differences in how to specify the location of the program on the
23174 @strong{With target remote mode:} You must either specify the program to debug
23175 on the @code{gdbserver} command line or use the @option{--attach} option
23176 (@pxref{Attaching to a program,,Attaching to a Running Program}).
23178 @cindex @option{--multi}, @code{gdbserver} option
23179 @strong{With target extended-remote mode:} You may specify the program to debug
23180 on the @code{gdbserver} command line, or you can load the program or attach
23181 to it using @value{GDBN} commands after connecting to @code{gdbserver}.
23183 @anchor{--multi Option in Types of Remote Connnections}
23184 You can start @code{gdbserver} without supplying an initial command to run
23185 or process ID to attach. To do this, use the @option{--multi} command line
23186 option. Then you can connect using @code{target extended-remote} and start
23187 the program you want to debug (see below for details on using the
23188 @code{run} command in this scenario). Note that the conditions under which
23189 @code{gdbserver} terminates depend on how @value{GDBN} connects to it
23190 (@code{target remote} or @code{target extended-remote}). The
23191 @option{--multi} option to @code{gdbserver} has no influence on that.
23193 @item The @code{run} command
23194 @strong{With target remote mode:} The @code{run} command is not
23195 supported. Once a connection has been established, you can use all
23196 the usual @value{GDBN} commands to examine and change data. The
23197 remote program is already running, so you can use commands like
23198 @kbd{step} and @kbd{continue}.
23200 @strong{With target extended-remote mode:} The @code{run} command is
23201 supported. The @code{run} command uses the value set by
23202 @code{set remote exec-file} (@pxref{set remote exec-file}) to select
23203 the program to run. Command line arguments are supported, except for
23204 wildcard expansion and I/O redirection (@pxref{Arguments}).
23206 If you specify the program to debug on the command line, then the
23207 @code{run} command is not required to start execution, and you can
23208 resume using commands like @kbd{step} and @kbd{continue} as with
23209 @code{target remote} mode.
23211 @anchor{Attaching in Types of Remote Connections}
23213 @strong{With target remote mode:} The @value{GDBN} command @code{attach} is
23214 not supported. To attach to a running program using @code{gdbserver}, you
23215 must use the @option{--attach} option (@pxref{Running gdbserver}).
23217 @strong{With target extended-remote mode:} To attach to a running program,
23218 you may use the @code{attach} command after the connection has been
23219 established. If you are using @code{gdbserver}, you may also invoke
23220 @code{gdbserver} using the @option{--attach} option
23221 (@pxref{Running gdbserver}).
23223 Some remote targets allow @value{GDBN} to determine the executable file running
23224 in the process the debugger is attaching to. In such a case, @value{GDBN}
23225 uses the value of @code{exec-file-mismatch} to handle a possible mismatch
23226 between the executable file name running in the process and the name of the
23227 current exec-file loaded by @value{GDBN} (@pxref{set exec-file-mismatch}).
23231 @anchor{Host and target files}
23232 @subsection Host and Target Files
23233 @cindex remote debugging, symbol files
23234 @cindex symbol files, remote debugging
23236 @value{GDBN}, running on the host, needs access to symbol and debugging
23237 information for your program running on the target. This requires
23238 access to an unstripped copy of your program, and possibly any associated
23239 symbol files. Note that this section applies equally to both @code{target
23240 remote} mode and @code{target extended-remote} mode.
23242 Some remote targets (@pxref{qXfer executable filename read}, and
23243 @pxref{Host I/O Packets}) allow @value{GDBN} to access program files over
23244 the same connection used to communicate with @value{GDBN}. With such a
23245 target, if the remote program is unstripped, the only command you need is
23246 @code{target remote} (or @code{target extended-remote}).
23248 If the remote program is stripped, or the target does not support remote
23249 program file access, start up @value{GDBN} using the name of the local
23250 unstripped copy of your program as the first argument, or use the
23251 @code{file} command. Use @code{set sysroot} to specify the location (on
23252 the host) of target libraries (unless your @value{GDBN} was compiled with
23253 the correct sysroot using @code{--with-sysroot}). Alternatively, you
23254 may use @code{set solib-search-path} to specify how @value{GDBN} locates
23257 The symbol file and target libraries must exactly match the executable
23258 and libraries on the target, with one exception: the files on the host
23259 system should not be stripped, even if the files on the target system
23260 are. Mismatched or missing files will lead to confusing results
23261 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
23262 files may also prevent @code{gdbserver} from debugging multi-threaded
23265 @subsection Remote Connection Commands
23266 @cindex remote connection commands
23267 @value{GDBN} can communicate with the target over a serial line, a
23268 local Unix domain socket, or
23269 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
23270 each case, @value{GDBN} uses the same protocol for debugging your
23271 program; only the medium carrying the debugging packets varies. The
23272 @code{target remote} and @code{target extended-remote} commands
23273 establish a connection to the target. Both commands accept the same
23274 arguments, which indicate the medium to use:
23278 @item target remote @var{serial-device}
23279 @itemx target extended-remote @var{serial-device}
23280 @cindex serial line, @code{target remote}
23281 Use @var{serial-device} to communicate with the target. For example,
23282 to use a serial line connected to the device named @file{/dev/ttyb}:
23285 target remote /dev/ttyb
23288 If you're using a serial line, you may want to give @value{GDBN} the
23289 @samp{--baud} option, or use the @code{set serial baud} command
23290 (@pxref{Remote Configuration, set serial baud}) before the
23291 @code{target} command.
23293 @item target remote @var{local-socket}
23294 @itemx target extended-remote @var{local-socket}
23295 @cindex local socket, @code{target remote}
23296 @cindex Unix domain socket
23297 Use @var{local-socket} to communicate with the target. For example,
23298 to use a local Unix domain socket bound to the file system entry @file{/tmp/gdb-socket0}:
23301 target remote /tmp/gdb-socket0
23304 Note that this command has the same form as the command to connect
23305 to a serial line. @value{GDBN} will automatically determine which
23306 kind of file you have specified and will make the appropriate kind
23308 This feature is not available if the host system does not support
23309 Unix domain sockets.
23311 @item target remote @code{@var{host}:@var{port}}
23312 @itemx target remote @code{[@var{host}]:@var{port}}
23313 @itemx target remote @code{tcp:@var{host}:@var{port}}
23314 @itemx target remote @code{tcp:[@var{host}]:@var{port}}
23315 @itemx target remote @code{tcp4:@var{host}:@var{port}}
23316 @itemx target remote @code{tcp6:@var{host}:@var{port}}
23317 @itemx target remote @code{tcp6:[@var{host}]:@var{port}}
23318 @itemx target extended-remote @code{@var{host}:@var{port}}
23319 @itemx target extended-remote @code{[@var{host}]:@var{port}}
23320 @itemx target extended-remote @code{tcp:@var{host}:@var{port}}
23321 @itemx target extended-remote @code{tcp:[@var{host}]:@var{port}}
23322 @itemx target extended-remote @code{tcp4:@var{host}:@var{port}}
23323 @itemx target extended-remote @code{tcp6:@var{host}:@var{port}}
23324 @itemx target extended-remote @code{tcp6:[@var{host}]:@var{port}}
23325 @cindex @acronym{TCP} port, @code{target remote}
23326 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
23327 The @var{host} may be either a host name, a numeric @acronym{IPv4}
23328 address, or a numeric @acronym{IPv6} address (with or without the
23329 square brackets to separate the address from the port); @var{port}
23330 must be a decimal number. The @var{host} could be the target machine
23331 itself, if it is directly connected to the net, or it might be a
23332 terminal server which in turn has a serial line to the target.
23334 For example, to connect to port 2828 on a terminal server named
23338 target remote manyfarms:2828
23341 To connect to port 2828 on a terminal server whose address is
23342 @code{2001:0db8:85a3:0000:0000:8a2e:0370:7334}, you can either use the
23343 square bracket syntax:
23346 target remote [2001:0db8:85a3:0000:0000:8a2e:0370:7334]:2828
23350 or explicitly specify the @acronym{IPv6} protocol:
23353 target remote tcp6:2001:0db8:85a3:0000:0000:8a2e:0370:7334:2828
23356 This last example may be confusing to the reader, because there is no
23357 visible separation between the hostname and the port number.
23358 Therefore, we recommend the user to provide @acronym{IPv6} addresses
23359 using square brackets for clarity. However, it is important to
23360 mention that for @value{GDBN} there is no ambiguity: the number after
23361 the last colon is considered to be the port number.
23363 If your remote target is actually running on the same machine as your
23364 debugger session (e.g.@: a simulator for your target running on the
23365 same host), you can omit the hostname. For example, to connect to
23366 port 1234 on your local machine:
23369 target remote :1234
23373 Note that the colon is still required here.
23375 @item target remote @code{udp:@var{host}:@var{port}}
23376 @itemx target remote @code{udp:[@var{host}]:@var{port}}
23377 @itemx target remote @code{udp4:@var{host}:@var{port}}
23378 @itemx target remote @code{udp6:[@var{host}]:@var{port}}
23379 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
23380 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
23381 @itemx target extended-remote @code{udp:[@var{host}]:@var{port}}
23382 @itemx target extended-remote @code{udp4:@var{host}:@var{port}}
23383 @itemx target extended-remote @code{udp6:@var{host}:@var{port}}
23384 @itemx target extended-remote @code{udp6:[@var{host}]:@var{port}}
23385 @cindex @acronym{UDP} port, @code{target remote}
23386 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
23387 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
23390 target remote udp:manyfarms:2828
23393 When using a @acronym{UDP} connection for remote debugging, you should
23394 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
23395 can silently drop packets on busy or unreliable networks, which will
23396 cause havoc with your debugging session.
23398 @item target remote | @var{command}
23399 @itemx target extended-remote | @var{command}
23400 @cindex pipe, @code{target remote} to
23401 Run @var{command} in the background and communicate with it using a
23402 pipe. The @var{command} is a shell command, to be parsed and expanded
23403 by the system's command shell, @code{/bin/sh}; it should expect remote
23404 protocol packets on its standard input, and send replies on its
23405 standard output. You could use this to run a stand-alone simulator
23406 that speaks the remote debugging protocol, to make net connections
23407 using programs like @code{ssh}, or for other similar tricks.
23409 If @var{command} closes its standard output (perhaps by exiting),
23410 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
23411 program has already exited, this will have no effect.)
23415 @cindex interrupting remote programs
23416 @cindex remote programs, interrupting
23417 Whenever @value{GDBN} is waiting for the remote program, if you type the
23418 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
23419 program. This may or may not succeed, depending in part on the hardware
23420 and the serial drivers the remote system uses. If you type the
23421 interrupt character once again, @value{GDBN} displays this prompt:
23424 Interrupted while waiting for the program.
23425 Give up (and stop debugging it)? (y or n)
23428 In @code{target remote} mode, if you type @kbd{y}, @value{GDBN} abandons
23429 the remote debugging session. (If you decide you want to try again later,
23430 you can use @kbd{target remote} again to connect once more.) If you type
23431 @kbd{n}, @value{GDBN} goes back to waiting.
23433 In @code{target extended-remote} mode, typing @kbd{n} will leave
23434 @value{GDBN} connected to the target.
23437 @kindex detach (remote)
23439 When you have finished debugging the remote program, you can use the
23440 @code{detach} command to release it from @value{GDBN} control.
23441 Detaching from the target normally resumes its execution, but the results
23442 will depend on your particular remote stub. After the @code{detach}
23443 command in @code{target remote} mode, @value{GDBN} is free to connect to
23444 another target. In @code{target extended-remote} mode, @value{GDBN} is
23445 still connected to the target.
23449 The @code{disconnect} command closes the connection to the target, and
23450 the target is generally not resumed. It will wait for @value{GDBN}
23451 (this instance or another one) to connect and continue debugging. After
23452 the @code{disconnect} command, @value{GDBN} is again free to connect to
23455 @cindex send command to remote monitor
23456 @cindex extend @value{GDBN} for remote targets
23457 @cindex add new commands for external monitor
23459 @item monitor @var{cmd}
23460 This command allows you to send arbitrary commands directly to the
23461 remote monitor. Since @value{GDBN} doesn't care about the commands it
23462 sends like this, this command is the way to extend @value{GDBN}---you
23463 can add new commands that only the external monitor will understand
23467 @node File Transfer
23468 @section Sending files to a remote system
23469 @cindex remote target, file transfer
23470 @cindex file transfer
23471 @cindex sending files to remote systems
23473 Some remote targets offer the ability to transfer files over the same
23474 connection used to communicate with @value{GDBN}. This is convenient
23475 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
23476 running @code{gdbserver} over a network interface. For other targets,
23477 e.g.@: embedded devices with only a single serial port, this may be
23478 the only way to upload or download files.
23480 Not all remote targets support these commands.
23484 @item remote put @var{hostfile} @var{targetfile}
23485 Copy file @var{hostfile} from the host system (the machine running
23486 @value{GDBN}) to @var{targetfile} on the target system.
23489 @item remote get @var{targetfile} @var{hostfile}
23490 Copy file @var{targetfile} from the target system to @var{hostfile}
23491 on the host system.
23493 @kindex remote delete
23494 @item remote delete @var{targetfile}
23495 Delete @var{targetfile} from the target system.
23500 @section Using the @code{gdbserver} Program
23503 @cindex remote connection without stubs
23504 @code{gdbserver} is a control program for Unix-like systems, which
23505 allows you to connect your program with a remote @value{GDBN} via
23506 @code{target remote} or @code{target extended-remote}---but without
23507 linking in the usual debugging stub.
23509 @code{gdbserver} is not a complete replacement for the debugging stubs,
23510 because it requires essentially the same operating-system facilities
23511 that @value{GDBN} itself does. In fact, a system that can run
23512 @code{gdbserver} to connect to a remote @value{GDBN} could also run
23513 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
23514 because it is a much smaller program than @value{GDBN} itself. It is
23515 also easier to port than all of @value{GDBN}, so you may be able to get
23516 started more quickly on a new system by using @code{gdbserver}.
23517 Finally, if you develop code for real-time systems, you may find that
23518 the tradeoffs involved in real-time operation make it more convenient to
23519 do as much development work as possible on another system, for example
23520 by cross-compiling. You can use @code{gdbserver} to make a similar
23521 choice for debugging.
23523 @value{GDBN} and @code{gdbserver} communicate via either a serial line
23524 or a TCP connection, using the standard @value{GDBN} remote serial
23528 @emph{Warning:} @code{gdbserver} does not have any built-in security.
23529 Do not run @code{gdbserver} connected to any public network; a
23530 @value{GDBN} connection to @code{gdbserver} provides access to the
23531 target system with the same privileges as the user running
23535 @anchor{Running gdbserver}
23536 @subsection Running @code{gdbserver}
23537 @cindex arguments, to @code{gdbserver}
23538 @cindex @code{gdbserver}, command-line arguments
23540 Run @code{gdbserver} on the target system. You need a copy of the
23541 program you want to debug, including any libraries it requires.
23542 @code{gdbserver} does not need your program's symbol table, so you can
23543 strip the program if necessary to save space. @value{GDBN} on the host
23544 system does all the symbol handling.
23546 To use the server, you must tell it how to communicate with @value{GDBN};
23547 the name of your program; and the arguments for your program. The usual
23551 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
23554 @var{comm} is either a device name (to use a serial line), or a TCP
23555 hostname and portnumber, or @code{-} or @code{stdio} to use
23556 stdin/stdout of @code{gdbserver}.
23557 For example, to debug Emacs with the argument
23558 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
23562 target> gdbserver /dev/com1 emacs foo.txt
23565 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
23568 To use a TCP connection instead of a serial line:
23571 target> gdbserver host:2345 emacs foo.txt
23574 The only difference from the previous example is the first argument,
23575 specifying that you are communicating with the host @value{GDBN} via
23576 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
23577 expect a TCP connection from machine @samp{host} to local TCP port 2345.
23578 (Currently, the @samp{host} part is ignored.) You can choose any number
23579 you want for the port number as long as it does not conflict with any
23580 TCP ports already in use on the target system (for example, @code{23} is
23581 reserved for @code{telnet}).@footnote{If you choose a port number that
23582 conflicts with another service, @code{gdbserver} prints an error message
23583 and exits.} You must use the same port number with the host @value{GDBN}
23584 @code{target remote} command.
23586 The @code{stdio} connection is useful when starting @code{gdbserver}
23590 (@value{GDBP}) target remote | ssh -T hostname gdbserver - hello
23593 The @samp{-T} option to ssh is provided because we don't need a remote pty,
23594 and we don't want escape-character handling. Ssh does this by default when
23595 a command is provided, the flag is provided to make it explicit.
23596 You could elide it if you want to.
23598 Programs started with stdio-connected gdbserver have @file{/dev/null} for
23599 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
23600 display through a pipe connected to gdbserver.
23601 Both @code{stdout} and @code{stderr} use the same pipe.
23603 @anchor{Attaching to a program}
23604 @subsubsection Attaching to a Running Program
23605 @cindex attach to a program, @code{gdbserver}
23606 @cindex @option{--attach}, @code{gdbserver} option
23608 On some targets, @code{gdbserver} can also attach to running programs.
23609 This is accomplished via the @code{--attach} argument. The syntax is:
23612 target> gdbserver --attach @var{comm} @var{pid}
23615 @var{pid} is the process ID of a currently running process. It isn't
23616 necessary to point @code{gdbserver} at a binary for the running process.
23618 In @code{target extended-remote} mode, you can also attach using the
23619 @value{GDBN} attach command
23620 (@pxref{Attaching in Types of Remote Connections}).
23623 You can debug processes by name instead of process ID if your target has the
23624 @code{pidof} utility:
23627 target> gdbserver --attach @var{comm} `pidof @var{program}`
23630 In case more than one copy of @var{program} is running, or @var{program}
23631 has multiple threads, most versions of @code{pidof} support the
23632 @code{-s} option to only return the first process ID.
23634 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
23636 This section applies only when @code{gdbserver} is run to listen on a TCP
23639 @code{gdbserver} normally terminates after all of its debugged processes have
23640 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
23641 extended-remote}, @code{gdbserver} stays running even with no processes left.
23642 @value{GDBN} normally terminates the spawned debugged process on its exit,
23643 which normally also terminates @code{gdbserver} in the @kbd{target remote}
23644 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
23645 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
23646 stays running even in the @kbd{target remote} mode.
23648 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
23649 Such reconnecting is useful for features like @ref{disconnected tracing}. For
23650 completeness, at most one @value{GDBN} can be connected at a time.
23652 @cindex @option{--once}, @code{gdbserver} option
23653 By default, @code{gdbserver} keeps the listening TCP port open, so that
23654 subsequent connections are possible. However, if you start @code{gdbserver}
23655 with the @option{--once} option, it will stop listening for any further
23656 connection attempts after connecting to the first @value{GDBN} session. This
23657 means no further connections to @code{gdbserver} will be possible after the
23658 first one. It also means @code{gdbserver} will terminate after the first
23659 connection with remote @value{GDBN} has closed, even for unexpectedly closed
23660 connections and even in the @kbd{target extended-remote} mode. The
23661 @option{--once} option allows reusing the same port number for connecting to
23662 multiple instances of @code{gdbserver} running on the same host, since each
23663 instance closes its port after the first connection.
23665 @anchor{Other Command-Line Arguments for gdbserver}
23666 @subsubsection Other Command-Line Arguments for @code{gdbserver}
23668 You can use the @option{--multi} option to start @code{gdbserver} without
23669 specifying a program to debug or a process to attach to. Then you can
23670 attach in @code{target extended-remote} mode and run or attach to a
23671 program. For more information,
23672 @pxref{--multi Option in Types of Remote Connnections}.
23674 @cindex @option{--debug}, @code{gdbserver} option
23675 The @option{--debug} option tells @code{gdbserver} to display extra
23676 status information about the debugging process.
23677 @cindex @option{--remote-debug}, @code{gdbserver} option
23678 The @option{--remote-debug} option tells @code{gdbserver} to display
23679 remote protocol debug output.
23680 @cindex @option{--debug-file}, @code{gdbserver} option
23681 @cindex @code{gdbserver}, send all debug output to a single file
23682 The @option{--debug-file=@var{filename}} option tells @code{gdbserver} to
23683 write any debug output to the given @var{filename}. These options are intended
23684 for @code{gdbserver} development and for bug reports to the developers.
23686 @cindex @option{--debug-format}, @code{gdbserver} option
23687 The @option{--debug-format=option1[,option2,...]} option tells
23688 @code{gdbserver} to include additional information in each output.
23689 Possible options are:
23693 Turn off all extra information in debugging output.
23695 Turn on all extra information in debugging output.
23697 Include a timestamp in each line of debugging output.
23700 Options are processed in order. Thus, for example, if @option{none}
23701 appears last then no additional information is added to debugging output.
23703 @cindex @option{--wrapper}, @code{gdbserver} option
23704 The @option{--wrapper} option specifies a wrapper to launch programs
23705 for debugging. The option should be followed by the name of the
23706 wrapper, then any command-line arguments to pass to the wrapper, then
23707 @kbd{--} indicating the end of the wrapper arguments.
23709 @code{gdbserver} runs the specified wrapper program with a combined
23710 command line including the wrapper arguments, then the name of the
23711 program to debug, then any arguments to the program. The wrapper
23712 runs until it executes your program, and then @value{GDBN} gains control.
23714 You can use any program that eventually calls @code{execve} with
23715 its arguments as a wrapper. Several standard Unix utilities do
23716 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
23717 with @code{exec "$@@"} will also work.
23719 For example, you can use @code{env} to pass an environment variable to
23720 the debugged program, without setting the variable in @code{gdbserver}'s
23724 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
23727 @cindex @option{--selftest}
23728 The @option{--selftest} option runs the self tests in @code{gdbserver}:
23731 $ gdbserver --selftest
23732 Ran 2 unit tests, 0 failed
23735 These tests are disabled in release.
23736 @subsection Connecting to @code{gdbserver}
23738 The basic procedure for connecting to the remote target is:
23742 Run @value{GDBN} on the host system.
23745 Make sure you have the necessary symbol files
23746 (@pxref{Host and target files}).
23747 Load symbols for your application using the @code{file} command before you
23748 connect. Use @code{set sysroot} to locate target libraries (unless your
23749 @value{GDBN} was compiled with the correct sysroot using
23750 @code{--with-sysroot}).
23753 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
23754 For TCP connections, you must start up @code{gdbserver} prior to using
23755 the @code{target} command. Otherwise you may get an error whose
23756 text depends on the host system, but which usually looks something like
23757 @samp{Connection refused}. Don't use the @code{load}
23758 command in @value{GDBN} when using @code{target remote} mode, since the
23759 program is already on the target.
23763 @anchor{Monitor Commands for gdbserver}
23764 @subsection Monitor Commands for @code{gdbserver}
23765 @cindex monitor commands, for @code{gdbserver}
23767 During a @value{GDBN} session using @code{gdbserver}, you can use the
23768 @code{monitor} command to send special requests to @code{gdbserver}.
23769 Here are the available commands.
23773 List the available monitor commands.
23775 @item monitor set debug 0
23776 @itemx monitor set debug 1
23777 Disable or enable general debugging messages.
23779 @item monitor set remote-debug 0
23780 @itemx monitor set remote-debug 1
23781 Disable or enable specific debugging messages associated with the remote
23782 protocol (@pxref{Remote Protocol}).
23784 @item monitor set debug-file filename
23785 @itemx monitor set debug-file
23786 Send any debug output to the given file, or to stderr.
23788 @item monitor set debug-format option1@r{[},option2,...@r{]}
23789 Specify additional text to add to debugging messages.
23790 Possible options are:
23794 Turn off all extra information in debugging output.
23796 Turn on all extra information in debugging output.
23798 Include a timestamp in each line of debugging output.
23801 Options are processed in order. Thus, for example, if @option{none}
23802 appears last then no additional information is added to debugging output.
23804 @item monitor set libthread-db-search-path [PATH]
23805 @cindex gdbserver, search path for @code{libthread_db}
23806 When this command is issued, @var{path} is a colon-separated list of
23807 directories to search for @code{libthread_db} (@pxref{Threads,,set
23808 libthread-db-search-path}). If you omit @var{path},
23809 @samp{libthread-db-search-path} will be reset to its default value.
23811 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
23812 not supported in @code{gdbserver}.
23815 Tell gdbserver to exit immediately. This command should be followed by
23816 @code{disconnect} to close the debugging session. @code{gdbserver} will
23817 detach from any attached processes and kill any processes it created.
23818 Use @code{monitor exit} to terminate @code{gdbserver} at the end
23819 of a multi-process mode debug session.
23823 @subsection Tracepoints support in @code{gdbserver}
23824 @cindex tracepoints support in @code{gdbserver}
23826 On some targets, @code{gdbserver} supports tracepoints, fast
23827 tracepoints and static tracepoints.
23829 For fast or static tracepoints to work, a special library called the
23830 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
23831 This library is built and distributed as an integral part of
23832 @code{gdbserver}. In addition, support for static tracepoints
23833 requires building the in-process agent library with static tracepoints
23834 support. At present, the UST (LTTng Userspace Tracer,
23835 @url{http://lttng.org/ust}) tracing engine is supported. This support
23836 is automatically available if UST development headers are found in the
23837 standard include path when @code{gdbserver} is built, or if
23838 @code{gdbserver} was explicitly configured using @option{--with-ust}
23839 to point at such headers. You can explicitly disable the support
23840 using @option{--with-ust=no}.
23842 There are several ways to load the in-process agent in your program:
23845 @item Specifying it as dependency at link time
23847 You can link your program dynamically with the in-process agent
23848 library. On most systems, this is accomplished by adding
23849 @code{-linproctrace} to the link command.
23851 @item Using the system's preloading mechanisms
23853 You can force loading the in-process agent at startup time by using
23854 your system's support for preloading shared libraries. Many Unixes
23855 support the concept of preloading user defined libraries. In most
23856 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
23857 in the environment. See also the description of @code{gdbserver}'s
23858 @option{--wrapper} command line option.
23860 @item Using @value{GDBN} to force loading the agent at run time
23862 On some systems, you can force the inferior to load a shared library,
23863 by calling a dynamic loader function in the inferior that takes care
23864 of dynamically looking up and loading a shared library. On most Unix
23865 systems, the function is @code{dlopen}. You'll use the @code{call}
23866 command for that. For example:
23869 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
23872 Note that on most Unix systems, for the @code{dlopen} function to be
23873 available, the program needs to be linked with @code{-ldl}.
23876 On systems that have a userspace dynamic loader, like most Unix
23877 systems, when you connect to @code{gdbserver} using @code{target
23878 remote}, you'll find that the program is stopped at the dynamic
23879 loader's entry point, and no shared library has been loaded in the
23880 program's address space yet, including the in-process agent. In that
23881 case, before being able to use any of the fast or static tracepoints
23882 features, you need to let the loader run and load the shared
23883 libraries. The simplest way to do that is to run the program to the
23884 main procedure. E.g., if debugging a C or C@t{++} program, start
23885 @code{gdbserver} like so:
23888 $ gdbserver :9999 myprogram
23891 Start GDB and connect to @code{gdbserver} like so, and run to main:
23895 (@value{GDBP}) target remote myhost:9999
23896 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
23897 (@value{GDBP}) b main
23898 (@value{GDBP}) continue
23901 The in-process tracing agent library should now be loaded into the
23902 process; you can confirm it with the @code{info sharedlibrary}
23903 command, which will list @file{libinproctrace.so} as loaded in the
23904 process. You are now ready to install fast tracepoints, list static
23905 tracepoint markers, probe static tracepoints markers, and start
23908 @node Remote Configuration
23909 @section Remote Configuration
23912 @kindex show remote
23913 This section documents the configuration options available when
23914 debugging remote programs. For the options related to the File I/O
23915 extensions of the remote protocol, see @ref{system,
23916 system-call-allowed}.
23919 @item set remoteaddresssize @var{bits}
23920 @cindex address size for remote targets
23921 @cindex bits in remote address
23922 Set the maximum size of address in a memory packet to the specified
23923 number of bits. @value{GDBN} will mask off the address bits above
23924 that number, when it passes addresses to the remote target. The
23925 default value is the number of bits in the target's address.
23927 @item show remoteaddresssize
23928 Show the current value of remote address size in bits.
23930 @item set serial baud @var{n}
23931 @cindex baud rate for remote targets
23932 Set the baud rate for the remote serial I/O to @var{n} baud. The
23933 value is used to set the speed of the serial port used for debugging
23936 @item show serial baud
23937 Show the current speed of the remote connection.
23939 @item set serial parity @var{parity}
23940 Set the parity for the remote serial I/O. Supported values of @var{parity} are:
23941 @code{even}, @code{none}, and @code{odd}. The default is @code{none}.
23943 @item show serial parity
23944 Show the current parity of the serial port.
23946 @item set remotebreak
23947 @cindex interrupt remote programs
23948 @cindex BREAK signal instead of Ctrl-C
23949 @anchor{set remotebreak}
23950 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
23951 when you type @kbd{Ctrl-c} to interrupt the program running
23952 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
23953 character instead. The default is off, since most remote systems
23954 expect to see @samp{Ctrl-C} as the interrupt signal.
23956 @item show remotebreak
23957 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
23958 interrupt the remote program.
23960 @item set remoteflow on
23961 @itemx set remoteflow off
23962 @kindex set remoteflow
23963 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
23964 on the serial port used to communicate to the remote target.
23966 @item show remoteflow
23967 @kindex show remoteflow
23968 Show the current setting of hardware flow control.
23970 @item set remotelogbase @var{base}
23971 Set the base (a.k.a.@: radix) of logging serial protocol
23972 communications to @var{base}. Supported values of @var{base} are:
23973 @code{ascii}, @code{octal}, and @code{hex}. The default is
23976 @item show remotelogbase
23977 Show the current setting of the radix for logging remote serial
23980 @item set remotelogfile @var{file}
23981 @cindex record serial communications on file
23982 Record remote serial communications on the named @var{file}. The
23983 default is not to record at all.
23985 @item show remotelogfile
23986 Show the current setting of the file name on which to record the
23987 serial communications.
23989 @item set remotetimeout @var{num}
23990 @cindex timeout for serial communications
23991 @cindex remote timeout
23992 Set the timeout limit to wait for the remote target to respond to
23993 @var{num} seconds. The default is 2 seconds.
23995 @item show remotetimeout
23996 Show the current number of seconds to wait for the remote target
23999 @cindex limit hardware breakpoints and watchpoints
24000 @cindex remote target, limit break- and watchpoints
24001 @anchor{set remote hardware-watchpoint-limit}
24002 @anchor{set remote hardware-breakpoint-limit}
24003 @item set remote hardware-watchpoint-limit @var{limit}
24004 @itemx set remote hardware-breakpoint-limit @var{limit}
24005 Restrict @value{GDBN} to using @var{limit} remote hardware watchpoints
24006 or breakpoints. The @var{limit} can be set to 0 to disable hardware
24007 watchpoints or breakpoints, and @code{unlimited} for unlimited
24008 watchpoints or breakpoints.
24010 @item show remote hardware-watchpoint-limit
24011 @itemx show remote hardware-breakpoint-limit
24012 Show the current limit for the number of hardware watchpoints or
24013 breakpoints that @value{GDBN} can use.
24015 @cindex limit hardware watchpoints length
24016 @cindex remote target, limit watchpoints length
24017 @anchor{set remote hardware-watchpoint-length-limit}
24018 @item set remote hardware-watchpoint-length-limit @var{limit}
24019 Restrict @value{GDBN} to using @var{limit} bytes for the maximum
24020 length of a remote hardware watchpoint. A @var{limit} of 0 disables
24021 hardware watchpoints and @code{unlimited} allows watchpoints of any
24024 @item show remote hardware-watchpoint-length-limit
24025 Show the current limit (in bytes) of the maximum length of
24026 a remote hardware watchpoint.
24028 @item set remote exec-file @var{filename}
24029 @itemx show remote exec-file
24030 @anchor{set remote exec-file}
24031 @cindex executable file, for remote target
24032 Select the file used for @code{run} with @code{target
24033 extended-remote}. This should be set to a filename valid on the
24034 target system. If it is not set, the target will use a default
24035 filename (e.g.@: the last program run).
24037 @item set remote interrupt-sequence
24038 @cindex interrupt remote programs
24039 @cindex select Ctrl-C, BREAK or BREAK-g
24040 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
24041 @samp{BREAK-g} as the
24042 sequence to the remote target in order to interrupt the execution.
24043 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
24044 is high level of serial line for some certain time.
24045 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
24046 It is @code{BREAK} signal followed by character @code{g}.
24048 @item show remote interrupt-sequence
24049 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
24050 is sent by @value{GDBN} to interrupt the remote program.
24051 @code{BREAK-g} is BREAK signal followed by @code{g} and
24052 also known as Magic SysRq g.
24054 @item set remote interrupt-on-connect
24055 @cindex send interrupt-sequence on start
24056 Specify whether interrupt-sequence is sent to remote target when
24057 @value{GDBN} connects to it. This is mostly needed when you debug
24058 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
24059 which is known as Magic SysRq g in order to connect @value{GDBN}.
24061 @item show remote interrupt-on-connect
24062 Show whether interrupt-sequence is sent
24063 to remote target when @value{GDBN} connects to it.
24067 @item set tcp auto-retry on
24068 @cindex auto-retry, for remote TCP target
24069 Enable auto-retry for remote TCP connections. This is useful if the remote
24070 debugging agent is launched in parallel with @value{GDBN}; there is a race
24071 condition because the agent may not become ready to accept the connection
24072 before @value{GDBN} attempts to connect. When auto-retry is
24073 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
24074 to establish the connection using the timeout specified by
24075 @code{set tcp connect-timeout}.
24077 @item set tcp auto-retry off
24078 Do not auto-retry failed TCP connections.
24080 @item show tcp auto-retry
24081 Show the current auto-retry setting.
24083 @item set tcp connect-timeout @var{seconds}
24084 @itemx set tcp connect-timeout unlimited
24085 @cindex connection timeout, for remote TCP target
24086 @cindex timeout, for remote target connection
24087 Set the timeout for establishing a TCP connection to the remote target to
24088 @var{seconds}. The timeout affects both polling to retry failed connections
24089 (enabled by @code{set tcp auto-retry on}) and waiting for connections
24090 that are merely slow to complete, and represents an approximate cumulative
24091 value. If @var{seconds} is @code{unlimited}, there is no timeout and
24092 @value{GDBN} will keep attempting to establish a connection forever,
24093 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
24095 @item show tcp connect-timeout
24096 Show the current connection timeout setting.
24099 @cindex remote packets, enabling and disabling
24100 The @value{GDBN} remote protocol autodetects the packets supported by
24101 your debugging stub. If you need to override the autodetection, you
24102 can use these commands to enable or disable individual packets. Each
24103 packet can be set to @samp{on} (the remote target supports this
24104 packet), @samp{off} (the remote target does not support this packet),
24105 or @samp{auto} (detect remote target support for this packet). They
24106 all default to @samp{auto}. For more information about each packet,
24107 see @ref{Remote Protocol}.
24109 During normal use, you should not have to use any of these commands.
24110 If you do, that may be a bug in your remote debugging stub, or a bug
24111 in @value{GDBN}. You may want to report the problem to the
24112 @value{GDBN} developers.
24114 For each packet @var{name}, the command to enable or disable the
24115 packet is @code{set remote @var{name}-packet}. If you configure a packet, the
24116 configuration will apply for all future remote targets if no target is selected.
24117 In case there is a target selected, only the configuration of the current target
24118 is changed. All other existing remote targets' features are not affected.
24119 The command to print the current configuration of a packet is
24120 @code{show remote @var{name}-packet}. It displays the current remote target's
24121 configuration. If no remote target is selected, the default configuration for
24122 future connections is shown. The available settings are:
24124 @multitable @columnfractions 0.28 0.32 0.25
24127 @tab Related Features
24129 @item @code{fetch-register}
24131 @tab @code{info registers}
24133 @item @code{set-register}
24137 @item @code{binary-download}
24139 @tab @code{load}, @code{set}
24141 @item @code{read-aux-vector}
24142 @tab @code{qXfer:auxv:read}
24143 @tab @code{info auxv}
24145 @item @code{symbol-lookup}
24146 @tab @code{qSymbol}
24147 @tab Detecting multiple threads
24149 @item @code{attach}
24150 @tab @code{vAttach}
24153 @item @code{verbose-resume}
24155 @tab Stepping or resuming multiple threads
24161 @item @code{software-breakpoint}
24165 @item @code{hardware-breakpoint}
24169 @item @code{write-watchpoint}
24173 @item @code{read-watchpoint}
24177 @item @code{access-watchpoint}
24181 @item @code{pid-to-exec-file}
24182 @tab @code{qXfer:exec-file:read}
24183 @tab @code{attach}, @code{run}
24185 @item @code{target-features}
24186 @tab @code{qXfer:features:read}
24187 @tab @code{set architecture}
24189 @item @code{library-info}
24190 @tab @code{qXfer:libraries:read}
24191 @tab @code{info sharedlibrary}
24193 @item @code{memory-map}
24194 @tab @code{qXfer:memory-map:read}
24195 @tab @code{info mem}
24197 @item @code{read-sdata-object}
24198 @tab @code{qXfer:sdata:read}
24199 @tab @code{print $_sdata}
24201 @item @code{read-siginfo-object}
24202 @tab @code{qXfer:siginfo:read}
24203 @tab @code{print $_siginfo}
24205 @item @code{write-siginfo-object}
24206 @tab @code{qXfer:siginfo:write}
24207 @tab @code{set $_siginfo}
24209 @item @code{threads}
24210 @tab @code{qXfer:threads:read}
24211 @tab @code{info threads}
24213 @item @code{get-thread-local-@*storage-address}
24214 @tab @code{qGetTLSAddr}
24215 @tab Displaying @code{__thread} variables
24217 @item @code{get-thread-information-block-address}
24218 @tab @code{qGetTIBAddr}
24219 @tab Display MS-Windows Thread Information Block.
24221 @item @code{search-memory}
24222 @tab @code{qSearch:memory}
24225 @item @code{supported-packets}
24226 @tab @code{qSupported}
24227 @tab Remote communications parameters
24229 @item @code{catch-syscalls}
24230 @tab @code{QCatchSyscalls}
24231 @tab @code{catch syscall}
24233 @item @code{pass-signals}
24234 @tab @code{QPassSignals}
24235 @tab @code{handle @var{signal}}
24237 @item @code{program-signals}
24238 @tab @code{QProgramSignals}
24239 @tab @code{handle @var{signal}}
24241 @item @code{hostio-close-packet}
24242 @tab @code{vFile:close}
24243 @tab @code{remote get}, @code{remote put}
24245 @item @code{hostio-open-packet}
24246 @tab @code{vFile:open}
24247 @tab @code{remote get}, @code{remote put}
24249 @item @code{hostio-pread-packet}
24250 @tab @code{vFile:pread}
24251 @tab @code{remote get}, @code{remote put}
24253 @item @code{hostio-pwrite-packet}
24254 @tab @code{vFile:pwrite}
24255 @tab @code{remote get}, @code{remote put}
24257 @item @code{hostio-unlink-packet}
24258 @tab @code{vFile:unlink}
24259 @tab @code{remote delete}
24261 @item @code{hostio-readlink-packet}
24262 @tab @code{vFile:readlink}
24265 @item @code{hostio-fstat-packet}
24266 @tab @code{vFile:fstat}
24269 @item @code{hostio-setfs-packet}
24270 @tab @code{vFile:setfs}
24273 @item @code{noack-packet}
24274 @tab @code{QStartNoAckMode}
24275 @tab Packet acknowledgment
24277 @item @code{osdata}
24278 @tab @code{qXfer:osdata:read}
24279 @tab @code{info os}
24281 @item @code{query-attached}
24282 @tab @code{qAttached}
24283 @tab Querying remote process attach state.
24285 @item @code{trace-buffer-size}
24286 @tab @code{QTBuffer:size}
24287 @tab @code{set trace-buffer-size}
24289 @item @code{trace-status}
24290 @tab @code{qTStatus}
24291 @tab @code{tstatus}
24293 @item @code{traceframe-info}
24294 @tab @code{qXfer:traceframe-info:read}
24295 @tab Traceframe info
24297 @item @code{install-in-trace}
24298 @tab @code{InstallInTrace}
24299 @tab Install tracepoint in tracing
24301 @item @code{disable-randomization}
24302 @tab @code{QDisableRandomization}
24303 @tab @code{set disable-randomization}
24305 @item @code{startup-with-shell}
24306 @tab @code{QStartupWithShell}
24307 @tab @code{set startup-with-shell}
24309 @item @code{environment-hex-encoded}
24310 @tab @code{QEnvironmentHexEncoded}
24311 @tab @code{set environment}
24313 @item @code{environment-unset}
24314 @tab @code{QEnvironmentUnset}
24315 @tab @code{unset environment}
24317 @item @code{environment-reset}
24318 @tab @code{QEnvironmentReset}
24319 @tab @code{Reset the inferior environment (i.e., unset user-set variables)}
24321 @item @code{set-working-dir}
24322 @tab @code{QSetWorkingDir}
24323 @tab @code{set cwd}
24325 @item @code{conditional-breakpoints-packet}
24326 @tab @code{Z0 and Z1}
24327 @tab @code{Support for target-side breakpoint condition evaluation}
24329 @item @code{multiprocess-extensions}
24330 @tab @code{multiprocess extensions}
24331 @tab Debug multiple processes and remote process PID awareness
24333 @item @code{swbreak-feature}
24334 @tab @code{swbreak stop reason}
24337 @item @code{hwbreak-feature}
24338 @tab @code{hwbreak stop reason}
24341 @item @code{fork-event-feature}
24342 @tab @code{fork stop reason}
24345 @item @code{vfork-event-feature}
24346 @tab @code{vfork stop reason}
24349 @item @code{exec-event-feature}
24350 @tab @code{exec stop reason}
24353 @item @code{thread-events}
24354 @tab @code{QThreadEvents}
24355 @tab Tracking thread lifetime.
24357 @item @code{no-resumed-stop-reply}
24358 @tab @code{no resumed thread left stop reply}
24359 @tab Tracking thread lifetime.
24363 @cindex packet size, remote, configuring
24364 The number of bytes per memory-read or memory-write packet for a remote target
24365 can be configured using the commands
24366 @w{@code{set remote memory-read-packet-size}} and
24367 @w{@code{set remote memory-write-packet-size}}. If set to @samp{0} (zero) the
24368 default packet size will be used. The actual limit is further reduced depending
24369 on the target. Specify @samp{fixed} to disable the target-dependent restriction
24370 and @samp{limit} to enable it. Similar to the enabling and disabling of remote
24371 packets, the command applies to the currently selected target (if available).
24372 If no remote target is selected, it applies to all future remote connections.
24373 The configuration of the selected target can be displayed using the commands
24374 @w{@code{show remote memory-read-packet-size}} and
24375 @w{@code{show remote memory-write-packet-size}}. If no remote target is
24376 selected, the default configuration for future connections is shown.
24379 @section Implementing a Remote Stub
24381 @cindex debugging stub, example
24382 @cindex remote stub, example
24383 @cindex stub example, remote debugging
24384 The stub files provided with @value{GDBN} implement the target side of the
24385 communication protocol, and the @value{GDBN} side is implemented in the
24386 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
24387 these subroutines to communicate, and ignore the details. (If you're
24388 implementing your own stub file, you can still ignore the details: start
24389 with one of the existing stub files. @file{sparc-stub.c} is the best
24390 organized, and therefore the easiest to read.)
24392 @cindex remote serial debugging, overview
24393 To debug a program running on another machine (the debugging
24394 @dfn{target} machine), you must first arrange for all the usual
24395 prerequisites for the program to run by itself. For example, for a C
24400 A startup routine to set up the C runtime environment; these usually
24401 have a name like @file{crt0}. The startup routine may be supplied by
24402 your hardware supplier, or you may have to write your own.
24405 A C subroutine library to support your program's
24406 subroutine calls, notably managing input and output.
24409 A way of getting your program to the other machine---for example, a
24410 download program. These are often supplied by the hardware
24411 manufacturer, but you may have to write your own from hardware
24415 The next step is to arrange for your program to use a serial port to
24416 communicate with the machine where @value{GDBN} is running (the @dfn{host}
24417 machine). In general terms, the scheme looks like this:
24421 @value{GDBN} already understands how to use this protocol; when everything
24422 else is set up, you can simply use the @samp{target remote} command
24423 (@pxref{Targets,,Specifying a Debugging Target}).
24425 @item On the target,
24426 you must link with your program a few special-purpose subroutines that
24427 implement the @value{GDBN} remote serial protocol. The file containing these
24428 subroutines is called a @dfn{debugging stub}.
24430 On certain remote targets, you can use an auxiliary program
24431 @code{gdbserver} instead of linking a stub into your program.
24432 @xref{Server,,Using the @code{gdbserver} Program}, for details.
24435 The debugging stub is specific to the architecture of the remote
24436 machine; for example, use @file{sparc-stub.c} to debug programs on
24439 @cindex remote serial stub list
24440 These working remote stubs are distributed with @value{GDBN}:
24445 @cindex @file{i386-stub.c}
24448 For Intel 386 and compatible architectures.
24451 @cindex @file{m68k-stub.c}
24452 @cindex Motorola 680x0
24454 For Motorola 680x0 architectures.
24457 @cindex @file{sh-stub.c}
24460 For Renesas SH architectures.
24463 @cindex @file{sparc-stub.c}
24465 For @sc{sparc} architectures.
24467 @item sparcl-stub.c
24468 @cindex @file{sparcl-stub.c}
24471 For Fujitsu @sc{sparclite} architectures.
24475 The @file{README} file in the @value{GDBN} distribution may list other
24476 recently added stubs.
24479 * Stub Contents:: What the stub can do for you
24480 * Bootstrapping:: What you must do for the stub
24481 * Debug Session:: Putting it all together
24484 @node Stub Contents
24485 @subsection What the Stub Can Do for You
24487 @cindex remote serial stub
24488 The debugging stub for your architecture supplies these three
24492 @findex set_debug_traps
24493 @item set_debug_traps
24494 @cindex remote serial stub, initialization
24495 This routine arranges for @code{handle_exception} to run when your
24496 program stops. You must call this subroutine explicitly in your
24497 program's startup code.
24499 @findex handle_exception
24500 @item handle_exception
24501 @cindex remote serial stub, main routine
24502 This is the central workhorse, but your program never calls it
24503 explicitly---the setup code arranges for @code{handle_exception} to
24504 run when a trap is triggered.
24506 @code{handle_exception} takes control when your program stops during
24507 execution (for example, on a breakpoint), and mediates communications
24508 with @value{GDBN} on the host machine. This is where the communications
24509 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
24510 representative on the target machine. It begins by sending summary
24511 information on the state of your program, then continues to execute,
24512 retrieving and transmitting any information @value{GDBN} needs, until you
24513 execute a @value{GDBN} command that makes your program resume; at that point,
24514 @code{handle_exception} returns control to your own code on the target
24518 @cindex @code{breakpoint} subroutine, remote
24519 Use this auxiliary subroutine to make your program contain a
24520 breakpoint. Depending on the particular situation, this may be the only
24521 way for @value{GDBN} to get control. For instance, if your target
24522 machine has some sort of interrupt button, you won't need to call this;
24523 pressing the interrupt button transfers control to
24524 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
24525 simply receiving characters on the serial port may also trigger a trap;
24526 again, in that situation, you don't need to call @code{breakpoint} from
24527 your own program---simply running @samp{target remote} from the host
24528 @value{GDBN} session gets control.
24530 Call @code{breakpoint} if none of these is true, or if you simply want
24531 to make certain your program stops at a predetermined point for the
24532 start of your debugging session.
24535 @node Bootstrapping
24536 @subsection What You Must Do for the Stub
24538 @cindex remote stub, support routines
24539 The debugging stubs that come with @value{GDBN} are set up for a particular
24540 chip architecture, but they have no information about the rest of your
24541 debugging target machine.
24543 First of all you need to tell the stub how to communicate with the
24547 @findex getDebugChar
24548 @item int getDebugChar()
24549 Write this subroutine to read a single character from the serial port.
24550 It may be identical to @code{getchar} for your target system; a
24551 different name is used to allow you to distinguish the two if you wish.
24553 @findex putDebugChar
24554 @item void putDebugChar(int)
24555 Write this subroutine to write a single character to the serial port.
24556 It may be identical to @code{putchar} for your target system; a
24557 different name is used to allow you to distinguish the two if you wish.
24560 @cindex control C, and remote debugging
24561 @cindex interrupting remote targets
24562 If you want @value{GDBN} to be able to stop your program while it is
24563 running, you need to use an interrupt-driven serial driver, and arrange
24564 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
24565 character). That is the character which @value{GDBN} uses to tell the
24566 remote system to stop.
24568 Getting the debugging target to return the proper status to @value{GDBN}
24569 probably requires changes to the standard stub; one quick and dirty way
24570 is to just execute a breakpoint instruction (the ``dirty'' part is that
24571 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
24573 Other routines you need to supply are:
24576 @findex exceptionHandler
24577 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
24578 Write this function to install @var{exception_address} in the exception
24579 handling tables. You need to do this because the stub does not have any
24580 way of knowing what the exception handling tables on your target system
24581 are like (for example, the processor's table might be in @sc{rom},
24582 containing entries which point to a table in @sc{ram}).
24583 The @var{exception_number} specifies the exception which should be changed;
24584 its meaning is architecture-dependent (for example, different numbers
24585 might represent divide by zero, misaligned access, etc). When this
24586 exception occurs, control should be transferred directly to
24587 @var{exception_address}, and the processor state (stack, registers,
24588 and so on) should be just as it is when a processor exception occurs. So if
24589 you want to use a jump instruction to reach @var{exception_address}, it
24590 should be a simple jump, not a jump to subroutine.
24592 For the 386, @var{exception_address} should be installed as an interrupt
24593 gate so that interrupts are masked while the handler runs. The gate
24594 should be at privilege level 0 (the most privileged level). The
24595 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
24596 help from @code{exceptionHandler}.
24598 @findex flush_i_cache
24599 @item void flush_i_cache()
24600 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
24601 instruction cache, if any, on your target machine. If there is no
24602 instruction cache, this subroutine may be a no-op.
24604 On target machines that have instruction caches, @value{GDBN} requires this
24605 function to make certain that the state of your program is stable.
24609 You must also make sure this library routine is available:
24613 @item void *memset(void *, int, int)
24614 This is the standard library function @code{memset} that sets an area of
24615 memory to a known value. If you have one of the free versions of
24616 @code{libc.a}, @code{memset} can be found there; otherwise, you must
24617 either obtain it from your hardware manufacturer, or write your own.
24620 If you do not use the GNU C compiler, you may need other standard
24621 library subroutines as well; this varies from one stub to another,
24622 but in general the stubs are likely to use any of the common library
24623 subroutines which @code{@value{NGCC}} generates as inline code.
24626 @node Debug Session
24627 @subsection Putting it All Together
24629 @cindex remote serial debugging summary
24630 In summary, when your program is ready to debug, you must follow these
24635 Make sure you have defined the supporting low-level routines
24636 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
24638 @code{getDebugChar}, @code{putDebugChar},
24639 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
24643 Insert these lines in your program's startup code, before the main
24644 procedure is called:
24651 On some machines, when a breakpoint trap is raised, the hardware
24652 automatically makes the PC point to the instruction after the
24653 breakpoint. If your machine doesn't do that, you may need to adjust
24654 @code{handle_exception} to arrange for it to return to the instruction
24655 after the breakpoint on this first invocation, so that your program
24656 doesn't keep hitting the initial breakpoint instead of making
24660 For the 680x0 stub only, you need to provide a variable called
24661 @code{exceptionHook}. Normally you just use:
24664 void (*exceptionHook)() = 0;
24668 but if before calling @code{set_debug_traps}, you set it to point to a
24669 function in your program, that function is called when
24670 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
24671 error). The function indicated by @code{exceptionHook} is called with
24672 one parameter: an @code{int} which is the exception number.
24675 Compile and link together: your program, the @value{GDBN} debugging stub for
24676 your target architecture, and the supporting subroutines.
24679 Make sure you have a serial connection between your target machine and
24680 the @value{GDBN} host, and identify the serial port on the host.
24683 @c The "remote" target now provides a `load' command, so we should
24684 @c document that. FIXME.
24685 Download your program to your target machine (or get it there by
24686 whatever means the manufacturer provides), and start it.
24689 Start @value{GDBN} on the host, and connect to the target
24690 (@pxref{Connecting,,Connecting to a Remote Target}).
24694 @node Configurations
24695 @chapter Configuration-Specific Information
24697 While nearly all @value{GDBN} commands are available for all native and
24698 cross versions of the debugger, there are some exceptions. This chapter
24699 describes things that are only available in certain configurations.
24701 There are three major categories of configurations: native
24702 configurations, where the host and target are the same, embedded
24703 operating system configurations, which are usually the same for several
24704 different processor architectures, and bare embedded processors, which
24705 are quite different from each other.
24710 * Embedded Processors::
24717 This section describes details specific to particular native
24721 * BSD libkvm Interface:: Debugging BSD kernel memory images
24722 * Process Information:: Process information
24723 * DJGPP Native:: Features specific to the DJGPP port
24724 * Cygwin Native:: Features specific to the Cygwin port
24725 * Hurd Native:: Features specific to @sc{gnu} Hurd
24726 * Darwin:: Features specific to Darwin
24727 * FreeBSD:: Features specific to FreeBSD
24730 @node BSD libkvm Interface
24731 @subsection BSD libkvm Interface
24734 @cindex kernel memory image
24735 @cindex kernel crash dump
24737 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
24738 interface that provides a uniform interface for accessing kernel virtual
24739 memory images, including live systems and crash dumps. @value{GDBN}
24740 uses this interface to allow you to debug live kernels and kernel crash
24741 dumps on many native BSD configurations. This is implemented as a
24742 special @code{kvm} debugging target. For debugging a live system, load
24743 the currently running kernel into @value{GDBN} and connect to the
24747 (@value{GDBP}) @b{target kvm}
24750 For debugging crash dumps, provide the file name of the crash dump as an
24754 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
24757 Once connected to the @code{kvm} target, the following commands are
24763 Set current context from the @dfn{Process Control Block} (PCB) address.
24766 Set current context from proc address. This command isn't available on
24767 modern FreeBSD systems.
24770 @node Process Information
24771 @subsection Process Information
24773 @cindex examine process image
24774 @cindex process info via @file{/proc}
24776 Some operating systems provide interfaces to fetch additional
24777 information about running processes beyond memory and per-thread
24778 register state. If @value{GDBN} is configured for an operating system
24779 with a supported interface, the command @code{info proc} is available
24780 to report information about the process running your program, or about
24781 any process running on your system.
24783 One supported interface is a facility called @samp{/proc} that can be
24784 used to examine the image of a running process using file-system
24785 subroutines. This facility is supported on @sc{gnu}/Linux and Solaris
24788 On FreeBSD and NetBSD systems, system control nodes are used to query
24789 process information.
24791 In addition, some systems may provide additional process information
24792 in core files. Note that a core file may include a subset of the
24793 information available from a live process. Process information is
24794 currently available from cores created on @sc{gnu}/Linux and FreeBSD
24801 @itemx info proc @var{process-id}
24802 Summarize available information about a process. If a
24803 process ID is specified by @var{process-id}, display information about
24804 that process; otherwise display information about the program being
24805 debugged. The summary includes the debugged process ID, the command
24806 line used to invoke it, its current working directory, and its
24807 executable file's absolute file name.
24809 On some systems, @var{process-id} can be of the form
24810 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
24811 within a process. If the optional @var{pid} part is missing, it means
24812 a thread from the process being debugged (the leading @samp{/} still
24813 needs to be present, or else @value{GDBN} will interpret the number as
24814 a process ID rather than a thread ID).
24816 @item info proc cmdline
24817 @cindex info proc cmdline
24818 Show the original command line of the process. This command is
24819 supported on @sc{gnu}/Linux, FreeBSD and NetBSD.
24821 @item info proc cwd
24822 @cindex info proc cwd
24823 Show the current working directory of the process. This command is
24824 supported on @sc{gnu}/Linux, FreeBSD and NetBSD.
24826 @item info proc exe
24827 @cindex info proc exe
24828 Show the name of executable of the process. This command is supported
24829 on @sc{gnu}/Linux, FreeBSD and NetBSD.
24831 @item info proc files
24832 @cindex info proc files
24833 Show the file descriptors open by the process. For each open file
24834 descriptor, @value{GDBN} shows its number, type (file, directory,
24835 character device, socket), file pointer offset, and the name of the
24836 resource open on the descriptor. The resource name can be a file name
24837 (for files, directories, and devices) or a protocol followed by socket
24838 address (for network connections). This command is supported on
24841 This example shows the open file descriptors for a process using a
24842 tty for standard input and output as well as two network sockets:
24845 (@value{GDBP}) info proc files 22136
24849 FD Type Offset Flags Name
24850 text file - r-------- /usr/bin/ssh
24851 ctty chr - rw------- /dev/pts/20
24852 cwd dir - r-------- /usr/home/john
24853 root dir - r-------- /
24854 0 chr 0x32933a4 rw------- /dev/pts/20
24855 1 chr 0x32933a4 rw------- /dev/pts/20
24856 2 chr 0x32933a4 rw------- /dev/pts/20
24857 3 socket 0x0 rw----n-- tcp4 10.0.1.2:53014 -> 10.0.1.10:22
24858 4 socket 0x0 rw------- unix stream:/tmp/ssh-FIt89oAzOn5f/agent.2456
24861 @item info proc mappings
24862 @cindex memory address space mappings
24863 Report the memory address space ranges accessible in a process. On
24864 Solaris, FreeBSD and NetBSD systems, each memory range includes information
24865 on whether the process has read, write, or execute access rights to each
24866 range. On @sc{gnu}/Linux, FreeBSD and NetBSD systems, each memory range
24867 includes the object file which is mapped to that range.
24869 @item info proc stat
24870 @itemx info proc status
24871 @cindex process detailed status information
24872 Show additional process-related information, including the user ID and
24873 group ID; virtual memory usage; the signals that are pending, blocked,
24874 and ignored; its TTY; its consumption of system and user time; its
24875 stack size; its @samp{nice} value; etc. These commands are supported
24876 on @sc{gnu}/Linux, FreeBSD and NetBSD.
24878 For @sc{gnu}/Linux systems, see the @samp{proc} man page for more
24879 information (type @kbd{man 5 proc} from your shell prompt).
24881 For FreeBSD and NetBSD systems, @code{info proc stat} is an alias for
24882 @code{info proc status}.
24884 @item info proc all
24885 Show all the information about the process described under all of the
24886 above @code{info proc} subcommands.
24889 @comment These sub-options of 'info proc' were not included when
24890 @comment procfs.c was re-written. Keep their descriptions around
24891 @comment against the day when someone finds the time to put them back in.
24892 @kindex info proc times
24893 @item info proc times
24894 Starting time, user CPU time, and system CPU time for your program and
24897 @kindex info proc id
24899 Report on the process IDs related to your program: its own process ID,
24900 the ID of its parent, the process group ID, and the session ID.
24903 @item set procfs-trace
24904 @kindex set procfs-trace
24905 @cindex @code{procfs} API calls
24906 This command enables and disables tracing of @code{procfs} API calls.
24908 @item show procfs-trace
24909 @kindex show procfs-trace
24910 Show the current state of @code{procfs} API call tracing.
24912 @item set procfs-file @var{file}
24913 @kindex set procfs-file
24914 Tell @value{GDBN} to write @code{procfs} API trace to the named
24915 @var{file}. @value{GDBN} appends the trace info to the previous
24916 contents of the file. The default is to display the trace on the
24919 @item show procfs-file
24920 @kindex show procfs-file
24921 Show the file to which @code{procfs} API trace is written.
24923 @item proc-trace-entry
24924 @itemx proc-trace-exit
24925 @itemx proc-untrace-entry
24926 @itemx proc-untrace-exit
24927 @kindex proc-trace-entry
24928 @kindex proc-trace-exit
24929 @kindex proc-untrace-entry
24930 @kindex proc-untrace-exit
24931 These commands enable and disable tracing of entries into and exits
24932 from the @code{syscall} interface.
24935 @kindex info pidlist
24936 @cindex process list, QNX Neutrino
24937 For QNX Neutrino only, this command displays the list of all the
24938 processes and all the threads within each process.
24941 @kindex info meminfo
24942 @cindex mapinfo list, QNX Neutrino
24943 For QNX Neutrino only, this command displays the list of all mapinfos.
24947 @subsection Features for Debugging @sc{djgpp} Programs
24948 @cindex @sc{djgpp} debugging
24949 @cindex native @sc{djgpp} debugging
24950 @cindex MS-DOS-specific commands
24953 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
24954 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
24955 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
24956 top of real-mode DOS systems and their emulations.
24958 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
24959 defines a few commands specific to the @sc{djgpp} port. This
24960 subsection describes those commands.
24965 This is a prefix of @sc{djgpp}-specific commands which print
24966 information about the target system and important OS structures.
24969 @cindex MS-DOS system info
24970 @cindex free memory information (MS-DOS)
24971 @item info dos sysinfo
24972 This command displays assorted information about the underlying
24973 platform: the CPU type and features, the OS version and flavor, the
24974 DPMI version, and the available conventional and DPMI memory.
24979 @cindex segment descriptor tables
24980 @cindex descriptor tables display
24982 @itemx info dos ldt
24983 @itemx info dos idt
24984 These 3 commands display entries from, respectively, Global, Local,
24985 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
24986 tables are data structures which store a descriptor for each segment
24987 that is currently in use. The segment's selector is an index into a
24988 descriptor table; the table entry for that index holds the
24989 descriptor's base address and limit, and its attributes and access
24992 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
24993 segment (used for both data and the stack), and a DOS segment (which
24994 allows access to DOS/BIOS data structures and absolute addresses in
24995 conventional memory). However, the DPMI host will usually define
24996 additional segments in order to support the DPMI environment.
24998 @cindex garbled pointers
24999 These commands allow to display entries from the descriptor tables.
25000 Without an argument, all entries from the specified table are
25001 displayed. An argument, which should be an integer expression, means
25002 display a single entry whose index is given by the argument. For
25003 example, here's a convenient way to display information about the
25004 debugged program's data segment:
25007 @exdent @code{(@value{GDBP}) info dos ldt $ds}
25008 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
25012 This comes in handy when you want to see whether a pointer is outside
25013 the data segment's limit (i.e.@: @dfn{garbled}).
25015 @cindex page tables display (MS-DOS)
25017 @itemx info dos pte
25018 These two commands display entries from, respectively, the Page
25019 Directory and the Page Tables. Page Directories and Page Tables are
25020 data structures which control how virtual memory addresses are mapped
25021 into physical addresses. A Page Table includes an entry for every
25022 page of memory that is mapped into the program's address space; there
25023 may be several Page Tables, each one holding up to 4096 entries. A
25024 Page Directory has up to 4096 entries, one each for every Page Table
25025 that is currently in use.
25027 Without an argument, @kbd{info dos pde} displays the entire Page
25028 Directory, and @kbd{info dos pte} displays all the entries in all of
25029 the Page Tables. An argument, an integer expression, given to the
25030 @kbd{info dos pde} command means display only that entry from the Page
25031 Directory table. An argument given to the @kbd{info dos pte} command
25032 means display entries from a single Page Table, the one pointed to by
25033 the specified entry in the Page Directory.
25035 @cindex direct memory access (DMA) on MS-DOS
25036 These commands are useful when your program uses @dfn{DMA} (Direct
25037 Memory Access), which needs physical addresses to program the DMA
25040 These commands are supported only with some DPMI servers.
25042 @cindex physical address from linear address
25043 @item info dos address-pte @var{addr}
25044 This command displays the Page Table entry for a specified linear
25045 address. The argument @var{addr} is a linear address which should
25046 already have the appropriate segment's base address added to it,
25047 because this command accepts addresses which may belong to @emph{any}
25048 segment. For example, here's how to display the Page Table entry for
25049 the page where a variable @code{i} is stored:
25052 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
25053 @exdent @code{Page Table entry for address 0x11a00d30:}
25054 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
25058 This says that @code{i} is stored at offset @code{0xd30} from the page
25059 whose physical base address is @code{0x02698000}, and shows all the
25060 attributes of that page.
25062 Note that you must cast the addresses of variables to a @code{char *},
25063 since otherwise the value of @code{__djgpp_base_address}, the base
25064 address of all variables and functions in a @sc{djgpp} program, will
25065 be added using the rules of C pointer arithmetics: if @code{i} is
25066 declared an @code{int}, @value{GDBN} will add 4 times the value of
25067 @code{__djgpp_base_address} to the address of @code{i}.
25069 Here's another example, it displays the Page Table entry for the
25073 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
25074 @exdent @code{Page Table entry for address 0x29110:}
25075 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
25079 (The @code{+ 3} offset is because the transfer buffer's address is the
25080 3rd member of the @code{_go32_info_block} structure.) The output
25081 clearly shows that this DPMI server maps the addresses in conventional
25082 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
25083 linear (@code{0x29110}) addresses are identical.
25085 This command is supported only with some DPMI servers.
25088 @cindex DOS serial data link, remote debugging
25089 In addition to native debugging, the DJGPP port supports remote
25090 debugging via a serial data link. The following commands are specific
25091 to remote serial debugging in the DJGPP port of @value{GDBN}.
25094 @kindex set com1base
25095 @kindex set com1irq
25096 @kindex set com2base
25097 @kindex set com2irq
25098 @kindex set com3base
25099 @kindex set com3irq
25100 @kindex set com4base
25101 @kindex set com4irq
25102 @item set com1base @var{addr}
25103 This command sets the base I/O port address of the @file{COM1} serial
25106 @item set com1irq @var{irq}
25107 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
25108 for the @file{COM1} serial port.
25110 There are similar commands @samp{set com2base}, @samp{set com3irq},
25111 etc.@: for setting the port address and the @code{IRQ} lines for the
25114 @kindex show com1base
25115 @kindex show com1irq
25116 @kindex show com2base
25117 @kindex show com2irq
25118 @kindex show com3base
25119 @kindex show com3irq
25120 @kindex show com4base
25121 @kindex show com4irq
25122 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
25123 display the current settings of the base address and the @code{IRQ}
25124 lines used by the COM ports.
25127 @kindex info serial
25128 @cindex DOS serial port status
25129 This command prints the status of the 4 DOS serial ports. For each
25130 port, it prints whether it's active or not, its I/O base address and
25131 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
25132 counts of various errors encountered so far.
25136 @node Cygwin Native
25137 @subsection Features for Debugging MS Windows PE Executables
25138 @cindex MS Windows debugging
25139 @cindex native Cygwin debugging
25140 @cindex Cygwin-specific commands
25142 @value{GDBN} supports native debugging of MS Windows programs, including
25143 DLLs with and without symbolic debugging information.
25145 @cindex Ctrl-BREAK, MS-Windows
25146 @cindex interrupt debuggee on MS-Windows
25147 MS-Windows programs that call @code{SetConsoleMode} to switch off the
25148 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
25149 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
25150 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
25151 sequence, which can be used to interrupt the debuggee even if it
25154 There are various additional Cygwin-specific commands, described in
25155 this section. Working with DLLs that have no debugging symbols is
25156 described in @ref{Non-debug DLL Symbols}.
25161 This is a prefix of MS Windows-specific commands which print
25162 information about the target system and important OS structures.
25164 @item info w32 selector
25165 This command displays information returned by
25166 the Win32 API @code{GetThreadSelectorEntry} function.
25167 It takes an optional argument that is evaluated to
25168 a long value to give the information about this given selector.
25169 Without argument, this command displays information
25170 about the six segment registers.
25172 @item info w32 thread-information-block
25173 This command displays thread specific information stored in the
25174 Thread Information Block (readable on the X86 CPU family using @code{$fs}
25175 selector for 32-bit programs and @code{$gs} for 64-bit programs).
25177 @kindex signal-event
25178 @item signal-event @var{id}
25179 This command signals an event with user-provided @var{id}. Used to resume
25180 crashing process when attached to it using MS-Windows JIT debugging (AeDebug).
25182 To use it, create or edit the following keys in
25183 @code{HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\AeDebug} and/or
25184 @code{HKLM\SOFTWARE\Wow6432Node\Microsoft\Windows NT\CurrentVersion\AeDebug}
25185 (for x86_64 versions):
25189 @code{Debugger} (REG_SZ) --- a command to launch the debugger.
25190 Suggested command is: @code{@var{fully-qualified-path-to-gdb.exe} -ex
25191 "attach %ld" -ex "signal-event %ld" -ex "continue"}.
25193 The first @code{%ld} will be replaced by the process ID of the
25194 crashing process, the second @code{%ld} will be replaced by the ID of
25195 the event that blocks the crashing process, waiting for @value{GDBN}
25199 @code{Auto} (REG_SZ) --- either @code{1} or @code{0}. @code{1} will
25200 make the system run debugger specified by the Debugger key
25201 automatically, @code{0} will cause a dialog box with ``OK'' and
25202 ``Cancel'' buttons to appear, which allows the user to either
25203 terminate the crashing process (OK) or debug it (Cancel).
25206 @kindex set cygwin-exceptions
25207 @cindex debugging the Cygwin DLL
25208 @cindex Cygwin DLL, debugging
25209 @item set cygwin-exceptions @var{mode}
25210 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
25211 happen inside the Cygwin DLL. If @var{mode} is @code{off},
25212 @value{GDBN} will delay recognition of exceptions, and may ignore some
25213 exceptions which seem to be caused by internal Cygwin DLL
25214 ``bookkeeping''. This option is meant primarily for debugging the
25215 Cygwin DLL itself; the default value is @code{off} to avoid annoying
25216 @value{GDBN} users with false @code{SIGSEGV} signals.
25218 @kindex show cygwin-exceptions
25219 @item show cygwin-exceptions
25220 Displays whether @value{GDBN} will break on exceptions that happen
25221 inside the Cygwin DLL itself.
25223 @kindex set new-console
25224 @item set new-console @var{mode}
25225 If @var{mode} is @code{on} the debuggee will
25226 be started in a new console on next start.
25227 If @var{mode} is @code{off}, the debuggee will
25228 be started in the same console as the debugger.
25230 @kindex show new-console
25231 @item show new-console
25232 Displays whether a new console is used
25233 when the debuggee is started.
25235 @kindex set new-group
25236 @item set new-group @var{mode}
25237 This boolean value controls whether the debuggee should
25238 start a new group or stay in the same group as the debugger.
25239 This affects the way the Windows OS handles
25242 @kindex show new-group
25243 @item show new-group
25244 Displays current value of new-group boolean.
25246 @kindex set debugevents
25247 @item set debugevents
25248 This boolean value adds debug output concerning kernel events related
25249 to the debuggee seen by the debugger. This includes events that
25250 signal thread and process creation and exit, DLL loading and
25251 unloading, console interrupts, and debugging messages produced by the
25252 Windows @code{OutputDebugString} API call.
25254 @kindex set debugexec
25255 @item set debugexec
25256 This boolean value adds debug output concerning execute events
25257 (such as resume thread) seen by the debugger.
25259 @kindex set debugexceptions
25260 @item set debugexceptions
25261 This boolean value adds debug output concerning exceptions in the
25262 debuggee seen by the debugger.
25264 @kindex set debugmemory
25265 @item set debugmemory
25266 This boolean value adds debug output concerning debuggee memory reads
25267 and writes by the debugger.
25271 This boolean values specifies whether the debuggee is called
25272 via a shell or directly (default value is on).
25276 Displays if the debuggee will be started with a shell.
25281 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
25284 @node Non-debug DLL Symbols
25285 @subsubsection Support for DLLs without Debugging Symbols
25286 @cindex DLLs with no debugging symbols
25287 @cindex Minimal symbols and DLLs
25289 Very often on windows, some of the DLLs that your program relies on do
25290 not include symbolic debugging information (for example,
25291 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
25292 symbols in a DLL, it relies on the minimal amount of symbolic
25293 information contained in the DLL's export table. This section
25294 describes working with such symbols, known internally to @value{GDBN} as
25295 ``minimal symbols''.
25297 Note that before the debugged program has started execution, no DLLs
25298 will have been loaded. The easiest way around this problem is simply to
25299 start the program --- either by setting a breakpoint or letting the
25300 program run once to completion.
25302 @subsubsection DLL Name Prefixes
25304 In keeping with the naming conventions used by the Microsoft debugging
25305 tools, DLL export symbols are made available with a prefix based on the
25306 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
25307 also entered into the symbol table, so @code{CreateFileA} is often
25308 sufficient. In some cases there will be name clashes within a program
25309 (particularly if the executable itself includes full debugging symbols)
25310 necessitating the use of the fully qualified name when referring to the
25311 contents of the DLL. Use single-quotes around the name to avoid the
25312 exclamation mark (``!'') being interpreted as a language operator.
25314 Note that the internal name of the DLL may be all upper-case, even
25315 though the file name of the DLL is lower-case, or vice-versa. Since
25316 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
25317 some confusion. If in doubt, try the @code{info functions} and
25318 @code{info variables} commands or even @code{maint print msymbols}
25319 (@pxref{Symbols}). Here's an example:
25322 (@value{GDBP}) info function CreateFileA
25323 All functions matching regular expression "CreateFileA":
25325 Non-debugging symbols:
25326 0x77e885f4 CreateFileA
25327 0x77e885f4 KERNEL32!CreateFileA
25331 (@value{GDBP}) info function !
25332 All functions matching regular expression "!":
25334 Non-debugging symbols:
25335 0x6100114c cygwin1!__assert
25336 0x61004034 cygwin1!_dll_crt0@@0
25337 0x61004240 cygwin1!dll_crt0(per_process *)
25341 @subsubsection Working with Minimal Symbols
25343 Symbols extracted from a DLL's export table do not contain very much
25344 type information. All that @value{GDBN} can do is guess whether a symbol
25345 refers to a function or variable depending on the linker section that
25346 contains the symbol. Also note that the actual contents of the memory
25347 contained in a DLL are not available unless the program is running. This
25348 means that you cannot examine the contents of a variable or disassemble
25349 a function within a DLL without a running program.
25351 Variables are generally treated as pointers and dereferenced
25352 automatically. For this reason, it is often necessary to prefix a
25353 variable name with the address-of operator (``&'') and provide explicit
25354 type information in the command. Here's an example of the type of
25358 (@value{GDBP}) print 'cygwin1!__argv'
25359 'cygwin1!__argv' has unknown type; cast it to its declared type
25363 (@value{GDBP}) x 'cygwin1!__argv'
25364 'cygwin1!__argv' has unknown type; cast it to its declared type
25367 And two possible solutions:
25370 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
25371 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
25375 (@value{GDBP}) x/2x &'cygwin1!__argv'
25376 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
25377 (@value{GDBP}) x/x 0x10021608
25378 0x10021608: 0x0022fd98
25379 (@value{GDBP}) x/s 0x0022fd98
25380 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
25383 Setting a break point within a DLL is possible even before the program
25384 starts execution. However, under these circumstances, @value{GDBN} can't
25385 examine the initial instructions of the function in order to skip the
25386 function's frame set-up code. You can work around this by using ``*&''
25387 to set the breakpoint at a raw memory address:
25390 (@value{GDBP}) break *&'python22!PyOS_Readline'
25391 Breakpoint 1 at 0x1e04eff0
25394 The author of these extensions is not entirely convinced that setting a
25395 break point within a shared DLL like @file{kernel32.dll} is completely
25399 @subsection Commands Specific to @sc{gnu} Hurd Systems
25400 @cindex @sc{gnu} Hurd debugging
25402 This subsection describes @value{GDBN} commands specific to the
25403 @sc{gnu} Hurd native debugging.
25408 @kindex set signals@r{, Hurd command}
25409 @kindex set sigs@r{, Hurd command}
25410 This command toggles the state of inferior signal interception by
25411 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
25412 affected by this command. @code{sigs} is a shorthand alias for
25417 @kindex show signals@r{, Hurd command}
25418 @kindex show sigs@r{, Hurd command}
25419 Show the current state of intercepting inferior's signals.
25421 @item set signal-thread
25422 @itemx set sigthread
25423 @kindex set signal-thread
25424 @kindex set sigthread
25425 This command tells @value{GDBN} which thread is the @code{libc} signal
25426 thread. That thread is run when a signal is delivered to a running
25427 process. @code{set sigthread} is the shorthand alias of @code{set
25430 @item show signal-thread
25431 @itemx show sigthread
25432 @kindex show signal-thread
25433 @kindex show sigthread
25434 These two commands show which thread will run when the inferior is
25435 delivered a signal.
25438 @kindex set stopped@r{, Hurd command}
25439 This commands tells @value{GDBN} that the inferior process is stopped,
25440 as with the @code{SIGSTOP} signal. The stopped process can be
25441 continued by delivering a signal to it.
25444 @kindex show stopped@r{, Hurd command}
25445 This command shows whether @value{GDBN} thinks the debuggee is
25448 @item set exceptions
25449 @kindex set exceptions@r{, Hurd command}
25450 Use this command to turn off trapping of exceptions in the inferior.
25451 When exception trapping is off, neither breakpoints nor
25452 single-stepping will work. To restore the default, set exception
25455 @item show exceptions
25456 @kindex show exceptions@r{, Hurd command}
25457 Show the current state of trapping exceptions in the inferior.
25459 @item set task pause
25460 @kindex set task@r{, Hurd commands}
25461 @cindex task attributes (@sc{gnu} Hurd)
25462 @cindex pause current task (@sc{gnu} Hurd)
25463 This command toggles task suspension when @value{GDBN} has control.
25464 Setting it to on takes effect immediately, and the task is suspended
25465 whenever @value{GDBN} gets control. Setting it to off will take
25466 effect the next time the inferior is continued. If this option is set
25467 to off, you can use @code{set thread default pause on} or @code{set
25468 thread pause on} (see below) to pause individual threads.
25470 @item show task pause
25471 @kindex show task@r{, Hurd commands}
25472 Show the current state of task suspension.
25474 @item set task detach-suspend-count
25475 @cindex task suspend count
25476 @cindex detach from task, @sc{gnu} Hurd
25477 This command sets the suspend count the task will be left with when
25478 @value{GDBN} detaches from it.
25480 @item show task detach-suspend-count
25481 Show the suspend count the task will be left with when detaching.
25483 @item set task exception-port
25484 @itemx set task excp
25485 @cindex task exception port, @sc{gnu} Hurd
25486 This command sets the task exception port to which @value{GDBN} will
25487 forward exceptions. The argument should be the value of the @dfn{send
25488 rights} of the task. @code{set task excp} is a shorthand alias.
25490 @item set noninvasive
25491 @cindex noninvasive task options
25492 This command switches @value{GDBN} to a mode that is the least
25493 invasive as far as interfering with the inferior is concerned. This
25494 is the same as using @code{set task pause}, @code{set exceptions}, and
25495 @code{set signals} to values opposite to the defaults.
25497 @item info send-rights
25498 @itemx info receive-rights
25499 @itemx info port-rights
25500 @itemx info port-sets
25501 @itemx info dead-names
25504 @cindex send rights, @sc{gnu} Hurd
25505 @cindex receive rights, @sc{gnu} Hurd
25506 @cindex port rights, @sc{gnu} Hurd
25507 @cindex port sets, @sc{gnu} Hurd
25508 @cindex dead names, @sc{gnu} Hurd
25509 These commands display information about, respectively, send rights,
25510 receive rights, port rights, port sets, and dead names of a task.
25511 There are also shorthand aliases: @code{info ports} for @code{info
25512 port-rights} and @code{info psets} for @code{info port-sets}.
25514 @item set thread pause
25515 @kindex set thread@r{, Hurd command}
25516 @cindex thread properties, @sc{gnu} Hurd
25517 @cindex pause current thread (@sc{gnu} Hurd)
25518 This command toggles current thread suspension when @value{GDBN} has
25519 control. Setting it to on takes effect immediately, and the current
25520 thread is suspended whenever @value{GDBN} gets control. Setting it to
25521 off will take effect the next time the inferior is continued.
25522 Normally, this command has no effect, since when @value{GDBN} has
25523 control, the whole task is suspended. However, if you used @code{set
25524 task pause off} (see above), this command comes in handy to suspend
25525 only the current thread.
25527 @item show thread pause
25528 @kindex show thread@r{, Hurd command}
25529 This command shows the state of current thread suspension.
25531 @item set thread run
25532 This command sets whether the current thread is allowed to run.
25534 @item show thread run
25535 Show whether the current thread is allowed to run.
25537 @item set thread detach-suspend-count
25538 @cindex thread suspend count, @sc{gnu} Hurd
25539 @cindex detach from thread, @sc{gnu} Hurd
25540 This command sets the suspend count @value{GDBN} will leave on a
25541 thread when detaching. This number is relative to the suspend count
25542 found by @value{GDBN} when it notices the thread; use @code{set thread
25543 takeover-suspend-count} to force it to an absolute value.
25545 @item show thread detach-suspend-count
25546 Show the suspend count @value{GDBN} will leave on the thread when
25549 @item set thread exception-port
25550 @itemx set thread excp
25551 Set the thread exception port to which to forward exceptions. This
25552 overrides the port set by @code{set task exception-port} (see above).
25553 @code{set thread excp} is the shorthand alias.
25555 @item set thread takeover-suspend-count
25556 Normally, @value{GDBN}'s thread suspend counts are relative to the
25557 value @value{GDBN} finds when it notices each thread. This command
25558 changes the suspend counts to be absolute instead.
25560 @item set thread default
25561 @itemx show thread default
25562 @cindex thread default settings, @sc{gnu} Hurd
25563 Each of the above @code{set thread} commands has a @code{set thread
25564 default} counterpart (e.g., @code{set thread default pause}, @code{set
25565 thread default exception-port}, etc.). The @code{thread default}
25566 variety of commands sets the default thread properties for all
25567 threads; you can then change the properties of individual threads with
25568 the non-default commands.
25575 @value{GDBN} provides the following commands specific to the Darwin target:
25578 @item set debug darwin @var{num}
25579 @kindex set debug darwin
25580 When set to a non zero value, enables debugging messages specific to
25581 the Darwin support. Higher values produce more verbose output.
25583 @item show debug darwin
25584 @kindex show debug darwin
25585 Show the current state of Darwin messages.
25587 @item set debug mach-o @var{num}
25588 @kindex set debug mach-o
25589 When set to a non zero value, enables debugging messages while
25590 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
25591 file format used on Darwin for object and executable files.) Higher
25592 values produce more verbose output. This is a command to diagnose
25593 problems internal to @value{GDBN} and should not be needed in normal
25596 @item show debug mach-o
25597 @kindex show debug mach-o
25598 Show the current state of Mach-O file messages.
25600 @item set mach-exceptions on
25601 @itemx set mach-exceptions off
25602 @kindex set mach-exceptions
25603 On Darwin, faults are first reported as a Mach exception and are then
25604 mapped to a Posix signal. Use this command to turn on trapping of
25605 Mach exceptions in the inferior. This might be sometimes useful to
25606 better understand the cause of a fault. The default is off.
25608 @item show mach-exceptions
25609 @kindex show mach-exceptions
25610 Show the current state of exceptions trapping.
25614 @subsection FreeBSD
25617 When the ABI of a system call is changed in the FreeBSD kernel, this
25618 is implemented by leaving a compatibility system call using the old
25619 ABI at the existing number and allocating a new system call number for
25620 the version using the new ABI. As a convenience, when a system call
25621 is caught by name (@pxref{catch syscall}), compatibility system calls
25624 For example, FreeBSD 12 introduced a new variant of the @code{kevent}
25625 system call and catching the @code{kevent} system call by name catches
25629 (@value{GDBP}) catch syscall kevent
25630 Catchpoint 1 (syscalls 'freebsd11_kevent' [363] 'kevent' [560])
25636 @section Embedded Operating Systems
25638 This section describes configurations involving the debugging of
25639 embedded operating systems that are available for several different
25642 @value{GDBN} includes the ability to debug programs running on
25643 various real-time operating systems.
25645 @node Embedded Processors
25646 @section Embedded Processors
25648 This section goes into details specific to particular embedded
25651 @cindex send command to simulator
25652 Whenever a specific embedded processor has a simulator, @value{GDBN}
25653 allows to send an arbitrary command to the simulator.
25656 @item sim @var{command}
25657 @kindex sim@r{, a command}
25658 Send an arbitrary @var{command} string to the simulator. Consult the
25659 documentation for the specific simulator in use for information about
25660 acceptable commands.
25665 * ARC:: Synopsys ARC
25668 * M68K:: Motorola M68K
25669 * MicroBlaze:: Xilinx MicroBlaze
25670 * MIPS Embedded:: MIPS Embedded
25671 * OpenRISC 1000:: OpenRISC 1000 (or1k)
25672 * PowerPC Embedded:: PowerPC Embedded
25675 * Super-H:: Renesas Super-H
25679 @subsection Synopsys ARC
25680 @cindex Synopsys ARC
25681 @cindex ARC specific commands
25687 @value{GDBN} provides the following ARC-specific commands:
25690 @item set debug arc
25691 @kindex set debug arc
25692 Control the level of ARC specific debug messages. Use 0 for no messages (the
25693 default), 1 for debug messages, and 2 for even more debug messages.
25695 @item show debug arc
25696 @kindex show debug arc
25697 Show the level of ARC specific debugging in operation.
25699 @item maint print arc arc-instruction @var{address}
25700 @kindex maint print arc arc-instruction
25701 Print internal disassembler information about instruction at a given address.
25708 @value{GDBN} provides the following ARM-specific commands:
25711 @item set arm disassembler
25713 This commands selects from a list of disassembly styles. The
25714 @code{"std"} style is the standard style.
25716 @item show arm disassembler
25718 Show the current disassembly style.
25720 @item set arm apcs32
25721 @cindex ARM 32-bit mode
25722 This command toggles ARM operation mode between 32-bit and 26-bit.
25724 @item show arm apcs32
25725 Display the current usage of the ARM 32-bit mode.
25727 @item set arm fpu @var{fputype}
25728 This command sets the ARM floating-point unit (FPU) type. The
25729 argument @var{fputype} can be one of these:
25733 Determine the FPU type by querying the OS ABI.
25735 Software FPU, with mixed-endian doubles on little-endian ARM
25738 GCC-compiled FPA co-processor.
25740 Software FPU with pure-endian doubles.
25746 Show the current type of the FPU.
25749 This command forces @value{GDBN} to use the specified ABI.
25752 Show the currently used ABI.
25754 @item set arm fallback-mode (arm|thumb|auto)
25755 @value{GDBN} uses the symbol table, when available, to determine
25756 whether instructions are ARM or Thumb. This command controls
25757 @value{GDBN}'s default behavior when the symbol table is not
25758 available. The default is @samp{auto}, which causes @value{GDBN} to
25759 use the current execution mode (from the @code{T} bit in the @code{CPSR}
25762 @item show arm fallback-mode
25763 Show the current fallback instruction mode.
25765 @item set arm force-mode (arm|thumb|auto)
25766 This command overrides use of the symbol table to determine whether
25767 instructions are ARM or Thumb. The default is @samp{auto}, which
25768 causes @value{GDBN} to use the symbol table and then the setting
25769 of @samp{set arm fallback-mode}.
25771 @item show arm force-mode
25772 Show the current forced instruction mode.
25774 @item set arm unwind-secure-frames
25775 This command enables unwinding from Non-secure to Secure mode on
25776 Cortex-M with Security extension.
25777 This can trigger security exceptions when unwinding the exception
25779 It is enabled by default.
25781 @item show arm unwind-secure-frames
25782 Show whether unwinding from Non-secure to Secure mode is enabled.
25784 @item set debug arm
25785 Toggle whether to display ARM-specific debugging messages from the ARM
25786 target support subsystem.
25788 @item show debug arm
25789 Show whether ARM-specific debugging messages are enabled.
25793 @item target sim @r{[}@var{simargs}@r{]} @dots{}
25794 The @value{GDBN} ARM simulator accepts the following optional arguments.
25797 @item --swi-support=@var{type}
25798 Tell the simulator which SWI interfaces to support. The argument
25799 @var{type} may be a comma separated list of the following values.
25800 The default value is @code{all}.
25816 @item target sim @r{[}@var{simargs}@r{]} @dots{}
25817 The @value{GDBN} BPF simulator accepts the following optional arguments.
25820 @item --skb-data-offset=@var{offset}
25821 Tell the simulator the offset, measured in bytes, of the
25822 @code{skb_data} field in the kernel @code{struct sk_buff} structure.
25823 This offset is used by some BPF specific-purpose load/store
25824 instructions. Defaults to 0.
25831 The Motorola m68k configuration includes ColdFire support.
25834 @subsection MicroBlaze
25835 @cindex Xilinx MicroBlaze
25836 @cindex XMD, Xilinx Microprocessor Debugger
25838 The MicroBlaze is a soft-core processor supported on various Xilinx
25839 FPGAs, such as Spartan or Virtex series. Boards with these processors
25840 usually have JTAG ports which connect to a host system running the Xilinx
25841 Embedded Development Kit (EDK) or Software Development Kit (SDK).
25842 This host system is used to download the configuration bitstream to
25843 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
25844 communicates with the target board using the JTAG interface and
25845 presents a @code{gdbserver} interface to the board. By default
25846 @code{xmd} uses port @code{1234}. (While it is possible to change
25847 this default port, it requires the use of undocumented @code{xmd}
25848 commands. Contact Xilinx support if you need to do this.)
25850 Use these GDB commands to connect to the MicroBlaze target processor.
25853 @item target remote :1234
25854 Use this command to connect to the target if you are running @value{GDBN}
25855 on the same system as @code{xmd}.
25857 @item target remote @var{xmd-host}:1234
25858 Use this command to connect to the target if it is connected to @code{xmd}
25859 running on a different system named @var{xmd-host}.
25862 Use this command to download a program to the MicroBlaze target.
25864 @item set debug microblaze @var{n}
25865 Enable MicroBlaze-specific debugging messages if non-zero.
25867 @item show debug microblaze @var{n}
25868 Show MicroBlaze-specific debugging level.
25871 @node MIPS Embedded
25872 @subsection @acronym{MIPS} Embedded
25875 @value{GDBN} supports these special commands for @acronym{MIPS} targets:
25878 @item set mipsfpu double
25879 @itemx set mipsfpu single
25880 @itemx set mipsfpu none
25881 @itemx set mipsfpu auto
25882 @itemx show mipsfpu
25883 @kindex set mipsfpu
25884 @kindex show mipsfpu
25885 @cindex @acronym{MIPS} remote floating point
25886 @cindex floating point, @acronym{MIPS} remote
25887 If your target board does not support the @acronym{MIPS} floating point
25888 coprocessor, you should use the command @samp{set mipsfpu none} (if you
25889 need this, you may wish to put the command in your @value{GDBN} init
25890 file). This tells @value{GDBN} how to find the return value of
25891 functions which return floating point values. It also allows
25892 @value{GDBN} to avoid saving the floating point registers when calling
25893 functions on the board. If you are using a floating point coprocessor
25894 with only single precision floating point support, as on the @sc{r4650}
25895 processor, use the command @samp{set mipsfpu single}. The default
25896 double precision floating point coprocessor may be selected using
25897 @samp{set mipsfpu double}.
25899 In previous versions the only choices were double precision or no
25900 floating point, so @samp{set mipsfpu on} will select double precision
25901 and @samp{set mipsfpu off} will select no floating point.
25903 As usual, you can inquire about the @code{mipsfpu} variable with
25904 @samp{show mipsfpu}.
25907 @node OpenRISC 1000
25908 @subsection OpenRISC 1000
25909 @cindex OpenRISC 1000
25912 The OpenRISC 1000 provides a free RISC instruction set architecture. It is
25913 mainly provided as a soft-core which can run on Xilinx, Altera and other
25916 @value{GDBN} for OpenRISC supports the below commands when connecting to
25924 Runs the builtin CPU simulator which can run very basic
25925 programs but does not support most hardware functions like MMU.
25926 For more complex use cases the user is advised to run an external
25927 target, and connect using @samp{target remote}.
25929 Example: @code{target sim}
25931 @item set debug or1k
25932 Toggle whether to display OpenRISC-specific debugging messages from the
25933 OpenRISC target support subsystem.
25935 @item show debug or1k
25936 Show whether OpenRISC-specific debugging messages are enabled.
25939 @node PowerPC Embedded
25940 @subsection PowerPC Embedded
25942 @cindex DVC register
25943 @value{GDBN} supports using the DVC (Data Value Compare) register to
25944 implement in hardware simple hardware watchpoint conditions of the form:
25947 (@value{GDBP}) watch @var{address|variable} \
25948 if @var{address|variable} == @var{constant expression}
25951 The DVC register will be automatically used when @value{GDBN} detects
25952 such pattern in a condition expression, and the created watchpoint uses one
25953 debug register (either the @code{exact-watchpoints} option is on and the
25954 variable is scalar, or the variable has a length of one byte). This feature
25955 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
25958 When running on PowerPC embedded processors, @value{GDBN} automatically uses
25959 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
25960 in which case watchpoints using only one debug register are created when
25961 watching variables of scalar types.
25963 You can create an artificial array to watch an arbitrary memory
25964 region using one of the following commands (@pxref{Expressions}):
25967 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
25968 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
25971 PowerPC embedded processors support masked watchpoints. See the discussion
25972 about the @code{mask} argument in @ref{Set Watchpoints}.
25974 @cindex ranged breakpoint
25975 PowerPC embedded processors support hardware accelerated
25976 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
25977 the inferior whenever it executes an instruction at any address within
25978 the range it was set at. To set a ranged breakpoint in @value{GDBN},
25979 use the @code{break-range} command.
25981 @value{GDBN} provides the following PowerPC-specific commands:
25984 @kindex break-range
25985 @item break-range @var{start-locspec}, @var{end-locspec}
25986 Set a breakpoint for an address range given by @var{start-locspec} and
25987 @var{end-locspec}, which are location specs. @xref{Location
25988 Specifications}, for a list of all the possible forms of location
25989 specs. @value{GDBN} resolves both @var{start-locspec} and
25990 @var{end-locspec}, and uses the addresses of the resolved code
25991 locations as start and end addresses of the range to break at. The
25992 breakpoint will stop execution of the inferior whenever it executes an
25993 instruction at any address between the start and end addresses,
25994 inclusive. If either @var{start-locspec} or @var{end-locspec} resolve
25995 to multiple code locations in the program, then the command aborts
25996 with an error without creating a breakpoint.
25998 @kindex set powerpc
25999 @item set powerpc soft-float
26000 @itemx show powerpc soft-float
26001 Force @value{GDBN} to use (or not use) a software floating point calling
26002 convention. By default, @value{GDBN} selects the calling convention based
26003 on the selected architecture and the provided executable file.
26005 @item set powerpc vector-abi
26006 @itemx show powerpc vector-abi
26007 Force @value{GDBN} to use the specified calling convention for vector
26008 arguments and return values. The valid options are @samp{auto};
26009 @samp{generic}, to avoid vector registers even if they are present;
26010 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
26011 registers. By default, @value{GDBN} selects the calling convention
26012 based on the selected architecture and the provided executable file.
26014 @item set powerpc exact-watchpoints
26015 @itemx show powerpc exact-watchpoints
26016 Allow @value{GDBN} to use only one debug register when watching a variable
26017 of scalar type, thus assuming that the variable is accessed through the
26018 address of its first byte.
26023 @subsection Atmel AVR
26026 When configured for debugging the Atmel AVR, @value{GDBN} supports the
26027 following AVR-specific commands:
26030 @item info io_registers
26031 @kindex info io_registers@r{, AVR}
26032 @cindex I/O registers (Atmel AVR)
26033 This command displays information about the AVR I/O registers. For
26034 each register, @value{GDBN} prints its number and value.
26041 When configured for debugging CRIS, @value{GDBN} provides the
26042 following CRIS-specific commands:
26045 @item set cris-version @var{ver}
26046 @cindex CRIS version
26047 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
26048 The CRIS version affects register names and sizes. This command is useful in
26049 case autodetection of the CRIS version fails.
26051 @item show cris-version
26052 Show the current CRIS version.
26054 @item set cris-dwarf2-cfi
26055 @cindex DWARF-2 CFI and CRIS
26056 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
26057 Change to @samp{off} when using @code{gcc-cris} whose version is below
26060 @item show cris-dwarf2-cfi
26061 Show the current state of using DWARF-2 CFI.
26063 @item set cris-mode @var{mode}
26065 Set the current CRIS mode to @var{mode}. It should only be changed when
26066 debugging in guru mode, in which case it should be set to
26067 @samp{guru} (the default is @samp{normal}).
26069 @item show cris-mode
26070 Show the current CRIS mode.
26074 @subsection Renesas Super-H
26077 For the Renesas Super-H processor, @value{GDBN} provides these
26081 @item set sh calling-convention @var{convention}
26082 @kindex set sh calling-convention
26083 Set the calling-convention used when calling functions from @value{GDBN}.
26084 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
26085 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
26086 convention. If the DWARF-2 information of the called function specifies
26087 that the function follows the Renesas calling convention, the function
26088 is called using the Renesas calling convention. If the calling convention
26089 is set to @samp{renesas}, the Renesas calling convention is always used,
26090 regardless of the DWARF-2 information. This can be used to override the
26091 default of @samp{gcc} if debug information is missing, or the compiler
26092 does not emit the DWARF-2 calling convention entry for a function.
26094 @item show sh calling-convention
26095 @kindex show sh calling-convention
26096 Show the current calling convention setting.
26101 @node Architectures
26102 @section Architectures
26104 This section describes characteristics of architectures that affect
26105 all uses of @value{GDBN} with the architecture, both native and cross.
26112 * HPPA:: HP PA architecture
26117 * AMD GPU:: @acronym{AMD GPU} architectures
26121 @subsection AArch64
26122 @cindex AArch64 support
26124 When @value{GDBN} is debugging the AArch64 architecture, it provides the
26125 following special commands:
26128 @item set debug aarch64
26129 @kindex set debug aarch64
26130 This command determines whether AArch64 architecture-specific debugging
26131 messages are to be displayed.
26133 @item show debug aarch64
26134 Show whether AArch64 debugging messages are displayed.
26138 @subsubsection AArch64 SVE.
26139 @cindex AArch64 SVE.
26141 When @value{GDBN} is debugging the AArch64 architecture, if the Scalable Vector
26142 Extension (SVE) is present, then @value{GDBN} will provide the vector registers
26143 @code{$z0} through @code{$z31}, vector predicate registers @code{$p0} through
26144 @code{$p15}, and the @code{$ffr} register. In addition, the pseudo register
26145 @code{$vg} will be provided. This is the vector granule for the current thread
26146 and represents the number of 64-bit chunks in an SVE @code{z} register.
26148 If the vector length changes, then the @code{$vg} register will be updated,
26149 but the lengths of the @code{z} and @code{p} registers will not change. This
26150 is a known limitation of @value{GDBN} and does not affect the execution of the
26153 For SVE, the following definitions are used throughout @value{GDBN}'s source
26154 code and in this document:
26159 @var{vl}: The vector length, in bytes. It defines the size of each @code{Z}
26165 @var{vq}: The number of 128 bit units in @var{vl}. This is mostly used
26166 internally by @value{GDBN} and the Linux Kernel.
26171 @var{vg}: The number of 64 bit units in @var{vl}. This is mostly used
26172 internally by @value{GDBN} and the Linux Kernel.
26178 @subsubsection AArch64 SME.
26179 @anchor{AArch64 SME}
26181 @cindex AArch64 SME
26182 @cindex Scalable Matrix Extension
26184 The Scalable Matrix Extension (@url{https://community.arm.com/arm-community-blogs/b/architectures-and-processors-blog/posts/scalable-matrix-extension-armv9-a-architecture, @acronym{SME}})
26185 is an AArch64 architecture extension that expands on the concept of the
26186 Scalable Vector Extension (@url{https://developer.arm.com/documentation/101726/4-0/Learn-about-the-Scalable-Vector-Extension--SVE-/What-is-the-Scalable-Vector-Extension-, @acronym{SVE}})
26187 by providing a 2-dimensional register @code{ZA}, which is a square
26188 matrix of variable size, just like SVE provides a group of vector registers of
26191 Similarly to SVE, where the size of each @code{Z} register is directly related
26192 to the vector length (@var{vl} for short), the @acronym{SME} @code{ZA} matrix
26193 register's size is directly related to the streaming vector length
26194 (@var{svl} for short). @xref{vl}. @xref{svl}.
26196 The @code{ZA} register state can be either active or inactive, if it is not in
26199 @acronym{SME} also introduces a new execution mode called streaming
26200 @acronym{SVE} mode (streaming mode for short). When streaming mode is
26201 enabled, the program supports execution of @acronym{SVE2} instructions and the
26202 @acronym{SVE} registers will have vector length @var{svl}. When streaming
26203 mode is disabled, the SVE registers have vector length @var{vl}.
26205 For more information about @acronym{SME} and @acronym{SVE}, please refer to
26206 official @url{https://developer.arm.com/documentation/ddi0487/latest,
26207 architecture documentation}.
26209 The following definitions are used throughout @value{GDBN}'s source code and
26215 @var{svl}: The streaming vector length, in bytes. It defines the size of each
26216 dimension of the 2-dimensional square @code{ZA} matrix. The total size of
26217 @code{ZA} is therefore @var{svl} by @var{svl}.
26219 When streaming mode is enabled, it defines the size of the @acronym{SVE}
26225 @var{svq}: The number of 128 bit units in @var{svl}, also known as streaming
26226 vector granule. This is mostly used internally by @value{GDBN} and the Linux
26232 @var{svg}: The number of 64 bit units in @var{svl}. This is mostly used
26233 internally by @value{GDBN} and the Linux Kernel.
26239 When @value{GDBN} is debugging the AArch64 architecture, if the Scalable Matrix
26240 Extension (@acronym{SME}) is present, then @value{GDBN} will make the @code{ZA}
26241 register available. @value{GDBN} will also make the @code{SVG} register and
26242 @code{SVCR} pseudo-register available.
26244 The @code{ZA} register is a 2-dimensional square @var{svl} by @var{svl}
26245 matrix of bytes. To simplify the representation and access to the @code{ZA}
26246 register in @value{GDBN}, it is defined as a vector of
26247 @var{svl}x@var{svl} bytes.
26249 If the user wants to index the @code{ZA} register as a matrix, it is possible
26250 to reference @code{ZA} as @code{ZA[@var{i}][@var{j}]}, where @var{i} is the
26251 row number and @var{j} is the column number.
26253 The @code{SVG} register always contains the streaming vector granule
26254 (@var{svg}) for the current thread. From the value of register @code{SVG} we
26255 can easily derive the @var{svl} value.
26257 @anchor{aarch64 sme svcr}
26258 The @code{SVCR} pseudo-register (streaming vector control register) is a status
26259 register that holds two state bits: @sc{sm} in bit 0 and @sc{za} in bit 1.
26261 If the @sc{sm} bit is 1, it means the current thread is in streaming
26262 mode, and the @acronym{SVE} registers will use @var{svl} for their sizes. If
26263 the @sc{sm} bit is 0, the current thread is not in streaming mode, and the
26264 @acronym{SVE} registers will use @var{vl} for their sizes. @xref{vl}.
26266 If the @sc{za} bit is 1, it means the @code{ZA} register is being used and
26267 has meaningful contents. If the @sc{za} bit is 0, the @code{ZA} register is
26268 unavailable and its contents are undefined.
26270 For convenience and simplicity, if the @sc{za} bit is 0, the @code{ZA}
26271 register and all of its pseudo-registers will read as zero.
26273 If @var{svl} changes during the execution of a program, then the @code{ZA}
26274 register size and the bits in the @code{SVCR} pseudo-register will be updated
26277 It is possible for users to change @var{svl} during the execution of a
26278 program by modifying the @code{SVG} register value.
26280 Whenever the @code{SVG} register is modified with a new value, the
26281 following will be observed:
26285 @item The @sc{za} and @sc{sm} bits will be cleared in the @code{SVCR}
26288 @item The @code{ZA} register will have a new size and its state will be
26289 cleared, forcing its contents and the contents of all of its pseudo-registers
26292 @item If the @sc{sm} bit was 1, the @acronym{SVE} registers will be reset to
26293 having their sizes based on @var{vl} as opposed to @var{svl}. If the
26294 @sc{sm} bit was 0 prior to modifying the @code{SVG} register, there will be no
26295 observable effect on the @acronym{SVE} registers.
26299 The possible values for the @code{SVG} register are 2, 4, 8, 16, 32. These
26300 numbers correspond to streaming vector length (@var{svl}) values of 16
26301 bytes, 32 bytes, 64 bytes, 128 bytes and 256 bytes respectively.
26303 The minimum size of the @code{ZA} register is 16 x 16 (256) bytes, and the
26304 maximum size is 256 x 256 (65536) bytes. In streaming mode, with bit @sc{sm}
26305 set, the size of the @code{ZA} register is the size of all the SVE @code{Z}
26306 registers combined.
26308 The @code{ZA} register can also be accessed using tiles and tile slices.
26310 Tile pseudo-registers are square, 2-dimensional sub-arrays of elements within
26311 the @code{ZA} register.
26313 The tile pseudo-registers have the following naming pattern:
26314 @code{ZA<@var{tile number}><@var{qualifier}>}.
26316 There is a total of 31 @code{ZA} tile pseudo-registers. They are
26317 @code{ZA0B}, @code{ZA0H} through @code{ZA1H}, @code{ZA0S} through @code{ZA3S},
26318 @code{ZA0D} through @code{ZA7D} and @code{ZA0Q} through @code{ZA15Q}.
26320 Tile slice pseudo-registers are vectors of horizontally or vertically
26321 contiguous elements within the @code{ZA} register.
26323 The tile slice pseudo-registers have the following naming pattern:
26324 @code{ZA<@var{tile number}><@var{direction}><@var{qualifier}>
26325 <@var{slice number}>}.
26327 There are up to 16 tiles (0 ~ 15), the direction can be either @code{v}
26328 (vertical) or @code{h} (horizontal), the qualifiers can be @code{b} (byte),
26329 @code{h} (halfword), @code{s} (word), @code{d} (doubleword) and @code{q}
26330 (quadword) and there are up to 256 slices (0 ~ 255) depending on the value
26331 of @var{svl}. The number of slices is the same as the value of @var{svl}.
26333 The number of available tile slice pseudo-registers can be large. For a
26334 minimum @var{svl} of 16 bytes, there are 5 (number of qualifiers) x
26335 2 (number of directions) x 16 (@var{svl}) pseudo-registers. For the
26336 maximum @var{svl} of 256 bytes, there are 5 x 2 x 256 pseudo-registers.
26338 When listing all the available registers, users will see the
26339 currently-available @code{ZA} pseudo-registers. Pseudo-registers that don't
26340 exist for a given @var{svl} value will not be displayed.
26342 For more information on @acronym{SME} and its terminology, please refer to the
26343 @url{https://developer.arm.com/documentation/ddi0616/aa/,
26344 Arm Architecture Reference Manual Supplement}, The Scalable Matrix Extension
26345 (@acronym{SME}), for Armv9-A.
26347 Some features are still under development and rely on
26348 @url{https://github.com/ARM-software/acle/releases/latest, ACLE} and
26349 @url{https://github.com/ARM-software/abi-aa/blob/main/aapcs64/aapcs64.rst, ABI}
26350 definitions, so there are known limitations to the current @acronym{SME}
26351 support in @value{GDBN}.
26353 One such example is calling functions in the program being debugged by
26354 @value{GDBN}. Such calls are not @acronym{SME}-aware and thus don't take into
26355 account the @code{SVCR} pseudo-register bits nor the @code{ZA} register
26356 contents. @xref{Calling}.
26358 The @url{https://github.com/ARM-software/abi-aa/blob/main/aapcs64/aapcs64.rst#the-za-lazy-saving-scheme,
26359 lazy saving scheme} involving the @code{TPIDR2} register is not yet supported
26360 by @value{GDBN}, though the @code{TPIDR2} register is known and supported
26363 Lastly, an important limitation for @command{gdbserver} is its inability to
26364 communicate @var{svl} changes to @value{GDBN}. This means @command{gdbserver},
26365 even though it is capable of adjusting its internal caches to reflect a change
26366 in the value of @var{svl} mid-execution, will operate with a potentially
26367 different @var{svl} value compared to @value{GDBN}. This can lead to
26368 @value{GDBN} showing incorrect values for the @code{ZA} register and
26369 incorrect values for SVE registers (when in streaming mode).
26371 This is the same limitation we have for the @acronym{SVE} registers, and there
26372 are plans to address this limitation going forward.
26374 @subsubsection AArch64 SME2.
26375 @anchor{AArch64 SME2}
26377 @cindex AArch64 SME2
26378 @cindex Scalable Matrix Extension 2
26380 The Scalable Matrix Extension 2 is an AArch64 architecture extension that
26381 further expands the @acronym{SME} extension with the following:
26385 @item The ability to address the @code{ZA} array through groups of
26386 one-dimensional @code{ZA} array vectors, as opposed to @code{ZA} tiles
26389 @item Instructions to operate on groups of @acronym{SVE} @code{Z} registers and
26390 @code{ZA} array vectors.
26392 @item A new 512 bit @code{ZT0} lookup table register, for data decompression.
26396 When @value{GDBN} is debugging the AArch64 architecture, if the Scalable Matrix
26397 Extension 2 (@acronym{SME2}) is present, then @value{GDBN} will make the
26398 @code{ZT0} register available.
26400 The @code{ZT0} register is only considered active when the @code{ZA} register
26401 state is active, therefore when the @sc{za} bit of the @code{SVCR} is 1.
26403 When the @sc{za} bit of @code{SVCR} is 0, that means the @code{ZA} register
26404 state is not active, which means the @code{ZT0} register state is also not
26407 When @code{ZT0} is not active, it is comprised of zeroes, just like @code{ZA}.
26409 Similarly to the @code{ZA} register, if the @code{ZT0} state is not active and
26410 the user attempts to modify its value such that any of its bytes is non-zero,
26411 then @value{GDBN} will initialize the @code{ZA} register state as well, which
26412 means the @code{SVCR} @sc{za} bit gets set to 1.
26414 For more information about @acronym{SME2}, please refer to the
26415 official @url{https://developer.arm.com/documentation/ddi0487/latest,
26416 architecture documentation}.
26418 @subsubsection AArch64 Pointer Authentication.
26419 @cindex AArch64 Pointer Authentication.
26420 @anchor{AArch64 PAC}
26422 When @value{GDBN} is debugging the AArch64 architecture, and the program is
26423 using the v8.3-A feature Pointer Authentication (PAC), then whenever the link
26424 register @code{$lr} is pointing to an PAC function its value will be masked.
26425 When GDB prints a backtrace, any addresses that required unmasking will be
26426 postfixed with the marker [PAC]. When using the MI, this is printed as part
26427 of the @code{addr_flags} field.
26429 @subsubsection AArch64 Memory Tagging Extension.
26430 @cindex AArch64 Memory Tagging Extension.
26432 When @value{GDBN} is debugging the AArch64 architecture, the program is
26433 using the v8.5-A feature Memory Tagging Extension (MTE) and there is support
26434 in the kernel for MTE, @value{GDBN} will make memory tagging functionality
26435 available for inspection and editing of logical and allocation tags.
26436 @xref{Memory Tagging}.
26438 To aid debugging, @value{GDBN} will output additional information when SIGSEGV
26439 signals are generated as a result of memory tag failures.
26441 If the tag violation is synchronous, the following will be shown:
26444 Program received signal SIGSEGV, Segmentation fault
26445 Memory tag violation while accessing address 0x0500fffff7ff8000
26450 If the tag violation is asynchronous, the fault address is not available.
26451 In this case @value{GDBN} will show the following:
26454 Program received signal SIGSEGV, Segmentation fault
26455 Memory tag violation
26456 Fault address unavailable.
26459 A special register, @code{tag_ctl}, is made available through the
26460 @code{org.gnu.gdb.aarch64.mte} feature. This register exposes some
26461 options that can be controlled at runtime and emulates the @code{prctl}
26462 option @code{PR_SET_TAGGED_ADDR_CTRL}. For further information, see the
26463 documentation in the Linux kernel.
26465 @value{GDBN} supports dumping memory tag data to core files through the
26466 @command{gcore} command and reading memory tag data from core files generated
26467 by the @command{gcore} command or the Linux kernel.
26469 When a process uses memory-mapped pages protected by memory tags (for
26470 example, AArch64 MTE), this additional information will be recorded in
26471 the core file in the event of a crash or if @value{GDBN} generates a core file
26472 from the current process state.
26474 The memory tag data will be used so developers can display the memory
26475 tags from a particular memory region (using the @samp{m} modifier to the
26476 @command{x} command, using the @command{print} command or using the various
26477 @command{memory-tag} subcommands.
26479 In the case of a crash, @value{GDBN} will attempt to retrieve the memory tag
26480 information automatically from the core file, and will show one of the above
26481 messages depending on whether the synchronous or asynchronous mode is selected.
26482 @xref{Memory Tagging}. @xref{Memory}.
26488 @item set struct-convention @var{mode}
26489 @kindex set struct-convention
26490 @cindex struct return convention
26491 @cindex struct/union returned in registers
26492 Set the convention used by the inferior to return @code{struct}s and
26493 @code{union}s from functions to @var{mode}. Possible values of
26494 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
26495 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
26496 are returned on the stack, while @code{"reg"} means that a
26497 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
26498 be returned in a register.
26500 @item show struct-convention
26501 @kindex show struct-convention
26502 Show the current setting of the convention to return @code{struct}s
26507 @subsubsection Intel @dfn{Memory Protection Extensions} (MPX).
26508 @cindex Intel Memory Protection Extensions (MPX).
26510 Memory Protection Extension (MPX) adds the bound registers @samp{BND0}
26511 @footnote{The register named with capital letters represent the architecture
26512 registers.} through @samp{BND3}. Bound registers store a pair of 64-bit values
26513 which are the lower bound and upper bound. Bounds are effective addresses or
26514 memory locations. The upper bounds are architecturally represented in 1's
26515 complement form. A bound having lower bound = 0, and upper bound = 0
26516 (1's complement of all bits set) will allow access to the entire address space.
26518 @samp{BND0} through @samp{BND3} are represented in @value{GDBN} as @samp{bnd0raw}
26519 through @samp{bnd3raw}. Pseudo registers @samp{bnd0} through @samp{bnd3}
26520 display the upper bound performing the complement of one operation on the
26521 upper bound value, i.e.@ when upper bound in @samp{bnd0raw} is 0 in the
26522 @value{GDBN} @samp{bnd0} it will be @code{0xfff@dots{}}. In this sense it
26523 can also be noted that the upper bounds are inclusive.
26525 As an example, assume that the register BND0 holds bounds for a pointer having
26526 access allowed for the range between 0x32 and 0x71. The values present on
26527 bnd0raw and bnd registers are presented as follows:
26530 bnd0raw = @{0x32, 0xffffffff8e@}
26531 bnd0 = @{lbound = 0x32, ubound = 0x71@} : size 64
26534 This way the raw value can be accessed via bnd0raw@dots{}bnd3raw. Any
26535 change on bnd0@dots{}bnd3 or bnd0raw@dots{}bnd3raw is reflect on its
26536 counterpart. When the bnd0@dots{}bnd3 registers are displayed via
26537 Python, the display includes the memory size, in bits, accessible to
26540 Bounds can also be stored in bounds tables, which are stored in
26541 application memory. These tables store bounds for pointers by specifying
26542 the bounds pointer's value along with its bounds. Evaluating and changing
26543 bounds located in bound tables is therefore interesting while investigating
26544 bugs on MPX context. @value{GDBN} provides commands for this purpose:
26547 @item show mpx bound @var{pointer}
26548 @kindex show mpx bound
26549 Display bounds of the given @var{pointer}.
26551 @item set mpx bound @var{pointer}, @var{lbound}, @var{ubound}
26552 @kindex set mpx bound
26553 Set the bounds of a pointer in the bound table.
26554 This command takes three parameters: @var{pointer} is the pointers
26555 whose bounds are to be changed, @var{lbound} and @var{ubound} are new values
26556 for lower and upper bounds respectively.
26559 When you call an inferior function on an Intel MPX enabled program,
26560 GDB sets the inferior's bound registers to the init (disabled) state
26561 before calling the function. As a consequence, bounds checks for the
26562 pointer arguments passed to the function will always pass.
26564 This is necessary because when you call an inferior function, the
26565 program is usually in the middle of the execution of other function.
26566 Since at that point bound registers are in an arbitrary state, not
26567 clearing them would lead to random bound violations in the called
26570 You can still examine the influence of the bound registers on the
26571 execution of the called function by stopping the execution of the
26572 called function at its prologue, setting bound registers, and
26573 continuing the execution. For example:
26577 Breakpoint 2 at 0x4009de: file i386-mpx-call.c, line 47.
26578 $ print upper (a, b, c, d, 1)
26579 Breakpoint 2, upper (a=0x0, b=0x6e0000005b, c=0x0, d=0x0, len=48)....
26581 @{lbound = 0x0, ubound = ffffffff@} : size -1
26584 At this last step the value of bnd0 can be changed for investigation of bound
26585 violations caused along the execution of the call. In order to know how to
26586 set the bound registers or bound table for the call consult the ABI.
26588 @subsubsection x87 registers
26590 @value{GDBN} provides access to the x87 state through the following registers:
26594 @item @code{$st0} to @code{st7}: @code{ST(0)} to @code{ST(7)} floating-point
26596 @item @code{$fctrl}: control word register (@code{FCW})
26597 @item @code{$fstat}: status word register (@code{FSW})
26598 @item @code{$ftag}: tag word (@code{FTW})
26599 @item @code{$fiseg}: last instruction pointer segment
26600 @item @code{$fioff}: last instruction pointer
26601 @item @code{$foseg}: last data pointer segment
26602 @item @code{$fooff}: last data pointer
26603 @item @code{$fop}: last opcode
26610 See the following section.
26613 @subsection @acronym{MIPS}
26615 @cindex stack on Alpha
26616 @cindex stack on @acronym{MIPS}
26617 @cindex Alpha stack
26618 @cindex @acronym{MIPS} stack
26619 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
26620 sometimes requires @value{GDBN} to search backward in the object code to
26621 find the beginning of a function.
26623 @cindex response time, @acronym{MIPS} debugging
26624 To improve response time (especially for embedded applications, where
26625 @value{GDBN} may be restricted to a slow serial line for this search)
26626 you may want to limit the size of this search, using one of these
26630 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
26631 @item set heuristic-fence-post @var{limit}
26632 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
26633 search for the beginning of a function. A value of @var{0} (the
26634 default) means there is no limit. However, except for @var{0}, the
26635 larger the limit the more bytes @code{heuristic-fence-post} must search
26636 and therefore the longer it takes to run. You should only need to use
26637 this command when debugging a stripped executable.
26639 @item show heuristic-fence-post
26640 Display the current limit.
26644 These commands are available @emph{only} when @value{GDBN} is configured
26645 for debugging programs on Alpha or @acronym{MIPS} processors.
26647 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
26651 @item set mips abi @var{arg}
26652 @kindex set mips abi
26653 @cindex set ABI for @acronym{MIPS}
26654 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
26655 values of @var{arg} are:
26659 The default ABI associated with the current binary (this is the
26669 @item show mips abi
26670 @kindex show mips abi
26671 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
26673 @item set mips compression @var{arg}
26674 @kindex set mips compression
26675 @cindex code compression, @acronym{MIPS}
26676 Tell @value{GDBN} which @acronym{MIPS} compressed
26677 @acronym{ISA, Instruction Set Architecture} encoding is used by the
26678 inferior. @value{GDBN} uses this for code disassembly and other
26679 internal interpretation purposes. This setting is only referred to
26680 when no executable has been associated with the debugging session or
26681 the executable does not provide information about the encoding it uses.
26682 Otherwise this setting is automatically updated from information
26683 provided by the executable.
26685 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
26686 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
26687 executables containing @acronym{MIPS16} code frequently are not
26688 identified as such.
26690 This setting is ``sticky''; that is, it retains its value across
26691 debugging sessions until reset either explicitly with this command or
26692 implicitly from an executable.
26694 The compiler and/or assembler typically add symbol table annotations to
26695 identify functions compiled for the @acronym{MIPS16} or
26696 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
26697 are present, @value{GDBN} uses them in preference to the global
26698 compressed @acronym{ISA} encoding setting.
26700 @item show mips compression
26701 @kindex show mips compression
26702 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
26703 @value{GDBN} to debug the inferior.
26706 @itemx show mipsfpu
26707 @xref{MIPS Embedded, set mipsfpu}.
26709 @item set mips mask-address @var{arg}
26710 @kindex set mips mask-address
26711 @cindex @acronym{MIPS} addresses, masking
26712 This command determines whether the most-significant 32 bits of 64-bit
26713 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
26714 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
26715 setting, which lets @value{GDBN} determine the correct value.
26717 @item show mips mask-address
26718 @kindex show mips mask-address
26719 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
26722 @item set remote-mips64-transfers-32bit-regs
26723 @kindex set remote-mips64-transfers-32bit-regs
26724 This command controls compatibility with 64-bit @acronym{MIPS} targets that
26725 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
26726 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
26727 and 64 bits for other registers, set this option to @samp{on}.
26729 @item show remote-mips64-transfers-32bit-regs
26730 @kindex show remote-mips64-transfers-32bit-regs
26731 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
26733 @item set debug mips
26734 @kindex set debug mips
26735 This command turns on and off debugging messages for the @acronym{MIPS}-specific
26736 target code in @value{GDBN}.
26738 @item show debug mips
26739 @kindex show debug mips
26740 Show the current setting of @acronym{MIPS} debugging messages.
26746 @cindex HPPA support
26748 When @value{GDBN} is debugging the HP PA architecture, it provides the
26749 following special commands:
26752 @item set debug hppa
26753 @kindex set debug hppa
26754 This command determines whether HPPA architecture-specific debugging
26755 messages are to be displayed.
26757 @item show debug hppa
26758 Show whether HPPA debugging messages are displayed.
26760 @item maint print unwind @var{address}
26761 @kindex maint print unwind@r{, HPPA}
26762 This command displays the contents of the unwind table entry at the
26763 given @var{address}.
26769 @subsection PowerPC
26770 @cindex PowerPC architecture
26772 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
26773 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
26774 numbers stored in the floating point registers. These values must be stored
26775 in two consecutive registers, always starting at an even register like
26776 @code{f0} or @code{f2}.
26778 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
26779 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
26780 @code{f2} and @code{f3} for @code{$dl1} and so on.
26782 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
26783 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
26786 @subsection Nios II
26787 @cindex Nios II architecture
26789 When @value{GDBN} is debugging the Nios II architecture,
26790 it provides the following special commands:
26794 @item set debug nios2
26795 @kindex set debug nios2
26796 This command turns on and off debugging messages for the Nios II
26797 target code in @value{GDBN}.
26799 @item show debug nios2
26800 @kindex show debug nios2
26801 Show the current setting of Nios II debugging messages.
26805 @subsection Sparc64
26806 @cindex Sparc64 support
26807 @cindex Application Data Integrity
26808 @subsubsection ADI Support
26810 The M7 processor supports an Application Data Integrity (ADI) feature that
26811 detects invalid data accesses. When software allocates memory and enables
26812 ADI on the allocated memory, it chooses a 4-bit version number, sets the
26813 version in the upper 4 bits of the 64-bit pointer to that data, and stores
26814 the 4-bit version in every cacheline of that data. Hardware saves the latter
26815 in spare bits in the cache and memory hierarchy. On each load and store,
26816 the processor compares the upper 4 VA (virtual address) bits to the
26817 cacheline's version. If there is a mismatch, the processor generates a
26818 version mismatch trap which can be either precise or disrupting. The trap
26819 is an error condition which the kernel delivers to the process as a SIGSEGV
26822 Note that only 64-bit applications can use ADI and need to be built with
26825 Values of the ADI version tags, which are in granularity of a
26826 cacheline (64 bytes), can be viewed or modified.
26830 @kindex adi examine
26831 @item adi (examine | x) [ / @var{n} ] @var{addr}
26833 The @code{adi examine} command displays the value of one ADI version tag per
26836 @var{n} is a decimal integer specifying the number in bytes; the default
26837 is 1. It specifies how much ADI version information, at the ratio of 1:ADI
26838 block size, to display.
26840 @var{addr} is the address in user address space where you want @value{GDBN}
26841 to begin displaying the ADI version tags.
26843 Below is an example of displaying ADI versions of variable "shmaddr".
26846 (@value{GDBP}) adi x/100 shmaddr
26847 0xfff800010002c000: 0 0
26851 @item adi (assign | a) [ / @var{n} ] @var{addr} = @var{tag}
26853 The @code{adi assign} command is used to assign new ADI version tag
26856 @var{n} is a decimal integer specifying the number in bytes;
26857 the default is 1. It specifies how much ADI version information, at the
26858 ratio of 1:ADI block size, to modify.
26860 @var{addr} is the address in user address space where you want @value{GDBN}
26861 to begin modifying the ADI version tags.
26863 @var{tag} is the new ADI version tag.
26865 For example, do the following to modify then verify ADI versions of
26866 variable "shmaddr":
26869 (@value{GDBP}) adi a/100 shmaddr = 7
26870 (@value{GDBP}) adi x/100 shmaddr
26871 0xfff800010002c000: 7 7
26878 @cindex S12Z support
26880 When @value{GDBN} is debugging the S12Z architecture,
26881 it provides the following special command:
26884 @item maint info bdccsr
26885 @kindex maint info bdccsr@r{, S12Z}
26886 This command displays the current value of the microprocessor's
26891 @subsection @acronym{AMD GPU}
26892 @cindex @acronym{AMD GPU} support
26894 @value{GDBN} supports debugging programs offloaded to @acronym{AMD GPU} devices
26895 using the @url{https://docs.amd.com/, @acronym{AMD ROCm}} platform.
26896 @value{GDBN} presents host threads alongside GPU wavefronts, allowing debugging
26897 both the host and device parts of the program simultaneously.
26899 @subsubsection @acronym{AMD GPU} Architectures
26901 The list of @acronym{AMD GPU} architectures supported by @value{GDBN} depends
26902 on the version of the AMD Debugger API library used. See its
26903 @uref{https://docs.amd.com/bundle/ROCDebugger_User_and_API, documentation} for
26906 @subsubsection @acronym{AMD GPU} Device Driver and @acronym{AMD ROCm} Runtime
26908 @value{GDBN} requires a compatible @acronym{AMD GPU} device driver to
26909 be installed. A warning message is displayed if either the device
26910 driver version or the version of the debug support it implements is
26911 unsupported. @value{GDBN} will continue to function except no
26912 @acronym{AMD GPU} debugging will be possible.
26914 @value{GDBN} requires each agent to have compatible firmware installed
26915 by the device driver. A warning message is displayed if unsupported
26916 firmware is detected. @value{GDBN} will continue to function except
26917 no @acronym{AMD GPU} debugging will be possible on the agent.
26919 @value{GDBN} requires a compatible @acronym{AMD ROCm} runtime to be
26920 loaded in order to detect @acronym{AMD GPU} code objects and
26921 wavefronts. A warning message is displayed if an unsupported
26922 @acronym{AMD ROCm} runtime is detected, or there is an error or
26923 restriction that prevents debugging. @value{GDBN} will continue to
26924 function except no @acronym{AMD GPU} debugging will be possible.
26926 @subsubsection @acronym{AMD GPU} Wavefronts
26929 An @acronym{AMD GPU} wavefront is represented in @value{GDBN} as a
26932 Note that some @acronym{AMD GPU} architectures may have restrictions
26933 on providing information about @acronym{AMD GPU} wavefronts created
26934 when @value{GDBN} is not attached (@pxref{AMD GPU Attaching
26935 Restrictions, , @acronym{AMD GPU} Attaching Restrictions}).
26937 When scheduler-locking is in effect (@pxref{set scheduler-locking}),
26938 new wavefronts created by the resumed thread (either CPU thread or GPU
26939 wavefront) are held in the halt state.
26941 @subsubsection @acronym{AMD GPU} Code Objects
26943 The @samp{info sharedlibrary} command will show the @acronym{AMD GPU}
26944 code objects as file or memory URIs, together with the host's shared
26945 libraries. For example:
26948 (@value{GDBP}) info sharedlibrary
26949 From To Syms Read Shared Object Library
26950 0x1111 0x2222 Yes (*) /lib64/ld-linux-x86-64.so.2
26952 0x3333 0x4444 Yes (*) /opt/rocm-4.5.0/.../libamd_comgr.so
26953 0x5555 0x6666 Yes (*) /lib/x86_64-linux-gnu/libtinfo.so.5
26954 0x7777 0x8888 Yes file:///tmp/a.out#offset=6477&size=10832
26955 0x9999 0xaaaa Yes (*) memory://95557/mem#offset=0x1234&size=100
26956 (*): Shared library is missing debugging information.
26960 For a @samp{file} URI, the path portion is the file on disk containing
26961 the code object. The @var{offset} parameter is a 0-based offset in
26962 this file, to the start of the code object. If omitted, it defaults to
26963 0. The @var{size} parameter is the size of the code object in bytes.
26964 If omitted, it defaults to the size of the file.
26966 For a @samp{memory} URI, the path portion is the process id of the
26967 process owning the memory containing the code object. The @var{offset}
26968 parameter is the memory address where the code object is found, and
26969 the @var{size} parameter is its size in bytes.
26971 @acronym{AMD GPU} code objects are loaded into each @acronym{AMD GPU}
26972 device separately. The @samp{info sharedlibrary} command may
26973 therefore show the same code object loaded multiple times. As a
26974 consequence, setting a breakpoint in @acronym{AMD GPU} code will
26975 result in multiple breakpoint locations if there are multiple
26976 @acronym{AMD GPU} devices.
26978 @subsubsection @acronym{AMD GPU} Entity Target Identifiers and Convenience Variables
26980 The @acronym{AMD GPU} entities have the following target identifier formats:
26984 @item Thread Target ID
26985 The @acronym{AMD GPU} thread target identifier (@var{systag}) string has the
26989 AMDGPU Wave @var{agent-id}:@var{queue-id}:@var{dispatch-id}:@var{wave-id} (@var{work-group-x},@var{work-group-y},@var{work-group-z})/@var{work-group-thread-index}
26994 @anchor{AMD GPU Signals}
26995 @subsubsection @acronym{AMD GPU} Signals
26997 For @acronym{AMD GPU} wavefronts, @value{GDBN} maps target conditions to stop
26998 signals in the following way:
27003 Execution of an illegal instruction.
27006 Execution of a @code{S_TRAP} instruction other than:
27011 @code{S_TRAP 1} which is used by @value{GDBN} to insert breakpoints.
27014 @code{S_TRAP 2} which raises @code{SIGABRT}.
27019 Execution of a @code{S_TRAP 2} instruction.
27022 Execution of a floating point or integer instruction detects a
27023 condition that is enabled to raise a signal. The conditions include:
27028 Floating point operation is invalid.
27031 Floating point operation had subnormal input that was rounded to zero.
27034 Floating point operation performed a division by zero.
27037 Floating point operation produced an overflow result. The result was
27038 rounded to infinity.
27041 Floating point operation produced an underflow result. A subnormal
27042 result was rounded to zero.
27045 Floating point operation produced an inexact result.
27048 Integer operation performed a division by zero.
27052 By default, these conditions are not enabled to raise signals. The
27053 @samp{set $mode} command can be used to change the @acronym{AMD GPU}
27054 wavefront's register that has bits controlling which conditions are
27055 enabled to raise signals. The @samp{print $trapsts} command can be
27056 used to inspect which conditions have been detected even if they are
27057 not enabled to raise a signal.
27060 Execution of an instruction that accessed global memory using an
27061 address that is outside the virtual address range.
27064 Execution of an instruction that accessed a global memory page that is
27065 either not mapped or accessed with incompatible permissions.
27069 If a single instruction raises more than one signal, they will be
27070 reported one at a time each time the wavefront is continued.
27072 @subsubsection @acronym{AMD GPU} Memory Violation Reporting
27074 A wavefront can report memory violation events. However, the program
27075 location at which they are reported may be after the machine instruction
27076 that caused them. This can result in the reported source statement
27077 being incorrect. The following commands can be used to control this
27082 @kindex set amdgpu precise-memory
27083 @cindex AMD GPU precise memory event reporting
27084 @item set amdgpu precise-memory @var{mode}
27085 Controls how @acronym{AMD GPU} devices detect memory violations, where
27091 The program location may not be immediately after the instruction that
27092 caused the memory violation. This is the default.
27095 Requests that the program location will be immediately after the
27096 instruction that caused a memory violation. Enabling this mode may make
27097 the @acronym{AMD GPU} device execution significantly slower as it has to
27098 wait for each memory operation to complete before executing the next
27103 The @code{amdgpu precise-memory} parameter is per-inferior. When an
27104 inferior forks or execs, or the user uses the @code{clone-inferior} command,
27105 and an inferior is created as a result, the newly created inferior inherits
27106 the parameter value of the original inferior.
27108 @kindex show amdgpu precise-memory
27109 @cindex AMD GPU precise memory event reporting
27110 @item show amdgpu precise-memory
27111 Displays the currently requested AMD GPU precise memory setting.
27115 @subsubsection @acronym{AMD GPU} Logging
27117 The @samp{set debug amd-dbgapi} command can be used
27118 to enable diagnostic messages in the @samp{amd-dbgapi} target. The
27119 @samp{show debug amd-dbgapi} command displays the current setting.
27120 @xref{set debug amd-dbgapi}.
27122 The @samp{set debug amd-dbgapi-lib log-level @var{level}} command can be used
27123 to enable diagnostic messages from the @samp{amd-dbgapi} library (which
27124 @value{GDBN} uses under the hood). The @samp{show debug amd-dbgapi-lib
27125 log-level} command displays the current @samp{amd-dbgapi} library log level.
27126 @xref{set debug amd-dbgapi-lib}.
27128 @subsubsection @acronym{AMD GPU} Restrictions
27133 When in non-stop mode, wavefronts may not hit breakpoints inserted
27134 while not stopped, nor see memory updates made while not stopped,
27135 until the wavefront is next stopped. Memory updated by non-stopped
27136 wavefronts may not be visible until the wavefront is next stopped.
27138 @item The HIP runtime performs deferred code object loading by default.
27139 @acronym{AMD GPU} code objects are not loaded until the first kernel is
27140 launched. Before then, all breakpoints have to be set as pending breakpoints.
27142 If source line positions are used that only correspond to source lines in
27143 unloaded code objects, then @value{GDBN} may not set pending breakpoints, and
27144 instead set breakpoints on the next following source line that maps to host
27145 code. This can result in unexpected breakpoint hits being reported. When the
27146 code object containing the source lines is loaded, the incorrect breakpoints
27147 will be removed and replaced by the correct ones. This problem can be avoided
27148 by only setting breakpoints in unloaded code objects using symbol or function
27151 Setting the @code{HIP_ENABLE_DEFERRED_LOADING} environment variable to @code{0}
27152 can be used to disable deferred code object loading by the HIP runtime. This
27153 ensures all code objects will be loaded when the inferior reaches the beginning
27154 of the @code{main} function.
27157 If no CPU thread is running, then @samp{Ctrl-C} is not able to stop
27158 @acronym{AMD GPU} threads. This can happen for example if you enable
27159 @code{scheduler-locking} after the whole program stopped, and then resume an
27160 @acronym{AMD GPU} thread. The only way to unblock the situation is to kill the
27161 @value{GDBN} process.
27163 @anchor{AMD GPU Attaching Restrictions}
27166 By default, for some architectures, the @acronym{AMD GPU} device driver causes
27167 all @acronym{AMD GPU} wavefronts created when @value{GDBN} is not attached to
27168 be unable to report the dispatch associated with the wavefront, or the
27169 wavefront's work-group position. The @samp{info threads} command will display
27170 this missing information with a @samp{?}.
27172 This does not affect wavefronts created while @value{GDBN} is attached which
27173 are always capable of reporting this information.
27175 If the @env{HSA_ENABLE_DEBUG} environment variable is set to @samp{1} when the
27176 @acronym{AMD ROCm} runtime is initialized, then this information will be
27177 available for all architectures even for wavefronts created when @value{GDBN}
27182 @node Controlling GDB
27183 @chapter Controlling @value{GDBN}
27185 You can alter the way @value{GDBN} interacts with you by using the
27186 @code{set} command. For commands controlling how @value{GDBN} displays
27187 data, see @ref{Print Settings, ,Print Settings}. Other settings are
27192 * Editing:: Command editing
27193 * Command History:: Command history
27194 * Screen Size:: Screen size
27195 * Output Styling:: Output styling
27196 * Numbers:: Numbers
27197 * ABI:: Configuring the current ABI
27198 * Auto-loading:: Automatically loading associated files
27199 * Messages/Warnings:: Optional warnings and messages
27200 * Debugging Output:: Optional messages about internal happenings
27201 * Other Misc Settings:: Other Miscellaneous Settings
27209 @value{GDBN} indicates its readiness to read a command by printing a string
27210 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
27211 can change the prompt string with the @code{set prompt} command. For
27212 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
27213 the prompt in one of the @value{GDBN} sessions so that you can always tell
27214 which one you are talking to.
27216 @emph{Note:} @code{set prompt} does not add a space for you after the
27217 prompt you set. This allows you to set a prompt which ends in a space
27218 or a prompt that does not.
27222 @item set prompt @var{newprompt}
27223 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
27225 @kindex show prompt
27227 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
27230 Versions of @value{GDBN} that ship with Python scripting enabled have
27231 prompt extensions. The commands for interacting with these extensions
27235 @kindex set extended-prompt
27236 @item set extended-prompt @var{prompt}
27237 Set an extended prompt that allows for substitutions.
27238 @xref{gdb.prompt}, for a list of escape sequences that can be used for
27239 substitution. Any escape sequences specified as part of the prompt
27240 string are replaced with the corresponding strings each time the prompt
27246 set extended-prompt Current working directory: \w (@value{GDBP})
27249 Note that when an extended-prompt is set, it takes control of the
27250 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
27252 @kindex show extended-prompt
27253 @item show extended-prompt
27254 Prints the extended prompt. Any escape sequences specified as part of
27255 the prompt string with @code{set extended-prompt}, are replaced with the
27256 corresponding strings each time the prompt is displayed.
27260 @section Command Editing
27262 @cindex command line editing
27264 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
27265 @sc{gnu} library provides consistent behavior for programs which provide a
27266 command line interface to the user. Advantages are @sc{gnu} Emacs-style
27267 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
27268 substitution, and a storage and recall of command history across
27269 debugging sessions.
27271 You may control the behavior of command line editing in @value{GDBN} with the
27272 command @code{set}.
27275 @kindex set editing
27278 @itemx set editing on
27279 Enable command line editing (enabled by default).
27281 @item set editing off
27282 Disable command line editing.
27284 @kindex show editing
27286 Show whether command line editing is enabled.
27289 @ifset SYSTEM_READLINE
27290 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
27292 @ifclear SYSTEM_READLINE
27293 @xref{Command Line Editing},
27295 for more details about the Readline
27296 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
27297 encouraged to read that chapter.
27299 @cindex Readline application name
27300 @value{GDBN} sets the Readline application name to @samp{gdb}. This
27301 is useful for conditions in @file{.inputrc}.
27303 @cindex operate-and-get-next
27304 @value{GDBN} defines a bindable Readline command,
27305 @code{operate-and-get-next}. This is bound to @kbd{C-o} by default.
27306 This command accepts the current line for execution and fetches the
27307 next line relative to the current line from the history for editing.
27308 Any argument is ignored.
27310 @node Command History
27311 @section Command History
27312 @cindex command history
27314 @value{GDBN} can keep track of the commands you type during your
27315 debugging sessions, so that you can be certain of precisely what
27316 happened. Use these commands to manage the @value{GDBN} command
27319 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
27320 package, to provide the history facility.
27321 @ifset SYSTEM_READLINE
27322 @xref{Using History Interactively, , , history, GNU History Library},
27324 @ifclear SYSTEM_READLINE
27325 @xref{Using History Interactively},
27327 for the detailed description of the History library.
27329 To issue a command to @value{GDBN} without affecting certain aspects of
27330 the state which is seen by users, prefix it with @samp{server }
27331 (@pxref{Server Prefix}). This
27332 means that this command will not affect the command history, nor will it
27333 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
27334 pressed on a line by itself.
27336 @cindex @code{server}, command prefix
27337 The server prefix does not affect the recording of values into the value
27338 history; to print a value without recording it into the value history,
27339 use the @code{output} command instead of the @code{print} command.
27341 Here is the description of @value{GDBN} commands related to command
27345 @cindex history substitution
27346 @cindex history file
27347 @kindex set history filename
27348 @cindex @env{GDBHISTFILE}, environment variable
27349 @item set history filename @r{[}@var{fname}@r{]}
27350 Set the name of the @value{GDBN} command history file to @var{fname}.
27351 This is the file where @value{GDBN} reads an initial command history
27352 list, and where it writes the command history from this session when it
27353 exits. You can access this list through history expansion or through
27354 the history command editing characters listed below. This file defaults
27355 to the value of the environment variable @env{GDBHISTFILE}, or to
27356 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
27359 The @env{GDBHISTFILE} environment variable is read after processing
27360 any @value{GDBN} initialization files (@pxref{Startup}) and after
27361 processing any commands passed using command line options (for
27362 example, @code{-ex}).
27364 If the @var{fname} argument is not given, or if the @env{GDBHISTFILE}
27365 is the empty string then @value{GDBN} will neither try to load an
27366 existing history file, nor will it try to save the history on exit.
27368 @cindex save command history
27369 @kindex set history save
27370 @item set history save
27371 @itemx set history save on
27372 Record command history in a file, whose name may be specified with the
27373 @code{set history filename} command. By default, this option is
27374 disabled. The command history will be recorded when @value{GDBN}
27375 exits. If @code{set history filename} is set to the empty string then
27376 history saving is disabled, even when @code{set history save} is
27379 @item set history save off
27380 Don't record the command history into the file specified by @code{set
27381 history filename} when @value{GDBN} exits.
27383 @cindex history size
27384 @kindex set history size
27385 @cindex @env{GDBHISTSIZE}, environment variable
27386 @item set history size @var{size}
27387 @itemx set history size unlimited
27388 Set the number of commands which @value{GDBN} keeps in its history list.
27389 This defaults to the value of the environment variable @env{GDBHISTSIZE}, or
27390 to 256 if this variable is not set. Non-numeric values of @env{GDBHISTSIZE}
27391 are ignored. If @var{size} is @code{unlimited} or if @env{GDBHISTSIZE} is
27392 either a negative number or the empty string, then the number of commands
27393 @value{GDBN} keeps in the history list is unlimited.
27395 The @env{GDBHISTSIZE} environment variable is read after processing
27396 any @value{GDBN} initialization files (@pxref{Startup}) and after
27397 processing any commands passed using command line options (for
27398 example, @code{-ex}).
27400 @cindex remove duplicate history
27401 @kindex set history remove-duplicates
27402 @item set history remove-duplicates @var{count}
27403 @itemx set history remove-duplicates unlimited
27404 Control the removal of duplicate history entries in the command history list.
27405 If @var{count} is non-zero, @value{GDBN} will look back at the last @var{count}
27406 history entries and remove the first entry that is a duplicate of the current
27407 entry being added to the command history list. If @var{count} is
27408 @code{unlimited} then this lookbehind is unbounded. If @var{count} is 0, then
27409 removal of duplicate history entries is disabled.
27411 Only history entries added during the current session are considered for
27412 removal. This option is set to 0 by default.
27416 History expansion assigns special meaning to the character @kbd{!}.
27417 @ifset SYSTEM_READLINE
27418 @xref{Event Designators, , , history, GNU History Library},
27420 @ifclear SYSTEM_READLINE
27421 @xref{Event Designators},
27425 @cindex history expansion, turn on/off
27426 Since @kbd{!} is also the logical not operator in C, history expansion
27427 is off by default. If you decide to enable history expansion with the
27428 @code{set history expansion on} command, you may sometimes need to
27429 follow @kbd{!} (when it is used as logical not, in an expression) with
27430 a space or a tab to prevent it from being expanded. The readline
27431 history facilities do not attempt substitution on the strings
27432 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
27434 The commands to control history expansion are:
27437 @item set history expansion on
27438 @itemx set history expansion
27439 @kindex set history expansion
27440 Enable history expansion. History expansion is off by default.
27442 @item set history expansion off
27443 Disable history expansion.
27446 @kindex show history
27448 @itemx show history filename
27449 @itemx show history save
27450 @itemx show history size
27451 @itemx show history expansion
27452 These commands display the state of the @value{GDBN} history parameters.
27453 @code{show history} by itself displays all four states.
27458 @kindex show commands
27459 @cindex show last commands
27460 @cindex display command history
27461 @item show commands
27462 Display the last ten commands in the command history.
27464 @item show commands @var{n}
27465 Print ten commands centered on command number @var{n}.
27467 @item show commands +
27468 Print ten commands just after the commands last printed.
27472 @section Screen Size
27473 @cindex size of screen
27474 @cindex screen size
27477 @cindex pauses in output
27479 Certain commands to @value{GDBN} may produce large amounts of
27480 information output to the screen. To help you read all of it,
27481 @value{GDBN} pauses and asks you for input at the end of each page of
27482 output. Type @key{RET} when you want to see one more page of output,
27483 @kbd{q} to discard the remaining output, or @kbd{c} to continue
27484 without paging for the rest of the current command. Also, the screen
27485 width setting determines when to wrap lines of output. Depending on
27486 what is being printed, @value{GDBN} tries to break the line at a
27487 readable place, rather than simply letting it overflow onto the
27490 Normally @value{GDBN} knows the size of the screen from the terminal
27491 driver software. For example, on Unix @value{GDBN} uses the termcap data base
27492 together with the value of the @env{TERM} environment variable and the
27493 @code{stty rows} and @code{stty cols} settings. If this is not correct,
27494 you can override it with the @code{set height} and @code{set
27501 @kindex show height
27502 @item set height @var{lpp}
27503 @itemx set height unlimited
27505 @itemx set width @var{cpl}
27506 @itemx set width unlimited
27508 These @code{set} commands specify a screen height of @var{lpp} lines and
27509 a screen width of @var{cpl} characters. The associated @code{show}
27510 commands display the current settings.
27512 If you specify a height of either @code{unlimited} or zero lines,
27513 @value{GDBN} does not pause during output no matter how long the
27514 output is. This is useful if output is to a file or to an editor
27517 Likewise, you can specify @samp{set width unlimited} or @samp{set
27518 width 0} to prevent @value{GDBN} from wrapping its output.
27520 @item set pagination on
27521 @itemx set pagination off
27522 @kindex set pagination
27523 Turn the output pagination on or off; the default is on. Turning
27524 pagination off is the alternative to @code{set height unlimited}. Note that
27525 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
27526 Options, -batch}) also automatically disables pagination.
27528 @item show pagination
27529 @kindex show pagination
27530 Show the current pagination mode.
27533 @node Output Styling
27534 @section Output Styling
27540 @value{GDBN} can style its output on a capable terminal. This is
27541 enabled by default on most systems, but disabled by default when in
27542 batch mode (@pxref{Mode Options}). Various style settings are available;
27543 and styles can also be disabled entirely.
27546 @item set style enabled @samp{on|off}
27547 Enable or disable all styling. The default is host-dependent, with
27548 most hosts defaulting to @samp{on}.
27550 If the @env{NO_COLOR} environment variable is set to a non-empty
27551 value, then @value{GDBN} will change this to @samp{off} at startup.
27553 @item show style enabled
27554 Show the current state of styling.
27556 @item set style sources @samp{on|off}
27557 Enable or disable source code styling. This affects whether source
27558 code, such as the output of the @code{list} command, is styled. The
27559 default is @samp{on}. Note that source styling only works if styling
27560 in general is enabled, and if a source highlighting library is
27561 available to @value{GDBN}.
27563 There are two ways that highlighting can be done. First, if
27564 @value{GDBN} was linked with the GNU Source Highlight library, then it
27565 is used. Otherwise, if @value{GDBN} was configured with Python
27566 scripting support, and if the Python Pygments package is available,
27567 then it will be used.
27569 @item show style sources
27570 Show the current state of source code styling.
27572 @item set style tui-current-position @samp{on|off}
27573 Enable or disable styling of the source and assembly code highlighted
27574 by the TUI's current position indicator. The default is @samp{off}.
27575 @xref{TUI, ,@value{GDBN} Text User Interface}.
27577 @item show style tui-current-position
27578 Show whether the source and assembly code highlighted by the TUI's
27579 current position indicator is styled.
27581 @anchor{style_disassembler_enabled}
27582 @item set style disassembler enabled @samp{on|off}
27583 Enable or disable disassembler styling. This affects whether
27584 disassembler output, such as the output of the @code{disassemble}
27585 command, is styled. Disassembler styling only works if styling in
27586 general is enabled (with @code{set style enabled on}), and if a source
27587 highlighting library is available to @value{GDBN}.
27589 The two source highlighting libraries that @value{GDBN} could use to
27590 style disassembler output are; @value{GDBN}'s builtin disassembler, or
27591 the Python Pygments package.
27593 @value{GDBN}'s first choice will be to use the builtin disassembler
27594 for styling, this usually provides better results, being able to style
27595 different types of instruction operands differently. However, the
27596 builtin disassembler is not able to style all architectures.
27598 For architectures that the builtin disassembler is unable to style,
27599 @value{GDBN} will fall back to use the Python Pygments package where
27600 possible. In order to use the Python Pygments package, @value{GDBN}
27601 must be built with Python support, and the Pygments package must be
27604 If neither of these options are available then @value{GDBN} will
27605 produce unstyled disassembler output, even when this setting is
27608 To discover if the current architecture supports styling using the
27609 builtin disassembler library see @ref{maint_libopcodes_styling,,@kbd{maint
27610 show libopcodes-styling enabled}}.
27612 @item show style disassembler enabled
27613 Show the current state of disassembler styling.
27617 Subcommands of @code{set style} control specific forms of styling.
27618 These subcommands all follow the same pattern: each style-able object
27619 can be styled with a foreground color, a background color, and an
27622 For example, the style of file names can be controlled using the
27623 @code{set style filename} group of commands:
27626 @item set style filename background @var{color}
27627 Set the background to @var{color}. Valid colors are @samp{none}
27628 (meaning the terminal's default color), @samp{black}, @samp{red},
27629 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
27632 @item set style filename foreground @var{color}
27633 Set the foreground to @var{color}. Valid colors are @samp{none}
27634 (meaning the terminal's default color), @samp{black}, @samp{red},
27635 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
27638 @item set style filename intensity @var{value}
27639 Set the intensity to @var{value}. Valid intensities are @samp{normal}
27640 (the default), @samp{bold}, and @samp{dim}.
27643 The @code{show style} command and its subcommands are styling
27644 a style name in their output using its own style.
27645 So, use @command{show style} to see the complete list of styles,
27646 their characteristics and the visual aspect of each style.
27648 The style-able objects are:
27651 Control the styling of file names and URLs. By default, this style's
27652 foreground color is green.
27655 Control the styling of function names. These are managed with the
27656 @code{set style function} family of commands. By default, this
27657 style's foreground color is yellow.
27659 This style is also used for symbol names in styled disassembler output
27660 if @value{GDBN} is using its builtin disassembler library for styling
27661 (@pxref{style_disassembler_enabled,,@kbd{set style disassembler
27665 Control the styling of variable names. These are managed with the
27666 @code{set style variable} family of commands. By default, this style's
27667 foreground color is cyan.
27670 Control the styling of addresses. These are managed with the
27671 @code{set style address} family of commands. By default, this style's
27672 foreground color is blue.
27674 This style is also used for addresses in styled disassembler output
27675 if @value{GDBN} is using its builtin disassembler library for styling
27676 (@pxref{style_disassembler_enabled,,@kbd{set style disassembler
27680 Control the styling of @value{GDBN}'s version number text. By
27681 default, this style's foreground color is magenta and it has bold
27682 intensity. The version number is displayed in two places, the output
27683 of @command{show version}, and when @value{GDBN} starts up.
27685 In order to control how @value{GDBN} styles the version number at
27686 startup, add the @code{set style version} family of commands to the
27687 early initialization command file (@pxref{Initialization
27691 Control the styling of titles. These are managed with the
27692 @code{set style title} family of commands. By default, this style's
27693 intensity is bold. Commands are using the title style to improve
27694 the readability of large output. For example, the commands
27695 @command{apropos} and @command{help} are using the title style
27696 for the command names.
27699 Control the styling of highlightings. These are managed with the
27700 @code{set style highlight} family of commands. By default, this style's
27701 foreground color is red. Commands are using the highlight style to draw
27702 the user attention to some specific parts of their output. For example,
27703 the command @command{apropos -v REGEXP} uses the highlight style to
27704 mark the documentation parts matching @var{regexp}.
27707 Control the styling of data annotations added by @value{GDBN} to data
27708 it displays. By default, this style's intensity is dim. Metadata
27709 annotations include the @samp{repeats @var{n} times} annotation for
27710 suppressed display of repeated array elements (@pxref{Print Settings}),
27711 @samp{<unavailable>} and @w{@samp{<error @var{descr}>}} annotations
27712 for errors and @samp{<optimized-out>} annotations for optimized-out
27713 values in displaying stack frame information in backtraces
27714 (@pxref{Backtrace}), etc.
27717 Control the styling of the TUI border. Note that, unlike other
27718 styling options, only the color of the border can be controlled via
27719 @code{set style}. This was done for compatibility reasons, as TUI
27720 controls to set the border's intensity predated the addition of
27721 general styling to @value{GDBN}. @xref{TUI Configuration}.
27723 @item tui-active-border
27724 Control the styling of the active TUI border; that is, the TUI window
27725 that has the focus.
27727 @item disassembler comment
27728 Control the styling of comments in the disassembler output. These are
27729 managed with the @code{set style disassembler comment} family of
27730 commands. This style is only used when @value{GDBN} is styling using
27731 its builtin disassembler library
27732 (@pxref{style_disassembler_enabled,,@kbd{set style disassembler
27733 enabled}}). By default, this style's intensity is dim, and its
27734 foreground color is white.
27736 @item disassembler immediate
27737 Control the styling of numeric operands in the disassembler output.
27738 These are managed with the @code{set style disassembler immediate}
27739 family of commands. This style is not used for instruction operands
27740 that represent addresses, in that case the @samp{disassembler address}
27741 style is used. This style is only used when @value{GDBN} is styling
27742 using its builtin disassembler library. By default, this style's
27743 foreground color is blue.
27745 @item disassembler address
27746 Control the styling of address operands in the disassembler output.
27747 This is an alias for the @samp{address} style.
27749 @item disassembler symbol
27750 Control the styling of symbol names in the disassembler output. This
27751 is an alias for the @samp{function} style.
27753 @item disassembler mnemonic
27754 Control the styling of instruction mnemonics in the disassembler
27755 output. These are managed with the @code{set style disassembler
27756 mnemonic} family of commands. This style is also used for assembler
27757 directives, e.g.@: @code{.byte}, @code{.word}, etc. This style is
27758 only used when @value{GDBN} is styling using its builtin disassembler
27759 library. By default, this style's foreground color is green.
27761 @item disassembler register
27762 Control the styling of register operands in the disassembler output.
27763 These are managed with the @code{set style disassembler register}
27764 family of commands. This style is only used when @value{GDBN} is
27765 styling using its builtin disassembler library. By default, this style's
27766 foreground color is red.
27772 @cindex number representation
27773 @cindex entering numbers
27775 You can always enter numbers in octal, decimal, or hexadecimal in
27776 @value{GDBN} by the usual conventions: octal numbers begin with
27777 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
27778 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
27779 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
27780 10; likewise, the default display for numbers---when no particular
27781 format is specified---is base 10. You can change the default base for
27782 both input and output with the commands described below.
27785 @kindex set input-radix
27786 @item set input-radix @var{base}
27787 Set the default base for numeric input. Supported choices
27788 for @var{base} are decimal 8, 10, or 16. The base must itself be
27789 specified either unambiguously or using the current input radix; for
27793 set input-radix 012
27794 set input-radix 10.
27795 set input-radix 0xa
27799 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
27800 leaves the input radix unchanged, no matter what it was, since
27801 @samp{10}, being without any leading or trailing signs of its base, is
27802 interpreted in the current radix. Thus, if the current radix is 16,
27803 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
27806 @kindex set output-radix
27807 @item set output-radix @var{base}
27808 Set the default base for numeric display. Supported choices
27809 for @var{base} are decimal 8, 10, or 16. The base must itself be
27810 specified either unambiguously or using the current input radix.
27812 @kindex show input-radix
27813 @item show input-radix
27814 Display the current default base for numeric input.
27816 @kindex show output-radix
27817 @item show output-radix
27818 Display the current default base for numeric display.
27820 @item set radix @r{[}@var{base}@r{]}
27824 These commands set and show the default base for both input and output
27825 of numbers. @code{set radix} sets the radix of input and output to
27826 the same base; without an argument, it resets the radix back to its
27827 default value of 10.
27832 @section Configuring the Current ABI
27834 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
27835 application automatically. However, sometimes you need to override its
27836 conclusions. Use these commands to manage @value{GDBN}'s view of the
27842 @cindex Newlib OS ABI and its influence on the longjmp handling
27844 One @value{GDBN} configuration can debug binaries for multiple operating
27845 system targets, either via remote debugging or native emulation.
27846 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
27847 but you can override its conclusion using the @code{set osabi} command.
27848 One example where this is useful is in debugging of binaries which use
27849 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
27850 not have the same identifying marks that the standard C library for your
27853 When @value{GDBN} is debugging the AArch64 architecture, it provides a
27854 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
27855 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
27856 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
27860 Show the OS ABI currently in use.
27863 With no argument, show the list of registered available OS ABI's.
27865 @item set osabi @var{abi}
27866 Set the current OS ABI to @var{abi}.
27869 @cindex float promotion
27871 Generally, the way that an argument of type @code{float} is passed to a
27872 function depends on whether the function is prototyped. For a prototyped
27873 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
27874 according to the architecture's convention for @code{float}. For unprototyped
27875 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
27876 @code{double} and then passed.
27878 Unfortunately, some forms of debug information do not reliably indicate whether
27879 a function is prototyped. If @value{GDBN} calls a function that is not marked
27880 as prototyped, it consults @kbd{set coerce-float-to-double}.
27883 @kindex set coerce-float-to-double
27884 @item set coerce-float-to-double
27885 @itemx set coerce-float-to-double on
27886 Arguments of type @code{float} will be promoted to @code{double} when passed
27887 to an unprototyped function. This is the default setting.
27889 @item set coerce-float-to-double off
27890 Arguments of type @code{float} will be passed directly to unprototyped
27893 @kindex show coerce-float-to-double
27894 @item show coerce-float-to-double
27895 Show the current setting of promoting @code{float} to @code{double}.
27899 @kindex show cp-abi
27900 @value{GDBN} needs to know the ABI used for your program's C@t{++}
27901 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
27902 used to build your application. @value{GDBN} only fully supports
27903 programs with a single C@t{++} ABI; if your program contains code using
27904 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
27905 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
27906 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
27907 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
27908 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
27909 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
27914 Show the C@t{++} ABI currently in use.
27917 With no argument, show the list of supported C@t{++} ABI's.
27919 @item set cp-abi @var{abi}
27920 @itemx set cp-abi auto
27921 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
27925 @section Automatically loading associated files
27926 @cindex auto-loading
27928 @value{GDBN} sometimes reads files with commands and settings automatically,
27929 without being explicitly told so by the user. We call this feature
27930 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
27931 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
27932 results or introduce security risks (e.g., if the file comes from untrusted
27935 There are various kinds of files @value{GDBN} can automatically load.
27936 In addition to these files, @value{GDBN} supports auto-loading code written
27937 in various extension languages. @xref{Auto-loading extensions}.
27939 Note that loading of these associated files (including the local @file{.gdbinit}
27940 file) requires accordingly configured @code{auto-load safe-path}
27941 (@pxref{Auto-loading safe path}).
27943 For these reasons, @value{GDBN} includes commands and options to let you
27944 control when to auto-load files and which files should be auto-loaded.
27947 @anchor{set auto-load off}
27948 @kindex set auto-load off
27949 @item set auto-load off
27950 Globally disable loading of all auto-loaded files.
27951 You may want to use this command with the @samp{-iex} option
27952 (@pxref{Option -init-eval-command}) such as:
27954 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
27957 Be aware that system init file (@pxref{System-wide configuration})
27958 and init files from your home directory (@pxref{Home Directory Init File})
27959 still get read (as they come from generally trusted directories).
27960 To prevent @value{GDBN} from auto-loading even those init files, use the
27961 @option{-nx} option (@pxref{Mode Options}), in addition to
27962 @code{set auto-load no}.
27964 @anchor{show auto-load}
27965 @kindex show auto-load
27966 @item show auto-load
27967 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
27971 (@value{GDBP}) show auto-load
27972 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
27973 libthread-db: Auto-loading of inferior specific libthread_db is on.
27974 local-gdbinit: Auto-loading of .gdbinit script from current directory
27976 python-scripts: Auto-loading of Python scripts is on.
27977 safe-path: List of directories from which it is safe to auto-load files
27978 is $debugdir:$datadir/auto-load.
27979 scripts-directory: List of directories from which to load auto-loaded scripts
27980 is $debugdir:$datadir/auto-load.
27983 @anchor{info auto-load}
27984 @kindex info auto-load
27985 @item info auto-load
27986 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
27990 (@value{GDBP}) info auto-load
27993 Yes /home/user/gdb/gdb-gdb.gdb
27994 libthread-db: No auto-loaded libthread-db.
27995 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
27999 Yes /home/user/gdb/gdb-gdb.py
28003 These are @value{GDBN} control commands for the auto-loading:
28005 @multitable @columnfractions .5 .5
28006 @item @xref{set auto-load off}.
28007 @tab Disable auto-loading globally.
28008 @item @xref{show auto-load}.
28009 @tab Show setting of all kinds of files.
28010 @item @xref{info auto-load}.
28011 @tab Show state of all kinds of files.
28012 @item @xref{set auto-load gdb-scripts}.
28013 @tab Control for @value{GDBN} command scripts.
28014 @item @xref{show auto-load gdb-scripts}.
28015 @tab Show setting of @value{GDBN} command scripts.
28016 @item @xref{info auto-load gdb-scripts}.
28017 @tab Show state of @value{GDBN} command scripts.
28018 @item @xref{set auto-load python-scripts}.
28019 @tab Control for @value{GDBN} Python scripts.
28020 @item @xref{show auto-load python-scripts}.
28021 @tab Show setting of @value{GDBN} Python scripts.
28022 @item @xref{info auto-load python-scripts}.
28023 @tab Show state of @value{GDBN} Python scripts.
28024 @item @xref{set auto-load guile-scripts}.
28025 @tab Control for @value{GDBN} Guile scripts.
28026 @item @xref{show auto-load guile-scripts}.
28027 @tab Show setting of @value{GDBN} Guile scripts.
28028 @item @xref{info auto-load guile-scripts}.
28029 @tab Show state of @value{GDBN} Guile scripts.
28030 @item @xref{set auto-load scripts-directory}.
28031 @tab Control for @value{GDBN} auto-loaded scripts location.
28032 @item @xref{show auto-load scripts-directory}.
28033 @tab Show @value{GDBN} auto-loaded scripts location.
28034 @item @xref{add-auto-load-scripts-directory}.
28035 @tab Add directory for auto-loaded scripts location list.
28036 @item @xref{set auto-load local-gdbinit}.
28037 @tab Control for init file in the current directory.
28038 @item @xref{show auto-load local-gdbinit}.
28039 @tab Show setting of init file in the current directory.
28040 @item @xref{info auto-load local-gdbinit}.
28041 @tab Show state of init file in the current directory.
28042 @item @xref{set auto-load libthread-db}.
28043 @tab Control for thread debugging library.
28044 @item @xref{show auto-load libthread-db}.
28045 @tab Show setting of thread debugging library.
28046 @item @xref{info auto-load libthread-db}.
28047 @tab Show state of thread debugging library.
28048 @item @xref{set auto-load safe-path}.
28049 @tab Control directories trusted for automatic loading.
28050 @item @xref{show auto-load safe-path}.
28051 @tab Show directories trusted for automatic loading.
28052 @item @xref{add-auto-load-safe-path}.
28053 @tab Add directory trusted for automatic loading.
28057 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
28058 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
28060 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
28061 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
28064 @node Init File in the Current Directory
28065 @subsection Automatically loading init file in the current directory
28066 @cindex auto-loading init file in the current directory
28068 By default, @value{GDBN} reads and executes the canned sequences of commands
28069 from init file (if any) in the current working directory,
28070 see @ref{Init File in the Current Directory during Startup}.
28072 Note that loading of this local @file{.gdbinit} file also requires accordingly
28073 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
28076 @anchor{set auto-load local-gdbinit}
28077 @kindex set auto-load local-gdbinit
28078 @item set auto-load local-gdbinit [on|off]
28079 Enable or disable the auto-loading of canned sequences of commands
28080 (@pxref{Sequences}) found in init file in the current directory.
28082 @anchor{show auto-load local-gdbinit}
28083 @kindex show auto-load local-gdbinit
28084 @item show auto-load local-gdbinit
28085 Show whether auto-loading of canned sequences of commands from init file in the
28086 current directory is enabled or disabled.
28088 @anchor{info auto-load local-gdbinit}
28089 @kindex info auto-load local-gdbinit
28090 @item info auto-load local-gdbinit
28091 Print whether canned sequences of commands from init file in the
28092 current directory have been auto-loaded.
28095 @node libthread_db.so.1 file
28096 @subsection Automatically loading thread debugging library
28097 @cindex auto-loading libthread_db.so.1
28099 This feature is currently present only on @sc{gnu}/Linux native hosts.
28101 @value{GDBN} reads in some cases thread debugging library from places specific
28102 to the inferior (@pxref{set libthread-db-search-path}).
28104 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
28105 without checking this @samp{set auto-load libthread-db} switch as system
28106 libraries have to be trusted in general. In all other cases of
28107 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
28108 auto-load libthread-db} is enabled before trying to open such thread debugging
28111 Note that loading of this debugging library also requires accordingly configured
28112 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
28115 @anchor{set auto-load libthread-db}
28116 @kindex set auto-load libthread-db
28117 @item set auto-load libthread-db [on|off]
28118 Enable or disable the auto-loading of inferior specific thread debugging library.
28120 @anchor{show auto-load libthread-db}
28121 @kindex show auto-load libthread-db
28122 @item show auto-load libthread-db
28123 Show whether auto-loading of inferior specific thread debugging library is
28124 enabled or disabled.
28126 @anchor{info auto-load libthread-db}
28127 @kindex info auto-load libthread-db
28128 @item info auto-load libthread-db
28129 Print the list of all loaded inferior specific thread debugging libraries and
28130 for each such library print list of inferior @var{pid}s using it.
28133 @node Auto-loading safe path
28134 @subsection Security restriction for auto-loading
28135 @cindex auto-loading safe-path
28137 As the files of inferior can come from untrusted source (such as submitted by
28138 an application user) @value{GDBN} does not always load any files automatically.
28139 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
28140 directories trusted for loading files not explicitly requested by user.
28141 Each directory can also be a shell wildcard pattern.
28143 If the path is not set properly you will see a warning and the file will not
28148 Reading symbols from /home/user/gdb/gdb...
28149 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
28150 declined by your `auto-load safe-path' set
28151 to "$debugdir:$datadir/auto-load".
28152 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
28153 declined by your `auto-load safe-path' set
28154 to "$debugdir:$datadir/auto-load".
28158 To instruct @value{GDBN} to go ahead and use the init files anyway,
28159 invoke @value{GDBN} like this:
28162 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
28165 The list of trusted directories is controlled by the following commands:
28168 @anchor{set auto-load safe-path}
28169 @kindex set auto-load safe-path
28170 @item set auto-load safe-path @r{[}@var{directories}@r{]}
28171 Set the list of directories (and their subdirectories) trusted for automatic
28172 loading and execution of scripts. You can also enter a specific trusted file.
28173 Each directory can also be a shell wildcard pattern; wildcards do not match
28174 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
28175 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
28176 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
28177 its default value as specified during @value{GDBN} compilation.
28179 The list of directories uses path separator (@samp{:} on GNU and Unix
28180 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
28181 to the @env{PATH} environment variable.
28183 @anchor{show auto-load safe-path}
28184 @kindex show auto-load safe-path
28185 @item show auto-load safe-path
28186 Show the list of directories trusted for automatic loading and execution of
28189 @anchor{add-auto-load-safe-path}
28190 @kindex add-auto-load-safe-path
28191 @item add-auto-load-safe-path
28192 Add an entry (or list of entries) to the list of directories trusted for
28193 automatic loading and execution of scripts. Multiple entries may be delimited
28194 by the host platform path separator in use.
28197 This variable defaults to what @code{--with-auto-load-dir} has been configured
28198 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
28199 substitution applies the same as for @ref{set auto-load scripts-directory}.
28200 The default @code{set auto-load safe-path} value can be also overriden by
28201 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
28203 Setting this variable to @file{/} disables this security protection,
28204 corresponding @value{GDBN} configuration option is
28205 @option{--without-auto-load-safe-path}.
28206 This variable is supposed to be set to the system directories writable by the
28207 system superuser only. Users can add their source directories in init files in
28208 their home directories (@pxref{Home Directory Init File}). See also deprecated
28209 init file in the current directory
28210 (@pxref{Init File in the Current Directory during Startup}).
28212 To force @value{GDBN} to load the files it declined to load in the previous
28213 example, you could use one of the following ways:
28216 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
28217 Specify this trusted directory (or a file) as additional component of the list.
28218 You have to specify also any existing directories displayed by
28219 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
28221 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
28222 Specify this directory as in the previous case but just for a single
28223 @value{GDBN} session.
28225 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
28226 Disable auto-loading safety for a single @value{GDBN} session.
28227 This assumes all the files you debug during this @value{GDBN} session will come
28228 from trusted sources.
28230 @item @kbd{./configure --without-auto-load-safe-path}
28231 During compilation of @value{GDBN} you may disable any auto-loading safety.
28232 This assumes all the files you will ever debug with this @value{GDBN} come from
28236 On the other hand you can also explicitly forbid automatic files loading which
28237 also suppresses any such warning messages:
28240 @item @kbd{gdb -iex "set auto-load no" @dots{}}
28241 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
28243 @item @file{~/.gdbinit}: @samp{set auto-load no}
28244 Disable auto-loading globally for the user
28245 (@pxref{Home Directory Init File}). While it is improbable, you could also
28246 use system init file instead (@pxref{System-wide configuration}).
28249 This setting applies to the file names as entered by user. If no entry matches
28250 @value{GDBN} tries as a last resort to also resolve all the file names into
28251 their canonical form (typically resolving symbolic links) and compare the
28252 entries again. @value{GDBN} already canonicalizes most of the filenames on its
28253 own before starting the comparison so a canonical form of directories is
28254 recommended to be entered.
28256 @node Auto-loading verbose mode
28257 @subsection Displaying files tried for auto-load
28258 @cindex auto-loading verbose mode
28260 For better visibility of all the file locations where you can place scripts to
28261 be auto-loaded with inferior --- or to protect yourself against accidental
28262 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
28263 all the files attempted to be loaded. Both existing and non-existing files may
28266 For example the list of directories from which it is safe to auto-load files
28267 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
28268 may not be too obvious while setting it up.
28271 (@value{GDBP}) set debug auto-load on
28272 (@value{GDBP}) file ~/src/t/true
28273 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
28274 for objfile "/tmp/true".
28275 auto-load: Updating directories of "/usr:/opt".
28276 auto-load: Using directory "/usr".
28277 auto-load: Using directory "/opt".
28278 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
28279 by your `auto-load safe-path' set to "/usr:/opt".
28283 @anchor{set debug auto-load}
28284 @kindex set debug auto-load
28285 @item set debug auto-load [on|off]
28286 Set whether to print the filenames attempted to be auto-loaded.
28288 @anchor{show debug auto-load}
28289 @kindex show debug auto-load
28290 @item show debug auto-load
28291 Show whether printing of the filenames attempted to be auto-loaded is turned
28295 @node Messages/Warnings
28296 @section Optional Warnings and Messages
28298 @cindex verbose operation
28299 @cindex optional warnings
28300 By default, @value{GDBN} is silent about its inner workings. If you are
28301 running on a slow machine, you may want to use the @code{set verbose}
28302 command. This makes @value{GDBN} tell you when it does a lengthy
28303 internal operation, so you will not think it has crashed.
28305 Currently, the messages controlled by @code{set verbose} are those
28306 which announce that the symbol table for a source file is being read;
28307 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
28310 @kindex set verbose
28311 @item set verbose on
28312 Enables @value{GDBN} output of certain informational messages.
28314 @item set verbose off
28315 Disables @value{GDBN} output of certain informational messages.
28317 @kindex show verbose
28319 Displays whether @code{set verbose} is on or off.
28322 By default, if @value{GDBN} encounters bugs in the symbol table of an
28323 object file, it is silent; but if you are debugging a compiler, you may
28324 find this information useful (@pxref{Symbol Errors, ,Errors Reading
28329 @kindex set complaints
28330 @item set complaints @var{limit}
28331 Permits @value{GDBN} to output @var{limit} complaints about each type of
28332 unusual symbols before becoming silent about the problem. Set
28333 @var{limit} to zero to suppress all complaints; set it to a large number
28334 to prevent complaints from being suppressed.
28336 @kindex show complaints
28337 @item show complaints
28338 Displays how many symbol complaints @value{GDBN} is permitted to produce.
28342 @anchor{confirmation requests}
28343 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
28344 lot of stupid questions to confirm certain commands. For example, if
28345 you try to run a program which is already running:
28349 The program being debugged has been started already.
28350 Start it from the beginning? (y or n)
28353 If you are willing to unflinchingly face the consequences of your own
28354 commands, you can disable this ``feature'':
28358 @kindex set confirm
28360 @cindex confirmation
28361 @cindex stupid questions
28362 @item set confirm off
28363 Disables confirmation requests. Note that running @value{GDBN} with
28364 the @option{--batch} option (@pxref{Mode Options, -batch}) also
28365 automatically disables confirmation requests.
28367 @item set confirm on
28368 Enables confirmation requests (the default).
28370 @kindex show confirm
28372 Displays state of confirmation requests.
28376 @cindex command tracing
28377 If you need to debug user-defined commands or sourced files you may find it
28378 useful to enable @dfn{command tracing}. In this mode each command will be
28379 printed as it is executed, prefixed with one or more @samp{+} symbols, the
28380 quantity denoting the call depth of each command.
28383 @kindex set trace-commands
28384 @cindex command scripts, debugging
28385 @item set trace-commands on
28386 Enable command tracing.
28387 @item set trace-commands off
28388 Disable command tracing.
28389 @item show trace-commands
28390 Display the current state of command tracing.
28393 @node Debugging Output
28394 @section Optional Messages about Internal Happenings
28395 @cindex optional debugging messages
28397 @value{GDBN} has commands that enable optional debugging messages from
28398 various @value{GDBN} subsystems; normally these commands are of
28399 interest to @value{GDBN} maintainers, or when reporting a bug. This
28400 section documents those commands.
28403 @kindex set exec-done-display
28404 @item set exec-done-display
28405 Turns on or off the notification of asynchronous commands'
28406 completion. When on, @value{GDBN} will print a message when an
28407 asynchronous command finishes its execution. The default is off.
28408 @kindex show exec-done-display
28409 @item show exec-done-display
28410 Displays the current setting of asynchronous command completion
28414 @cindex ARM AArch64
28415 @item set debug aarch64
28416 Turns on or off display of debugging messages related to ARM AArch64.
28417 The default is off.
28419 @item show debug aarch64
28420 Displays the current state of displaying debugging messages related to
28423 @cindex gdbarch debugging info
28424 @cindex architecture debugging info
28425 @item set debug arch
28426 Turns on or off display of gdbarch debugging info. The default is off
28427 @item show debug arch
28428 Displays the current state of displaying gdbarch debugging info.
28430 @item set debug aix-thread
28431 @cindex AIX threads
28432 Display debugging messages about inner workings of the AIX thread
28434 @item show debug aix-thread
28435 Show the current state of AIX thread debugging info display.
28437 @cindex AMD GPU debugging info
28438 @anchor{set debug amd-dbgapi-lib}
28439 @item set debug amd-dbgapi-lib
28440 @itemx show debug amd-dbgapi-lib
28442 The @code{set debug amd-dbgapi-lib log-level @var{level}} command can be used
28443 to enable diagnostic messages from the @samp{amd-dbgapi} library, where
28444 @var{level} can be:
28449 no logging is enabled
28452 fatal errors are reported
28455 fatal errors and warnings are reported
28458 fatal errors, warnings, and info messages are reported
28461 all messages are reported
28465 The @code{show debug amd-dbgapi-lib log-level} command displays the current
28466 @acronym{amd-dbgapi} library log level.
28468 @anchor{set debug amd-dbgapi}
28469 @item set debug amd-dbgapi
28470 @itemx show debug amd-dbgapi
28472 The @samp{set debug amd-dbgapi} command can be used
28473 to enable diagnostic messages in the @samp{amd-dbgapi} target. The
28474 @samp{show debug amd-dbgapi} command displays the current setting.
28475 @xref{set debug amd-dbgapi}.
28477 @item set debug check-physname
28479 Check the results of the ``physname'' computation. When reading DWARF
28480 debugging information for C@t{++}, @value{GDBN} attempts to compute
28481 each entity's name. @value{GDBN} can do this computation in two
28482 different ways, depending on exactly what information is present.
28483 When enabled, this setting causes @value{GDBN} to compute the names
28484 both ways and display any discrepancies.
28485 @item show debug check-physname
28486 Show the current state of ``physname'' checking.
28488 @item set debug coff-pe-read
28489 @cindex COFF/PE exported symbols
28490 Control display of debugging messages related to reading of COFF/PE
28491 exported symbols. The default is off.
28492 @item show debug coff-pe-read
28493 Displays the current state of displaying debugging messages related to
28494 reading of COFF/PE exported symbols.
28496 @item set debug dwarf-die
28498 Dump DWARF DIEs after they are read in.
28499 The value is the number of nesting levels to print.
28500 A value of zero turns off the display.
28501 @item show debug dwarf-die
28502 Show the current state of DWARF DIE debugging.
28504 @item set debug dwarf-line
28505 @cindex DWARF Line Tables
28506 Turns on or off display of debugging messages related to reading
28507 DWARF line tables. The default is 0 (off).
28508 A value of 1 provides basic information.
28509 A value greater than 1 provides more verbose information.
28510 @item show debug dwarf-line
28511 Show the current state of DWARF line table debugging.
28513 @item set debug dwarf-read
28514 @cindex DWARF Reading
28515 Turns on or off display of debugging messages related to reading
28516 DWARF debug info. The default is 0 (off).
28517 A value of 1 provides basic information.
28518 A value greater than 1 provides more verbose information.
28519 @item show debug dwarf-read
28520 Show the current state of DWARF reader debugging.
28522 @item set debug displaced
28523 @cindex displaced stepping debugging info
28524 Turns on or off display of @value{GDBN} debugging info for the
28525 displaced stepping support. The default is off.
28526 @item show debug displaced
28527 Displays the current state of displaying @value{GDBN} debugging info
28528 related to displaced stepping.
28530 @item set debug event
28531 @cindex event debugging info
28532 Turns on or off display of @value{GDBN} event debugging info. The
28534 @item show debug event
28535 Displays the current state of displaying @value{GDBN} event debugging
28538 @item set debug event-loop
28539 @cindex event-loop debugging
28540 Controls output of debugging info about the event loop. The possible
28541 values are @samp{off}, @samp{all} (shows all debugging info) and
28542 @samp{all-except-ui} (shows all debugging info except those about
28543 UI-related events).
28544 @item show debug event-loop
28545 Shows the current state of displaying debugging info about the event
28548 @item set debug expression
28549 @cindex expression debugging info
28550 Turns on or off display of debugging info about @value{GDBN}
28551 expression parsing. The default is off.
28552 @item show debug expression
28553 Displays the current state of displaying debugging info about
28554 @value{GDBN} expression parsing.
28556 @item set debug fbsd-lwp
28557 @cindex FreeBSD LWP debug messages
28558 Turns on or off debugging messages from the FreeBSD LWP debug support.
28559 @item show debug fbsd-lwp
28560 Show the current state of FreeBSD LWP debugging messages.
28562 @item set debug fbsd-nat
28563 @cindex FreeBSD native target debug messages
28564 Turns on or off debugging messages from the FreeBSD native target.
28565 @item show debug fbsd-nat
28566 Show the current state of FreeBSD native target debugging messages.
28568 @item set debug fortran-array-slicing
28569 @cindex fortran array slicing debugging info
28570 Turns on or off display of @value{GDBN} Fortran array slicing
28571 debugging info. The default is off.
28573 @item show debug fortran-array-slicing
28574 Displays the current state of displaying @value{GDBN} Fortran array
28575 slicing debugging info.
28577 @item set debug frame
28578 @cindex frame debugging info
28579 Turns on or off display of @value{GDBN} frame debugging info. The
28581 @item show debug frame
28582 Displays the current state of displaying @value{GDBN} frame debugging
28585 @item set debug gnu-nat
28586 @cindex @sc{gnu}/Hurd debug messages
28587 Turn on or off debugging messages from the @sc{gnu}/Hurd debug support.
28588 @item show debug gnu-nat
28589 Show the current state of @sc{gnu}/Hurd debugging messages.
28591 @item set debug infrun
28592 @cindex inferior debugging info
28593 Turns on or off display of @value{GDBN} debugging info for running the inferior.
28594 The default is off. @file{infrun.c} contains GDB's runtime state machine used
28595 for implementing operations such as single-stepping the inferior.
28596 @item show debug infrun
28597 Displays the current state of @value{GDBN} inferior debugging.
28599 @item set debug infcall
28600 @cindex inferior function call debugging info
28601 Turns on or off display of debugging info related to inferior function
28602 calls made by @value{GDBN}.
28603 @item show debug infcall
28604 Displays the current state of @value{GDBN} inferior function call debugging.
28606 @item set debug jit
28607 @cindex just-in-time compilation, debugging messages
28608 Turn on or off debugging messages from JIT debug support.
28609 @item show debug jit
28610 Displays the current state of @value{GDBN} JIT debugging.
28612 @item set debug linux-nat @r{[}on@r{|}off@r{]}
28613 @cindex @sc{gnu}/Linux native target debug messages
28614 @cindex Linux native targets
28615 Turn on or off debugging messages from the Linux native target debug support.
28616 @item show debug linux-nat
28617 Show the current state of Linux native target debugging messages.
28619 @item set debug linux-namespaces
28620 @cindex @sc{gnu}/Linux namespaces debug messages
28621 Turn on or off debugging messages from the Linux namespaces debug support.
28622 @item show debug linux-namespaces
28623 Show the current state of Linux namespaces debugging messages.
28625 @item set debug mach-o
28626 @cindex Mach-O symbols processing
28627 Control display of debugging messages related to Mach-O symbols
28628 processing. The default is off.
28629 @item show debug mach-o
28630 Displays the current state of displaying debugging messages related to
28631 reading of COFF/PE exported symbols.
28633 @item set debug notification
28634 @cindex remote async notification debugging info
28635 Turn on or off debugging messages about remote async notification.
28636 The default is off.
28637 @item show debug notification
28638 Displays the current state of remote async notification debugging messages.
28640 @item set debug observer
28641 @cindex observer debugging info
28642 Turns on or off display of @value{GDBN} observer debugging. This
28643 includes info such as the notification of observable events.
28644 @item show debug observer
28645 Displays the current state of observer debugging.
28647 @item set debug overload
28648 @cindex C@t{++} overload debugging info
28649 Turns on or off display of @value{GDBN} C@t{++} overload debugging
28650 info. This includes info such as ranking of functions, etc. The default
28652 @item show debug overload
28653 Displays the current state of displaying @value{GDBN} C@t{++} overload
28656 @cindex expression parser, debugging info
28657 @cindex debug expression parser
28658 @item set debug parser
28659 Turns on or off the display of expression parser debugging output.
28660 Internally, this sets the @code{yydebug} variable in the expression
28661 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
28662 details. The default is off.
28663 @item show debug parser
28664 Show the current state of expression parser debugging.
28666 @cindex packets, reporting on stdout
28667 @cindex serial connections, debugging
28668 @cindex debug remote protocol
28669 @cindex remote protocol debugging
28670 @cindex display remote packets
28671 @item set debug remote
28672 Turns on or off display of reports on all packets sent back and forth across
28673 the serial line to the remote machine. The info is printed on the
28674 @value{GDBN} standard output stream. The default is off.
28675 @item show debug remote
28676 Displays the state of display of remote packets.
28678 @item set debug remote-packet-max-chars
28679 Sets the maximum number of characters to display for each remote packet when
28680 @code{set debug remote} is on. This is useful to prevent @value{GDBN} from
28681 displaying lengthy remote packets and polluting the console.
28683 The default value is @code{512}, which means @value{GDBN} will truncate each
28684 remote packet after 512 bytes.
28686 Setting this option to @code{unlimited} will disable truncation and will output
28687 the full length of the remote packets.
28688 @item show debug remote-packet-max-chars
28689 Displays the number of bytes to output for remote packet debugging.
28691 @item set debug separate-debug-file
28692 Turns on or off display of debug output about separate debug file search.
28693 @item show debug separate-debug-file
28694 Displays the state of separate debug file search debug output.
28696 @item set debug serial
28697 Turns on or off display of @value{GDBN} serial debugging info. The
28699 @item show debug serial
28700 Displays the current state of displaying @value{GDBN} serial debugging
28703 @item set debug solib
28704 Turns on or off display of debugging messages related to shared libraries.
28705 The default is off.
28706 @item show debug solib
28707 Show the current state of solib debugging messages.
28709 @item set debug symbol-lookup
28710 @cindex symbol lookup
28711 Turns on or off display of debugging messages related to symbol lookup.
28712 The default is 0 (off).
28713 A value of 1 provides basic information.
28714 A value greater than 1 provides more verbose information.
28715 @item show debug symbol-lookup
28716 Show the current state of symbol lookup debugging messages.
28718 @item set debug symfile
28719 @cindex symbol file functions
28720 Turns on or off display of debugging messages related to symbol file functions.
28721 The default is off. @xref{Files}.
28722 @item show debug symfile
28723 Show the current state of symbol file debugging messages.
28725 @item set debug symtab-create
28726 @cindex symbol table creation
28727 Turns on or off display of debugging messages related to symbol table creation.
28728 The default is 0 (off).
28729 A value of 1 provides basic information.
28730 A value greater than 1 provides more verbose information.
28731 @item show debug symtab-create
28732 Show the current state of symbol table creation debugging.
28734 @item set debug target
28735 @cindex target debugging info
28736 Turns on or off display of @value{GDBN} target debugging info. This info
28737 includes what is going on at the target level of GDB, as it happens. The
28738 default is 0. Set it to 1 to track events, and to 2 to also track the
28739 value of large memory transfers.
28740 @item show debug target
28741 Displays the current state of displaying @value{GDBN} target debugging
28744 @item set debug timestamp
28745 @cindex timestamping debugging info
28746 Turns on or off display of timestamps with @value{GDBN} debugging info.
28747 When enabled, seconds and microseconds are displayed before each debugging
28749 @item show debug timestamp
28750 Displays the current state of displaying timestamps with @value{GDBN}
28753 @item set debug varobj
28754 @cindex variable object debugging info
28755 Turns on or off display of @value{GDBN} variable object debugging
28756 info. The default is off.
28757 @item show debug varobj
28758 Displays the current state of displaying @value{GDBN} variable object
28761 @item set debug xml
28762 @cindex XML parser debugging
28763 Turn on or off debugging messages for built-in XML parsers.
28764 @item show debug xml
28765 Displays the current state of XML debugging messages.
28767 @item set debug breakpoints
28768 @cindex breakpoint debugging info
28769 Turns on or off display of @value{GDBN} debugging info for breakpoint insertion
28770 and removal. The default is off.
28771 @item show debug breakpoints
28772 Displays the current state of displaying @value{GDBN} debugging info for
28773 breakpoint insertion and removal.
28776 @node Other Misc Settings
28777 @section Other Miscellaneous Settings
28778 @cindex miscellaneous settings
28781 @kindex set interactive-mode
28782 @item set interactive-mode
28783 If @code{on}, forces @value{GDBN} to assume that GDB was started
28784 in a terminal. In practice, this means that @value{GDBN} should wait
28785 for the user to answer queries generated by commands entered at
28786 the command prompt. If @code{off}, forces @value{GDBN} to operate
28787 in the opposite mode, and it uses the default answers to all queries.
28788 If @code{auto} (the default), @value{GDBN} tries to determine whether
28789 its standard input is a terminal, and works in interactive-mode if it
28790 is, non-interactively otherwise.
28792 In the vast majority of cases, the debugger should be able to guess
28793 correctly which mode should be used. But this setting can be useful
28794 in certain specific cases, such as running a MinGW @value{GDBN}
28795 inside a cygwin window.
28797 @kindex show interactive-mode
28798 @item show interactive-mode
28799 Displays whether the debugger is operating in interactive mode or not.
28803 @kindex set suppress-cli-notifications
28804 @item set suppress-cli-notifications
28805 If @code{on}, command-line-interface (CLI) notifications that are
28806 printed by @value{GDBN} are suppressed. If @code{off}, the
28807 notifications are printed as usual. The default value is @code{off}.
28808 CLI notifications occur when you change the selected context or when
28809 the program being debugged stops, as detailed below.
28812 @item User-selected context changes:
28813 When you change the selected context (i.e.@: the current inferior,
28814 thread and/or the frame), @value{GDBN} prints information about the
28815 new context. For example, the default behavior is below:
28819 [Switching to inferior 1 [process 634] (/tmp/test)]
28820 [Switching to thread 1 (process 634)]
28821 #0 main () at test.c:3
28826 When the notifications are suppressed, the new context is not printed:
28829 (gdb) set suppress-cli-notifications on
28834 @item The program being debugged stops:
28835 When the program you are debugging stops (e.g.@: because of hitting a
28836 breakpoint, completing source-stepping, an interrupt, etc.),
28837 @value{GDBN} prints information about the stop event. For example,
28838 below is a breakpoint hit:
28841 (gdb) break test.c:3
28842 Breakpoint 2 at 0x555555555155: file test.c, line 3.
28846 Breakpoint 2, main () at test.c:3
28851 When the notifications are suppressed, the output becomes:
28854 (gdb) break test.c:3
28855 Breakpoint 2 at 0x555555555155: file test.c, line 3.
28856 (gdb) set suppress-cli-notifications on
28862 Suppressing CLI notifications may be useful in scripts to obtain a
28863 reduced output from a list of commands.
28866 @kindex show suppress-cli-notifications
28867 @item show suppress-cli-notifications
28868 Displays whether printing CLI notifications is suppressed or not.
28871 @node Extending GDB
28872 @chapter Extending @value{GDBN}
28873 @cindex extending GDB
28875 @value{GDBN} provides several mechanisms for extension.
28876 @value{GDBN} also provides the ability to automatically load
28877 extensions when it reads a file for debugging. This allows the
28878 user to automatically customize @value{GDBN} for the program
28881 To facilitate the use of extension languages, @value{GDBN} is capable
28882 of evaluating the contents of a file. When doing so, @value{GDBN}
28883 can recognize which extension language is being used by looking at
28884 the filename extension. Files with an unrecognized filename extension
28885 are always treated as a @value{GDBN} Command Files.
28886 @xref{Command Files,, Command files}.
28888 You can control how @value{GDBN} evaluates these files with the following
28892 @kindex set script-extension
28893 @kindex show script-extension
28894 @item set script-extension off
28895 All scripts are always evaluated as @value{GDBN} Command Files.
28897 @item set script-extension soft
28898 The debugger determines the scripting language based on filename
28899 extension. If this scripting language is supported, @value{GDBN}
28900 evaluates the script using that language. Otherwise, it evaluates
28901 the file as a @value{GDBN} Command File.
28903 @item set script-extension strict
28904 The debugger determines the scripting language based on filename
28905 extension, and evaluates the script using that language. If the
28906 language is not supported, then the evaluation fails.
28908 @item show script-extension
28909 Display the current value of the @code{script-extension} option.
28913 @ifset SYSTEM_GDBINIT_DIR
28914 This setting is not used for files in the system-wide gdbinit directory.
28915 Files in that directory must have an extension matching their language,
28916 or have a @file{.gdb} extension to be interpreted as regular @value{GDBN}
28917 commands. @xref{Startup}.
28921 * Sequences:: Canned Sequences of @value{GDBN} Commands
28922 * Aliases:: Command Aliases
28923 * Python:: Extending @value{GDBN} using Python
28924 * Guile:: Extending @value{GDBN} using Guile
28925 * Auto-loading extensions:: Automatically loading extensions
28926 * Multiple Extension Languages:: Working with multiple extension languages
28930 @section Canned Sequences of Commands
28932 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
28933 Command Lists}), @value{GDBN} provides two ways to store sequences of
28934 commands for execution as a unit: user-defined commands and command
28938 * Define:: How to define your own commands
28939 * Hooks:: Hooks for user-defined commands
28940 * Command Files:: How to write scripts of commands to be stored in a file
28941 * Output:: Commands for controlled output
28942 * Auto-loading sequences:: Controlling auto-loaded command files
28946 @subsection User-defined Commands
28948 @cindex user-defined command
28949 @cindex arguments, to user-defined commands
28950 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
28951 which you assign a new name as a command. This is done with the
28952 @code{define} command. User commands may accept an unlimited number of arguments
28953 separated by whitespace. Arguments are accessed within the user command
28954 via @code{$arg0@dots{}$argN}. A trivial example:
28958 print $arg0 + $arg1 + $arg2
28963 To execute the command use:
28970 This defines the command @code{adder}, which prints the sum of
28971 its three arguments. Note the arguments are text substitutions, so they may
28972 reference variables, use complex expressions, or even perform inferior
28975 @cindex argument count in user-defined commands
28976 @cindex how many arguments (user-defined commands)
28977 In addition, @code{$argc} may be used to find out how many arguments have
28983 print $arg0 + $arg1
28986 print $arg0 + $arg1 + $arg2
28991 Combining with the @code{eval} command (@pxref{eval}) makes it easier
28992 to process a variable number of arguments:
28999 eval "set $sum = $sum + $arg%d", $i
29009 @item define @var{commandname}
29010 Define a command named @var{commandname}. If there is already a command
29011 by that name, you are asked to confirm that you want to redefine it.
29012 The argument @var{commandname} may be a bare command name consisting of letters,
29013 numbers, dashes, dots, and underscores. It may also start with any
29014 predefined or user-defined prefix command.
29015 For example, @samp{define target my-target} creates
29016 a user-defined @samp{target my-target} command.
29018 The definition of the command is made up of other @value{GDBN} command lines,
29019 which are given following the @code{define} command. The end of these
29020 commands is marked by a line containing @code{end}.
29023 @kindex end@r{ (user-defined commands)}
29024 @item document @var{commandname}
29025 Document the user-defined command @var{commandname}, so that it can be
29026 accessed by @code{help}. The command @var{commandname} must already be
29027 defined. This command reads lines of documentation just as @code{define}
29028 reads the lines of the command definition, ending with @code{end}.
29029 After the @code{document} command is finished, @code{help} on command
29030 @var{commandname} displays the documentation you have written.
29032 You may use the @code{document} command again to change the
29033 documentation of a command. Redefining the command with @code{define}
29034 does not change the documentation.
29036 It is also possible to document user-defined aliases. The alias documentation
29037 will then be used by the @code{help} and @code{apropos} commands
29038 instead of the documentation of the aliased command.
29039 Documenting a user-defined alias is particularly useful when defining
29040 an alias as a set of nested @code{with} commands
29041 (@pxref{Command aliases default args}).
29043 @kindex define-prefix
29044 @item define-prefix @var{commandname}
29045 Define or mark the command @var{commandname} as a user-defined prefix
29046 command. Once marked, @var{commandname} can be used as prefix command
29047 by the @code{define} command.
29048 Note that @code{define-prefix} can be used with a not yet defined
29049 @var{commandname}. In such a case, @var{commandname} is defined as
29050 an empty user-defined command.
29051 In case you redefine a command that was marked as a user-defined
29052 prefix command, the subcommands of the redefined command are kept
29053 (and @value{GDBN} indicates so to the user).
29057 (@value{GDBP}) define-prefix abc
29058 (@value{GDBP}) define-prefix abc def
29059 (@value{GDBP}) define abc def
29060 Type commands for definition of "abc def".
29061 End with a line saying just "end".
29062 >echo command initial def\n
29064 (@value{GDBP}) define abc def ghi
29065 Type commands for definition of "abc def ghi".
29066 End with a line saying just "end".
29067 >echo command ghi\n
29069 (@value{GDBP}) define abc def
29070 Keeping subcommands of prefix command "def".
29071 Redefine command "def"? (y or n) y
29072 Type commands for definition of "abc def".
29073 End with a line saying just "end".
29074 >echo command def\n
29076 (@value{GDBP}) abc def ghi
29078 (@value{GDBP}) abc def
29083 @kindex dont-repeat
29084 @cindex don't repeat command
29086 Used inside a user-defined command, this tells @value{GDBN} that this
29087 command should not be repeated when the user hits @key{RET}
29088 (@pxref{Command Syntax, repeat last command}).
29090 @kindex help user-defined
29091 @item help user-defined
29092 List all user-defined commands and all python commands defined in class
29093 COMMAND_USER. The first line of the documentation or docstring is
29098 @itemx show user @var{commandname}
29099 Display the @value{GDBN} commands used to define @var{commandname} (but
29100 not its documentation). If no @var{commandname} is given, display the
29101 definitions for all user-defined commands.
29102 This does not work for user-defined python commands.
29104 @cindex infinite recursion in user-defined commands
29105 @kindex show max-user-call-depth
29106 @kindex set max-user-call-depth
29107 @item show max-user-call-depth
29108 @itemx set max-user-call-depth
29109 The value of @code{max-user-call-depth} controls how many recursion
29110 levels are allowed in user-defined commands before @value{GDBN} suspects an
29111 infinite recursion and aborts the command.
29112 This does not apply to user-defined python commands.
29115 In addition to the above commands, user-defined commands frequently
29116 use control flow commands, described in @ref{Command Files}.
29118 When user-defined commands are executed, the
29119 commands of the definition are not printed. An error in any command
29120 stops execution of the user-defined command.
29122 If used interactively, commands that would ask for confirmation proceed
29123 without asking when used inside a user-defined command. Many @value{GDBN}
29124 commands that normally print messages to say what they are doing omit the
29125 messages when used in a user-defined command.
29128 @subsection User-defined Command Hooks
29129 @cindex command hooks
29130 @cindex hooks, for commands
29131 @cindex hooks, pre-command
29134 You may define @dfn{hooks}, which are a special kind of user-defined
29135 command. Whenever you run the command @samp{foo}, if the user-defined
29136 command @samp{hook-foo} exists, it is executed (with no arguments)
29137 before that command.
29139 @cindex hooks, post-command
29141 A hook may also be defined which is run after the command you executed.
29142 Whenever you run the command @samp{foo}, if the user-defined command
29143 @samp{hookpost-foo} exists, it is executed (with no arguments) after
29144 that command. Post-execution hooks may exist simultaneously with
29145 pre-execution hooks, for the same command.
29147 It is valid for a hook to call the command which it hooks. If this
29148 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
29150 @c It would be nice if hookpost could be passed a parameter indicating
29151 @c if the command it hooks executed properly or not. FIXME!
29153 @kindex stop@r{, a pseudo-command}
29154 In addition, a pseudo-command, @samp{stop} exists. Defining
29155 (@samp{hook-stop}) makes the associated commands execute every time
29156 execution stops in your program: before breakpoint commands are run,
29157 displays are printed, or the stack frame is printed.
29159 For example, to ignore @code{SIGALRM} signals while
29160 single-stepping, but treat them normally during normal execution,
29165 handle SIGALRM nopass
29169 handle SIGALRM pass
29172 define hook-continue
29173 handle SIGALRM pass
29177 As a further example, to hook at the beginning and end of the @code{echo}
29178 command, and to add extra text to the beginning and end of the message,
29186 define hookpost-echo
29190 (@value{GDBP}) echo Hello World
29191 <<<---Hello World--->>>
29196 You can define a hook for any single-word command in @value{GDBN}, but
29197 not for command aliases; you should define a hook for the basic command
29198 name, e.g.@: @code{backtrace} rather than @code{bt}.
29199 @c FIXME! So how does Joe User discover whether a command is an alias
29201 You can hook a multi-word command by adding @code{hook-} or
29202 @code{hookpost-} to the last word of the command, e.g.@:
29203 @samp{define target hook-remote} to add a hook to @samp{target remote}.
29205 If an error occurs during the execution of your hook, execution of
29206 @value{GDBN} commands stops and @value{GDBN} issues a prompt
29207 (before the command that you actually typed had a chance to run).
29209 If you try to define a hook which does not match any known command, you
29210 get a warning from the @code{define} command.
29212 @node Command Files
29213 @subsection Command Files
29215 @cindex command files
29216 @cindex scripting commands
29217 A command file for @value{GDBN} is a text file made of lines that are
29218 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
29219 also be included. An empty line in a command file does nothing; it
29220 does not mean to repeat the last command, as it would from the
29223 You can request the execution of a command file with the @code{source}
29224 command. Note that the @code{source} command is also used to evaluate
29225 scripts that are not Command Files. The exact behavior can be configured
29226 using the @code{script-extension} setting.
29227 @xref{Extending GDB,, Extending GDB}.
29231 @cindex execute commands from a file
29232 @item source [-s] [-v] @var{filename}
29233 Execute the command file @var{filename}.
29236 The lines in a command file are generally executed sequentially,
29237 unless the order of execution is changed by one of the
29238 @emph{flow-control commands} described below. The commands are not
29239 printed as they are executed. An error in any command terminates
29240 execution of the command file and control is returned to the console.
29242 @value{GDBN} first searches for @var{filename} in the current directory.
29243 If the file is not found there, and @var{filename} does not specify a
29244 directory, then @value{GDBN} also looks for the file on the source search path
29245 (specified with the @samp{directory} command);
29246 except that @file{$cdir} is not searched because the compilation directory
29247 is not relevant to scripts.
29249 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
29250 on the search path even if @var{filename} specifies a directory.
29251 The search is done by appending @var{filename} to each element of the
29252 search path. So, for example, if @var{filename} is @file{mylib/myscript}
29253 and the search path contains @file{/home/user} then @value{GDBN} will
29254 look for the script @file{/home/user/mylib/myscript}.
29255 The search is also done if @var{filename} is an absolute path.
29256 For example, if @var{filename} is @file{/tmp/myscript} and
29257 the search path contains @file{/home/user} then @value{GDBN} will
29258 look for the script @file{/home/user/tmp/myscript}.
29259 For DOS-like systems, if @var{filename} contains a drive specification,
29260 it is stripped before concatenation. For example, if @var{filename} is
29261 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
29262 will look for the script @file{c:/tmp/myscript}.
29264 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
29265 each command as it is executed. The option must be given before
29266 @var{filename}, and is interpreted as part of the filename anywhere else.
29268 Commands that would ask for confirmation if used interactively proceed
29269 without asking when used in a command file. Many @value{GDBN} commands that
29270 normally print messages to say what they are doing omit the messages
29271 when called from command files.
29273 @value{GDBN} also accepts command input from standard input. In this
29274 mode, normal output goes to standard output and error output goes to
29275 standard error. Errors in a command file supplied on standard input do
29276 not terminate execution of the command file---execution continues with
29280 gdb < cmds > log 2>&1
29283 (The syntax above will vary depending on the shell used.) This example
29284 will execute commands from the file @file{cmds}. All output and errors
29285 would be directed to @file{log}.
29287 Since commands stored on command files tend to be more general than
29288 commands typed interactively, they frequently need to deal with
29289 complicated situations, such as different or unexpected values of
29290 variables and symbols, changes in how the program being debugged is
29291 built, etc. @value{GDBN} provides a set of flow-control commands to
29292 deal with these complexities. Using these commands, you can write
29293 complex scripts that loop over data structures, execute commands
29294 conditionally, etc.
29301 This command allows to include in your script conditionally executed
29302 commands. The @code{if} command takes a single argument, which is an
29303 expression to evaluate. It is followed by a series of commands that
29304 are executed only if the expression is true (its value is nonzero).
29305 There can then optionally be an @code{else} line, followed by a series
29306 of commands that are only executed if the expression was false. The
29307 end of the list is marked by a line containing @code{end}.
29311 This command allows to write loops. Its syntax is similar to
29312 @code{if}: the command takes a single argument, which is an expression
29313 to evaluate, and must be followed by the commands to execute, one per
29314 line, terminated by an @code{end}. These commands are called the
29315 @dfn{body} of the loop. The commands in the body of @code{while} are
29316 executed repeatedly as long as the expression evaluates to true.
29320 This command exits the @code{while} loop in whose body it is included.
29321 Execution of the script continues after that @code{while}s @code{end}
29324 @kindex loop_continue
29325 @item loop_continue
29326 This command skips the execution of the rest of the body of commands
29327 in the @code{while} loop in whose body it is included. Execution
29328 branches to the beginning of the @code{while} loop, where it evaluates
29329 the controlling expression.
29331 @kindex end@r{ (if/else/while commands)}
29333 Terminate the block of commands that are the body of @code{if},
29334 @code{else}, or @code{while} flow-control commands.
29339 @subsection Commands for Controlled Output
29341 During the execution of a command file or a user-defined command, normal
29342 @value{GDBN} output is suppressed; the only output that appears is what is
29343 explicitly printed by the commands in the definition. This section
29344 describes three commands useful for generating exactly the output you
29349 @item echo @var{text}
29350 @c I do not consider backslash-space a standard C escape sequence
29351 @c because it is not in ANSI.
29352 Print @var{text}. Nonprinting characters can be included in
29353 @var{text} using C escape sequences, such as @samp{\n} to print a
29354 newline. @strong{No newline is printed unless you specify one.}
29355 In addition to the standard C escape sequences, a backslash followed
29356 by a space stands for a space. This is useful for displaying a
29357 string with spaces at the beginning or the end, since leading and
29358 trailing spaces are otherwise trimmed from all arguments.
29359 To print @samp{@w{ }and foo =@w{ }}, use the command
29360 @samp{echo \@w{ }and foo = \@w{ }}.
29362 A backslash at the end of @var{text} can be used, as in C, to continue
29363 the command onto subsequent lines. For example,
29366 echo This is some text\n\
29367 which is continued\n\
29368 onto several lines.\n
29371 produces the same output as
29374 echo This is some text\n
29375 echo which is continued\n
29376 echo onto several lines.\n
29380 @item output @var{expression}
29381 Print the value of @var{expression} and nothing but that value: no
29382 newlines, no @samp{$@var{nn} = }. The value is not entered in the
29383 value history either. @xref{Expressions, ,Expressions}, for more information
29386 @item output/@var{fmt} @var{expression}
29387 Print the value of @var{expression} in format @var{fmt}. You can use
29388 the same formats as for @code{print}. @xref{Output Formats,,Output
29389 Formats}, for more information.
29392 @item printf @var{template}, @var{expressions}@dots{}
29393 Print the values of one or more @var{expressions} under the control of
29394 the string @var{template}. To print several values, make
29395 @var{expressions} be a comma-separated list of individual expressions,
29396 which may be either numbers or pointers. Their values are printed as
29397 specified by @var{template}, exactly as a C program would do by
29398 executing the code below:
29401 printf (@var{template}, @var{expressions}@dots{});
29404 As in @code{C} @code{printf}, ordinary characters in @var{template}
29405 are printed verbatim, while @dfn{conversion specification} introduced
29406 by the @samp{%} character cause subsequent @var{expressions} to be
29407 evaluated, their values converted and formatted according to type and
29408 style information encoded in the conversion specifications, and then
29411 For example, you can print two values in hex like this:
29414 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
29417 @code{printf} supports all the standard @code{C} conversion
29418 specifications, including the flags and modifiers between the @samp{%}
29419 character and the conversion letter, with the following exceptions:
29423 The argument-ordering modifiers, such as @samp{2$}, are not supported.
29426 The modifier @samp{*} is not supported for specifying precision or
29430 The @samp{'} flag (for separation of digits into groups according to
29431 @code{LC_NUMERIC'}) is not supported.
29434 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
29438 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
29441 The conversion letters @samp{a} and @samp{A} are not supported.
29445 Note that the @samp{ll} type modifier is supported only if the
29446 underlying @code{C} implementation used to build @value{GDBN} supports
29447 the @code{long long int} type, and the @samp{L} type modifier is
29448 supported only if @code{long double} type is available.
29450 As in @code{C}, @code{printf} supports simple backslash-escape
29451 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
29452 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
29453 single character. Octal and hexadecimal escape sequences are not
29456 Additionally, @code{printf} supports conversion specifications for DFP
29457 (@dfn{Decimal Floating Point}) types using the following length modifiers
29458 together with a floating point specifier.
29463 @samp{H} for printing @code{Decimal32} types.
29466 @samp{D} for printing @code{Decimal64} types.
29469 @samp{DD} for printing @code{Decimal128} types.
29472 If the underlying @code{C} implementation used to build @value{GDBN} has
29473 support for the three length modifiers for DFP types, other modifiers
29474 such as width and precision will also be available for @value{GDBN} to use.
29476 In case there is no such @code{C} support, no additional modifiers will be
29477 available and the value will be printed in the standard way.
29479 Here's an example of printing DFP types using the above conversion letters:
29481 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
29484 @anchor{%V Format Specifier}
29485 Additionally, @code{printf} supports a special @samp{%V} output format.
29486 This format prints the string representation of an expression just as
29487 @value{GDBN} would produce with the standard @kbd{print} command
29488 (@pxref{Data, ,Examining Data}):
29491 (@value{GDBP}) print array
29492 $1 = @{0, 1, 2, 3, 4, 5@}
29493 (@value{GDBP}) printf "Array is: %V\n", array
29494 Array is: @{0, 1, 2, 3, 4, 5@}
29497 It is possible to include print options with the @samp{%V} format by
29498 placing them in @samp{[...]} immediately after the @samp{%V}, like
29502 (@value{GDBP}) printf "Array is: %V[-array-indexes on]\n", array
29503 Array is: @{[0] = 0, [1] = 1, [2] = 2, [3] = 3, [4] = 4, [5] = 5@}
29506 If you need to print a literal @samp{[} directly after a @samp{%V}, then
29507 just include an empty print options list:
29510 (@value{GDBP}) printf "Array is: %V[][Hello]\n", array
29511 Array is: @{0, 1, 2, 3, 4, 5@}[Hello]
29516 @item eval @var{template}, @var{expressions}@dots{}
29517 Convert the values of one or more @var{expressions} under the control of
29518 the string @var{template} to a command line, and call it.
29522 @node Auto-loading sequences
29523 @subsection Controlling auto-loading native @value{GDBN} scripts
29524 @cindex native script auto-loading
29526 When a new object file is read (for example, due to the @code{file}
29527 command, or because the inferior has loaded a shared library),
29528 @value{GDBN} will look for the command file @file{@var{objfile}-gdb.gdb}.
29529 @xref{Auto-loading extensions}.
29531 Auto-loading can be enabled or disabled,
29532 and the list of auto-loaded scripts can be printed.
29535 @anchor{set auto-load gdb-scripts}
29536 @kindex set auto-load gdb-scripts
29537 @item set auto-load gdb-scripts [on|off]
29538 Enable or disable the auto-loading of canned sequences of commands scripts.
29540 @anchor{show auto-load gdb-scripts}
29541 @kindex show auto-load gdb-scripts
29542 @item show auto-load gdb-scripts
29543 Show whether auto-loading of canned sequences of commands scripts is enabled or
29546 @anchor{info auto-load gdb-scripts}
29547 @kindex info auto-load gdb-scripts
29548 @cindex print list of auto-loaded canned sequences of commands scripts
29549 @item info auto-load gdb-scripts [@var{regexp}]
29550 Print the list of all canned sequences of commands scripts that @value{GDBN}
29554 If @var{regexp} is supplied only canned sequences of commands scripts with
29555 matching names are printed.
29558 @section Command Aliases
29559 @cindex aliases for commands
29561 Aliases allow you to define alternate spellings for existing commands.
29562 For example, if a new @value{GDBN} command defined in Python
29563 (@pxref{Python}) has a long name, it is handy to have an abbreviated
29564 version of it that involves less typing.
29566 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
29567 of the @samp{step} command even though it is otherwise an ambiguous
29568 abbreviation of other commands like @samp{set} and @samp{show}.
29570 Aliases are also used to provide shortened or more common versions
29571 of multi-word commands. For example, @value{GDBN} provides the
29572 @samp{tty} alias of the @samp{set inferior-tty} command.
29574 You can define a new alias with the @samp{alias} command.
29579 @item alias [-a] [--] @var{alias} = @var{command} [@var{default-args}]
29583 @var{alias} specifies the name of the new alias. Each word of
29584 @var{alias} must consist of letters, numbers, dashes and underscores.
29586 @var{command} specifies the name of an existing command
29587 that is being aliased.
29589 @var{command} can also be the name of an existing alias. In this
29590 case, @var{command} cannot be an alias that has default arguments.
29592 The @samp{-a} option specifies that the new alias is an abbreviation
29593 of the command. Abbreviations are not used in command completion.
29595 The @samp{--} option specifies the end of options,
29596 and is useful when @var{alias} begins with a dash.
29598 You can specify @var{default-args} for your alias. These
29599 @var{default-args} will be automatically added before the alias
29600 arguments typed explicitly on the command line.
29602 For example, the below defines an alias @code{btfullall} that shows all local
29603 variables and all frame arguments:
29605 (@value{GDBP}) alias btfullall = backtrace -full -frame-arguments all
29608 For more information about @var{default-args}, see @ref{Command
29609 aliases default args, ,Default Arguments}.
29611 Here is a simple example showing how to make an abbreviation of a
29612 command so that there is less to type. Suppose you were tired of
29613 typing @samp{disas}, the current shortest unambiguous abbreviation of
29614 the @samp{disassemble} command and you wanted an even shorter version
29615 named @samp{di}. The following will accomplish this.
29618 (@value{GDBP}) alias -a di = disas
29621 Note that aliases are different from user-defined commands. With a
29622 user-defined command, you also need to write documentation for it with
29623 the @samp{document} command. An alias automatically picks up the
29624 documentation of the existing command.
29626 Here is an example where we make @samp{elms} an abbreviation of
29627 @samp{elements} in the @samp{set print elements} command.
29628 This is to show that you can make an abbreviation of any part
29632 (@value{GDBP}) alias -a set print elms = set print elements
29633 (@value{GDBP}) alias -a show print elms = show print elements
29634 (@value{GDBP}) set p elms 200
29635 (@value{GDBP}) show p elms
29636 Limit on string chars or array elements to print is 200.
29639 Note that if you are defining an alias of a @samp{set} command,
29640 and you want to have an alias for the corresponding @samp{show}
29641 command, then you need to define the latter separately.
29643 Unambiguously abbreviated commands are allowed in @var{command} and
29644 @var{alias}, just as they are normally.
29647 (@value{GDBP}) alias -a set pr elms = set p ele
29650 Finally, here is an example showing the creation of a one word
29651 alias for a more complex command.
29652 This creates alias @samp{spe} of the command @samp{set print elements}.
29655 (@value{GDBP}) alias spe = set print elements
29656 (@value{GDBP}) spe 20
29660 * Command aliases default args:: Default arguments for aliases
29663 @node Command aliases default args
29664 @subsection Default Arguments
29665 @cindex aliases for commands, default arguments
29667 You can tell @value{GDBN} to always prepend some default arguments to
29668 the list of arguments provided explicitly by the user when using a
29669 user-defined alias.
29671 If you repeatedly use the same arguments or options for a command, you
29672 can define an alias for this command and tell @value{GDBN} to
29673 automatically prepend these arguments or options to the list of
29674 arguments you type explicitly when using the alias@footnote{@value{GDBN}
29675 could easily accept default arguments for pre-defined commands and aliases,
29676 but it was deemed this would be confusing, and so is not allowed.}.
29678 For example, if you often use the command @code{thread apply all}
29679 specifying to work on the threads in ascending order and to continue in case it
29680 encounters an error, you can tell @value{GDBN} to automatically preprend
29681 the @code{-ascending} and @code{-c} options by using:
29684 (@value{GDBP}) alias thread apply asc-all = thread apply all -ascending -c
29687 Once you have defined this alias with its default args, any time you type
29688 the @code{thread apply asc-all} followed by @code{some arguments},
29689 @value{GDBN} will execute @code{thread apply all -ascending -c some arguments}.
29691 To have even less to type, you can also define a one word alias:
29693 (@value{GDBP}) alias t_a_c = thread apply all -ascending -c
29696 As usual, unambiguous abbreviations can be used for @var{alias}
29697 and @var{default-args}.
29699 The different aliases of a command do not share their default args.
29700 For example, you define a new alias @code{bt_ALL} showing all possible
29701 information and another alias @code{bt_SMALL} showing very limited information
29704 (@value{GDBP}) alias bt_ALL = backtrace -entry-values both -frame-arg all \
29705 -past-main -past-entry -full
29706 (@value{GDBP}) alias bt_SMALL = backtrace -entry-values no -frame-arg none \
29707 -past-main off -past-entry off
29710 (For more on using the @code{alias} command, see @ref{Aliases}.)
29712 Default args are not limited to the arguments and options of @var{command},
29713 but can specify nested commands if @var{command} accepts such a nested command
29715 For example, the below defines @code{faalocalsoftype} that lists the
29716 frames having locals of a certain type, together with the matching
29719 (@value{GDBP}) alias faalocalsoftype = frame apply all info locals -q -t
29720 (@value{GDBP}) faalocalsoftype int
29721 #1 0x55554f5e in sleeper_or_burner (v=0xdf50) at sleepers.c:86
29726 This is also very useful to define an alias for a set of nested @code{with}
29727 commands to have a particular combination of temporary settings. For example,
29728 the below defines the alias @code{pp10} that pretty prints an expression
29729 argument, with a maximum of 10 elements if the expression is a string or
29732 (@value{GDBP}) alias pp10 = with print pretty -- with print elements 10 -- print
29734 This defines the alias @code{pp10} as being a sequence of 3 commands.
29735 The first part @code{with print pretty --} temporarily activates the setting
29736 @code{set print pretty}, then launches the command that follows the separator
29738 The command following the first part is also a @code{with} command that
29739 temporarily changes the setting @code{set print elements} to 10, then
29740 launches the command that follows the second separator @code{--}.
29741 The third part @code{print} is the command the @code{pp10} alias will launch,
29742 using the temporary values of the settings and the arguments explicitly given
29744 For more information about the @code{with} command usage,
29745 see @ref{Command Settings}.
29747 By default, asking the help for an alias shows the documentation of
29748 the aliased command. When the alias is a set of nested commands, @code{help}
29749 of an alias shows the documentation of the first command. This help
29750 is not particularly useful for an alias such as @code{pp10}.
29751 For such an alias, it is useful to give a specific documentation
29752 using the @code{document} command (@pxref{Define, document}).
29755 @c Python docs live in a separate file.
29756 @include python.texi
29758 @c Guile docs live in a separate file.
29759 @include guile.texi
29761 @node Auto-loading extensions
29762 @section Auto-loading extensions
29763 @cindex auto-loading extensions
29765 @value{GDBN} provides two mechanisms for automatically loading
29766 extensions when a new object file is read (for example, due to the
29767 @code{file} command, or because the inferior has loaded a shared
29768 library): @file{@var{objfile}-gdb.@var{ext}} (@pxref{objfile-gdbdotext
29769 file,,The @file{@var{objfile}-gdb.@var{ext}} file}) and the
29770 @code{.debug_gdb_scripts} section of modern file formats like ELF
29771 (@pxref{dotdebug_gdb_scripts section,,The @code{.debug_gdb_scripts}
29772 section}). For a discussion of the differences between these two
29773 approaches see @ref{Which flavor to choose?}.
29775 The auto-loading feature is useful for supplying application-specific
29776 debugging commands and features.
29778 Auto-loading can be enabled or disabled,
29779 and the list of auto-loaded scripts can be printed.
29780 See the @samp{auto-loading} section of each extension language
29781 for more information.
29782 For @value{GDBN} command files see @ref{Auto-loading sequences}.
29783 For Python files see @ref{Python Auto-loading}.
29785 Note that loading of this script file also requires accordingly configured
29786 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
29789 * objfile-gdbdotext file:: The @file{@var{objfile}-gdb.@var{ext}} file
29790 * dotdebug_gdb_scripts section:: The @code{.debug_gdb_scripts} section
29791 * Which flavor to choose?:: Choosing between these approaches
29794 @node objfile-gdbdotext file
29795 @subsection The @file{@var{objfile}-gdb.@var{ext}} file
29796 @cindex @file{@var{objfile}-gdb.gdb}
29797 @cindex @file{@var{objfile}-gdb.py}
29798 @cindex @file{@var{objfile}-gdb.scm}
29800 When a new object file is read, @value{GDBN} looks for a file named
29801 @file{@var{objfile}-gdb.@var{ext}} (we call it @var{script-name} below),
29802 where @var{objfile} is the object file's name and
29803 where @var{ext} is the file extension for the extension language:
29806 @item @file{@var{objfile}-gdb.gdb}
29807 GDB's own command language
29808 @item @file{@var{objfile}-gdb.py}
29810 @item @file{@var{objfile}-gdb.scm}
29814 @var{script-name} is formed by ensuring that the file name of @var{objfile}
29815 is absolute, following all symlinks, and resolving @code{.} and @code{..}
29816 components, and appending the @file{-gdb.@var{ext}} suffix.
29817 If this file exists and is readable, @value{GDBN} will evaluate it as a
29818 script in the specified extension language.
29820 If this file does not exist, then @value{GDBN} will look for
29821 @var{script-name} file in all of the directories as specified below.
29822 (On MS-Windows/MS-DOS, the drive letter of the executable's leading
29823 directories is converted to a one-letter subdirectory, i.e.@:
29824 @file{d:/usr/bin/} is converted to @file{/d/usr/bin/}, because Windows
29825 filesystems disallow colons in file names.)
29827 Note that loading of these files requires an accordingly configured
29828 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
29830 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
29831 scripts normally according to its @file{.exe} filename. But if no scripts are
29832 found @value{GDBN} also tries script filenames matching the object file without
29833 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
29834 is attempted on any platform. This makes the script filenames compatible
29835 between Unix and MS-Windows hosts.
29838 @anchor{set auto-load scripts-directory}
29839 @kindex set auto-load scripts-directory
29840 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
29841 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
29842 may be delimited by the host platform path separator in use
29843 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
29845 Each entry here needs to be covered also by the security setting
29846 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
29848 @anchor{with-auto-load-dir}
29849 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
29850 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
29851 configuration option @option{--with-auto-load-dir}.
29853 Any reference to @file{$debugdir} will get replaced by
29854 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
29855 reference to @file{$datadir} will get replaced by @var{data-directory} which is
29856 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
29857 @file{$datadir} must be placed as a directory component --- either alone or
29858 delimited by @file{/} or @file{\} directory separators, depending on the host
29861 The list of directories uses path separator (@samp{:} on GNU and Unix
29862 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
29863 to the @env{PATH} environment variable.
29865 @anchor{show auto-load scripts-directory}
29866 @kindex show auto-load scripts-directory
29867 @item show auto-load scripts-directory
29868 Show @value{GDBN} auto-loaded scripts location.
29870 @anchor{add-auto-load-scripts-directory}
29871 @kindex add-auto-load-scripts-directory
29872 @item add-auto-load-scripts-directory @r{[}@var{directories}@dots{}@r{]}
29873 Add an entry (or list of entries) to the list of auto-loaded scripts locations.
29874 Multiple entries may be delimited by the host platform path separator in use.
29877 @value{GDBN} does not track which files it has already auto-loaded this way.
29878 @value{GDBN} will load the associated script every time the corresponding
29879 @var{objfile} is opened.
29880 So your @file{-gdb.@var{ext}} file should be careful to avoid errors if it
29881 is evaluated more than once.
29883 @node dotdebug_gdb_scripts section
29884 @subsection The @code{.debug_gdb_scripts} section
29885 @cindex @code{.debug_gdb_scripts} section
29887 For systems using file formats like ELF and COFF,
29888 when @value{GDBN} loads a new object file
29889 it will look for a special section named @code{.debug_gdb_scripts}.
29890 If this section exists, its contents is a list of null-terminated entries
29891 specifying scripts to load. Each entry begins with a non-null prefix byte that
29892 specifies the kind of entry, typically the extension language and whether the
29893 script is in a file or inlined in @code{.debug_gdb_scripts}.
29895 The following entries are supported:
29898 @item SECTION_SCRIPT_ID_PYTHON_FILE = 1
29899 @item SECTION_SCRIPT_ID_SCHEME_FILE = 3
29900 @item SECTION_SCRIPT_ID_PYTHON_TEXT = 4
29901 @item SECTION_SCRIPT_ID_SCHEME_TEXT = 6
29904 @subsubsection Script File Entries
29906 If the entry specifies a file, @value{GDBN} will look for the file first
29907 in the current directory and then along the source search path
29908 (@pxref{Source Path, ,Specifying Source Directories}),
29909 except that @file{$cdir} is not searched, since the compilation
29910 directory is not relevant to scripts.
29912 File entries can be placed in section @code{.debug_gdb_scripts} with,
29913 for example, this GCC macro for Python scripts.
29916 /* Note: The "MS" section flags are to remove duplicates. */
29917 #define DEFINE_GDB_PY_SCRIPT(script_name) \
29919 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
29920 .byte 1 /* Python */\n\
29921 .asciz \"" script_name "\"\n\
29927 For Guile scripts, replace @code{.byte 1} with @code{.byte 3}.
29928 Then one can reference the macro in a header or source file like this:
29931 DEFINE_GDB_PY_SCRIPT ("my-app-scripts.py")
29934 The script name may include directories if desired.
29936 Note that loading of this script file also requires accordingly configured
29937 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
29939 If the macro invocation is put in a header, any application or library
29940 using this header will get a reference to the specified script,
29941 and with the use of @code{"MS"} attributes on the section, the linker
29942 will remove duplicates.
29944 @subsubsection Script Text Entries
29946 Script text entries allow to put the executable script in the entry
29947 itself instead of loading it from a file.
29948 The first line of the entry, everything after the prefix byte and up to
29949 the first newline (@code{0xa}) character, is the script name, and must not
29950 contain any kind of space character, e.g., spaces or tabs.
29951 The rest of the entry, up to the trailing null byte, is the script to
29952 execute in the specified language. The name needs to be unique among
29953 all script names, as @value{GDBN} executes each script only once based
29956 Here is an example from file @file{py-section-script.c} in the @value{GDBN}
29960 #include "symcat.h"
29961 #include "gdb/section-scripts.h"
29963 ".pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n"
29964 ".byte " XSTRING (SECTION_SCRIPT_ID_PYTHON_TEXT) "\n"
29965 ".ascii \"gdb.inlined-script\\n\"\n"
29966 ".ascii \"class test_cmd (gdb.Command):\\n\"\n"
29967 ".ascii \" def __init__ (self):\\n\"\n"
29968 ".ascii \" super (test_cmd, self).__init__ ("
29969 "\\\"test-cmd\\\", gdb.COMMAND_OBSCURE)\\n\"\n"
29970 ".ascii \" def invoke (self, arg, from_tty):\\n\"\n"
29971 ".ascii \" print (\\\"test-cmd output, arg = %s\\\" % arg)\\n\"\n"
29972 ".ascii \"test_cmd ()\\n\"\n"
29978 Loading of inlined scripts requires a properly configured
29979 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
29980 The path to specify in @code{auto-load safe-path} is the path of the file
29981 containing the @code{.debug_gdb_scripts} section.
29983 @node Which flavor to choose?
29984 @subsection Which flavor to choose?
29986 Given the multiple ways of auto-loading extensions, it might not always
29987 be clear which one to choose. This section provides some guidance.
29990 Benefits of the @file{-gdb.@var{ext}} way:
29994 Can be used with file formats that don't support multiple sections.
29997 Ease of finding scripts for public libraries.
29999 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
30000 in the source search path.
30001 For publicly installed libraries, e.g., @file{libstdc++}, there typically
30002 isn't a source directory in which to find the script.
30005 Doesn't require source code additions.
30009 Benefits of the @code{.debug_gdb_scripts} way:
30013 Works with static linking.
30015 Scripts for libraries done the @file{-gdb.@var{ext}} way require an objfile to
30016 trigger their loading. When an application is statically linked the only
30017 objfile available is the executable, and it is cumbersome to attach all the
30018 scripts from all the input libraries to the executable's
30019 @file{-gdb.@var{ext}} script.
30022 Works with classes that are entirely inlined.
30024 Some classes can be entirely inlined, and thus there may not be an associated
30025 shared library to attach a @file{-gdb.@var{ext}} script to.
30028 Scripts needn't be copied out of the source tree.
30030 In some circumstances, apps can be built out of large collections of internal
30031 libraries, and the build infrastructure necessary to install the
30032 @file{-gdb.@var{ext}} scripts in a place where @value{GDBN} can find them is
30033 cumbersome. It may be easier to specify the scripts in the
30034 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
30035 top of the source tree to the source search path.
30038 @node Multiple Extension Languages
30039 @section Multiple Extension Languages
30041 The Guile and Python extension languages do not share any state,
30042 and generally do not interfere with each other.
30043 There are some things to be aware of, however.
30045 @subsection Python comes first
30047 Python was @value{GDBN}'s first extension language, and to avoid breaking
30048 existing behaviour Python comes first. This is generally solved by the
30049 ``first one wins'' principle. @value{GDBN} maintains a list of enabled
30050 extension languages, and when it makes a call to an extension language,
30051 (say to pretty-print a value), it tries each in turn until an extension
30052 language indicates it has performed the request (e.g., has returned the
30053 pretty-printed form of a value).
30054 This extends to errors while performing such requests: If an error happens
30055 while, for example, trying to pretty-print an object then the error is
30056 reported and any following extension languages are not tried.
30059 @chapter Command Interpreters
30060 @cindex command interpreters
30062 @value{GDBN} supports multiple command interpreters, and some command
30063 infrastructure to allow users or user interface writers to switch
30064 between interpreters or run commands in other interpreters.
30066 @value{GDBN} currently supports two command interpreters, the console
30067 interpreter (sometimes called the command-line interpreter or @sc{cli})
30068 and the machine interface interpreter (or @sc{gdb/mi}). This manual
30069 describes both of these interfaces in great detail.
30071 By default, @value{GDBN} will start with the console interpreter.
30072 However, the user may choose to start @value{GDBN} with another
30073 interpreter by specifying the @option{-i} or @option{--interpreter}
30074 startup options. Defined interpreters include:
30078 @cindex console interpreter
30079 The traditional console or command-line interpreter. This is the most often
30080 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
30081 @value{GDBN} will use this interpreter.
30085 @cindex Debugger Adapter Protocol
30086 When @value{GDBN} has been built with Python support, it also supports
30087 the Debugger Adapter Protocol. This protocol can be used by a
30088 debugger GUI or an IDE to communicate with @value{GDBN}. This
30089 protocol is documented at
30090 @url{https://microsoft.github.io/debug-adapter-protocol/}.
30091 @xref{Debugger Adapter Protocol}, for information about @value{GDBN}
30092 extensions to the protocol.
30095 @cindex mi interpreter
30096 The newest @sc{gdb/mi} interface (currently @code{mi3}). Used primarily
30097 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
30098 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
30102 @cindex mi3 interpreter
30103 The @sc{gdb/mi} interface introduced in @value{GDBN} 9.1.
30106 @cindex mi2 interpreter
30107 The @sc{gdb/mi} interface introduced in @value{GDBN} 6.0.
30111 @cindex invoke another interpreter
30113 @kindex interpreter-exec
30114 You may execute commands in any interpreter from the current
30115 interpreter using the appropriate command. If you are running the
30116 console interpreter, simply use the @code{interpreter-exec} command:
30119 interpreter-exec mi "-data-list-register-names"
30122 @sc{gdb/mi} has a similar command, although it is only available in versions of
30123 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
30125 Note that @code{interpreter-exec} only changes the interpreter for the
30126 duration of the specified command. It does not change the interpreter
30129 @cindex start a new independent interpreter
30131 Although you may only choose a single interpreter at startup, it is
30132 possible to run an independent interpreter on a specified input/output
30133 device (usually a tty).
30135 For example, consider a debugger GUI or IDE that wants to provide a
30136 @value{GDBN} console view. It may do so by embedding a terminal
30137 emulator widget in its GUI, starting @value{GDBN} in the traditional
30138 command-line mode with stdin/stdout/stderr redirected to that
30139 terminal, and then creating an MI interpreter running on a specified
30140 input/output device. The console interpreter created by @value{GDBN}
30141 at startup handles commands the user types in the terminal widget,
30142 while the GUI controls and synchronizes state with @value{GDBN} using
30143 the separate MI interpreter.
30145 To start a new secondary @dfn{user interface} running MI, use the
30146 @code{new-ui} command:
30149 @cindex new user interface
30151 new-ui @var{interpreter} @var{tty}
30154 The @var{interpreter} parameter specifies the interpreter to run.
30155 This accepts the same values as the @code{interpreter-exec} command.
30156 For example, @samp{console}, @samp{mi}, @samp{mi2}, etc. The
30157 @var{tty} parameter specifies the name of the bidirectional file the
30158 interpreter uses for input/output, usually the name of a
30159 pseudoterminal slave on Unix systems. For example:
30162 (@value{GDBP}) new-ui mi /dev/pts/9
30166 runs an MI interpreter on @file{/dev/pts/9}.
30169 @chapter @value{GDBN} Text User Interface
30171 @cindex Text User Interface
30173 The @value{GDBN} Text User Interface (TUI) is a terminal
30174 interface which uses the @code{curses} library to show the source
30175 file, the assembly output, the program registers and @value{GDBN}
30176 commands in separate text windows. The TUI mode is supported only
30177 on platforms where a suitable version of the @code{curses} library
30180 The TUI mode is enabled by default when you invoke @value{GDBN} as
30181 @samp{@value{GDBP} -tui}.
30182 You can also switch in and out of TUI mode while @value{GDBN} runs by
30183 using various TUI commands and key bindings, such as @command{tui
30184 enable} or @kbd{C-x C-a}. @xref{TUI Commands, ,TUI Commands}, and
30185 @ref{TUI Keys, ,TUI Key Bindings}.
30188 * TUI Overview:: TUI overview
30189 * TUI Keys:: TUI key bindings
30190 * TUI Single Key Mode:: TUI single key mode
30191 * TUI Mouse Support:: TUI mouse support
30192 * TUI Commands:: TUI-specific commands
30193 * TUI Configuration:: TUI configuration variables
30197 @section TUI Overview
30199 In TUI mode, @value{GDBN} can display several text windows:
30203 This window is the @value{GDBN} command window with the @value{GDBN}
30204 prompt and the @value{GDBN} output. The @value{GDBN} input is still
30205 managed using readline.
30208 The source window shows the source file of the program. The current
30209 line and active breakpoints are displayed in this window.
30212 The assembly window shows the disassembly output of the program.
30215 This window shows the processor registers. Registers are highlighted
30216 when their values change.
30219 The source and assembly windows show the current program position by
30220 highlighting the current line and marking it with a @samp{>} marker.
30221 By default, source and assembly code styling is disabled for the
30222 highlighted text, but you can enable it with the @code{set style
30223 tui-current-position on} command. @xref{Output Styling}.
30225 Breakpoints are indicated with two markers. The first marker
30226 indicates the breakpoint type:
30230 Breakpoint which was hit at least once.
30233 Breakpoint which was never hit.
30236 Hardware breakpoint which was hit at least once.
30239 Hardware breakpoint which was never hit.
30242 The second marker indicates whether the breakpoint is enabled or not:
30246 Breakpoint is enabled.
30249 Breakpoint is disabled.
30252 The source, assembly and register windows are updated when the current
30253 thread changes, when the frame changes, or when the program counter
30256 These windows are not all visible at the same time. The command
30257 window is always visible. The others can be arranged in several
30268 source and assembly,
30271 source and registers, or
30274 assembly and registers.
30277 These are the standard layouts, but other layouts can be defined.
30279 A status line above the command window shows the following information:
30283 Indicates the current @value{GDBN} target.
30284 (@pxref{Targets, ,Specifying a Debugging Target}).
30287 Gives the current process or thread number.
30288 When no process is being debugged, this field is set to @code{No process}.
30291 Gives the current function name for the selected frame.
30292 The name is demangled if demangling is turned on (@pxref{Print Settings}).
30293 When there is no symbol corresponding to the current program counter,
30294 the string @code{??} is displayed.
30297 Indicates the current line number for the selected frame.
30298 When the current line number is not known, the string @code{??} is displayed.
30301 Indicates the current program counter address.
30305 @section TUI Key Bindings
30306 @cindex TUI key bindings
30308 The TUI installs several key bindings in the readline keymaps
30309 @ifset SYSTEM_READLINE
30310 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
30312 @ifclear SYSTEM_READLINE
30313 (@pxref{Command Line Editing}).
30315 The following key bindings are installed for both TUI mode and the
30316 @value{GDBN} standard mode.
30325 Enter or leave the TUI mode. When leaving the TUI mode,
30326 the curses window management stops and @value{GDBN} operates using
30327 its standard mode, writing on the terminal directly. When reentering
30328 the TUI mode, control is given back to the curses windows.
30329 The screen is then refreshed.
30331 This key binding uses the bindable Readline function
30332 @code{tui-switch-mode}.
30336 Use a TUI layout with only one window. The layout will
30337 either be @samp{source} or @samp{assembly}. When the TUI mode
30338 is not active, it will switch to the TUI mode.
30340 Think of this key binding as the Emacs @kbd{C-x 1} binding.
30342 This key binding uses the bindable Readline function
30343 @code{tui-delete-other-windows}.
30347 Use a TUI layout with at least two windows. When the current
30348 layout already has two windows, the next layout with two windows is used.
30349 When a new layout is chosen, one window will always be common to the
30350 previous layout and the new one.
30352 Think of it as the Emacs @kbd{C-x 2} binding.
30354 This key binding uses the bindable Readline function
30355 @code{tui-change-windows}.
30359 Change the active window. The TUI associates several key bindings
30360 (like scrolling and arrow keys) with the active window. This command
30361 gives the focus to the next TUI window.
30363 Think of it as the Emacs @kbd{C-x o} binding.
30365 This key binding uses the bindable Readline function
30366 @code{tui-other-window}.
30370 Switch in and out of the TUI SingleKey mode that binds single
30371 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
30373 This key binding uses the bindable Readline function
30374 @code{next-keymap}.
30377 The following key bindings only work in the TUI mode:
30382 Scroll the active window one page up.
30386 Scroll the active window one page down.
30390 Scroll the active window one line up.
30394 Scroll the active window one line down.
30398 Scroll the active window one column left.
30402 Scroll the active window one column right.
30406 Refresh the screen.
30409 Because the arrow keys scroll the active window in the TUI mode, they
30410 are not available for their normal use by readline unless the command
30411 window has the focus. When another window is active, you must use
30412 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
30413 and @kbd{C-f} to control the command window.
30415 @node TUI Single Key Mode
30416 @section TUI Single Key Mode
30417 @cindex TUI single key mode
30419 The TUI also provides a @dfn{SingleKey} mode, which binds several
30420 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
30421 switch into this mode, where the following key bindings are used:
30424 @kindex c @r{(SingleKey TUI key)}
30428 @kindex d @r{(SingleKey TUI key)}
30432 @kindex f @r{(SingleKey TUI key)}
30436 @kindex n @r{(SingleKey TUI key)}
30440 @kindex o @r{(SingleKey TUI key)}
30442 nexti. The shortcut letter @samp{o} stands for ``step Over''.
30444 @kindex q @r{(SingleKey TUI key)}
30446 exit the SingleKey mode.
30448 @kindex r @r{(SingleKey TUI key)}
30452 @kindex s @r{(SingleKey TUI key)}
30456 @kindex i @r{(SingleKey TUI key)}
30458 stepi. The shortcut letter @samp{i} stands for ``step Into''.
30460 @kindex u @r{(SingleKey TUI key)}
30464 @kindex v @r{(SingleKey TUI key)}
30468 @kindex w @r{(SingleKey TUI key)}
30473 Other keys temporarily switch to the @value{GDBN} command prompt.
30474 The key that was pressed is inserted in the editing buffer so that
30475 it is possible to type most @value{GDBN} commands without interaction
30476 with the TUI SingleKey mode. Once the command is entered the TUI
30477 SingleKey mode is restored. The only way to permanently leave
30478 this mode is by typing @kbd{q} or @kbd{C-x s}.
30480 @cindex SingleKey keymap name
30481 If @value{GDBN} was built with Readline 8.0 or later, the TUI
30482 SingleKey keymap will be named @samp{SingleKey}. This can be used in
30483 @file{.inputrc} to add additional bindings to this keymap.
30485 @node TUI Mouse Support
30486 @section TUI Mouse Support
30487 @cindex TUI mouse support
30489 If the curses library supports the mouse, the TUI supports mouse
30492 The mouse wheel scrolls the appropriate window under the mouse cursor.
30494 The TUI itself does not directly support copying/pasting with the
30495 mouse. However, on Unix terminals, you can typically press and hold
30496 the @key{SHIFT} key on your keyboard to temporarily bypass
30497 @value{GDBN}'s TUI and access the terminal's native mouse copy/paste
30498 functionality (commonly, click-drag-release or double-click to select
30499 text, middle-click to paste). This copy/paste works with the
30500 terminal's selection buffer, as opposed to the TUI's buffer. Alternatively, to
30501 disable mouse support in the TUI entirely and give the terminal control over
30502 mouse clicks, turn off the @code{tui mouse-events} setting
30503 (@pxref{tui-mouse-events,,set tui mouse-events}).
30505 Python extensions can react to mouse clicks
30506 (@pxref{python-window-click,,Window.click}).
30509 @section TUI-specific Commands
30510 @cindex TUI commands
30512 The TUI has specific commands to control the text windows.
30513 These commands are always available, even when @value{GDBN} is not in
30514 the TUI mode. When @value{GDBN} is in the standard mode, most
30515 of these commands will automatically switch to the TUI mode.
30517 Note that if @value{GDBN}'s @code{stdout} is not connected to a
30518 terminal, or @value{GDBN} has been started with the machine interface
30519 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
30520 these commands will fail with an error, because it would not be
30521 possible or desirable to enable curses window management.
30526 Activate TUI mode. The last active TUI window layout will be used if
30527 TUI mode has previously been used in the current debugging session,
30528 otherwise a default layout is used.
30531 @kindex tui disable
30532 Disable TUI mode, returning to the console interpreter.
30534 @anchor{info_win_command}
30537 List the names and sizes of all currently displayed windows.
30539 @item tui new-layout @var{name} @var{window} @var{weight} @r{[}@var{window} @var{weight}@dots{}@r{]}
30540 @kindex tui new-layout
30541 Create a new TUI layout. The new layout will be named @var{name}, and
30542 can be accessed using the @code{layout} command (see below).
30544 Each @var{window} parameter is either the name of a window to display,
30545 or a window description. The windows will be displayed from top to
30546 bottom in the order listed.
30548 The names of the windows are the same as the ones given to the
30549 @code{focus} command (see below); additional, the @code{status}
30550 window can be specified. Note that, because it is of fixed height,
30551 the weight assigned to the status window is of no importance. It is
30552 conventional to use @samp{0} here.
30554 A window description looks a bit like an invocation of @code{tui
30555 new-layout}, and is of the form
30556 @{@r{[}@code{-horizontal}@r{]}@var{window} @var{weight} @r{[}@var{window} @var{weight}@dots{}@r{]}@}.
30558 This specifies a sub-layout. If @code{-horizontal} is given, the
30559 windows in this description will be arranged side-by-side, rather than
30562 Each @var{weight} is an integer. It is the weight of this window
30563 relative to all the other windows in the layout. These numbers are
30564 used to calculate how much of the screen is given to each window.
30569 (gdb) tui new-layout example src 1 regs 1 status 0 cmd 1
30572 Here, the new layout is called @samp{example}. It shows the source
30573 and register windows, followed by the status window, and then finally
30574 the command window. The non-status windows all have the same weight,
30575 so the terminal will be split into three roughly equal sections.
30577 Here is a more complex example, showing a horizontal layout:
30580 (gdb) tui new-layout example @{-horizontal src 1 asm 1@} 2 status 0 cmd 1
30583 This will result in side-by-side source and assembly windows; with the
30584 status and command window being beneath these, filling the entire
30585 width of the terminal. Because they have weight 2, the source and
30586 assembly windows will be twice the height of the command window.
30590 @item tui layout @var{name}
30591 @itemx layout @var{name}
30592 Changes which TUI windows are displayed. The @var{name} parameter
30593 controls which layout is shown. It can be either one of the built-in
30594 layout names, or the name of a layout defined by the user using
30595 @code{tui new-layout}.
30597 The built-in layouts are as follows:
30601 Display the next layout.
30604 Display the previous layout.
30607 Display the source and command windows.
30610 Display the assembly and command windows.
30613 Display the source, assembly, and command windows.
30616 When in @code{src} layout display the register, source, and command
30617 windows. When in @code{asm} or @code{split} layout display the
30618 register, assembler, and command windows.
30622 @item tui focus @var{name}
30623 @itemx focus @var{name}
30624 Changes which TUI window is currently active for scrolling. The
30625 @var{name} parameter can be any of the following:
30629 Make the next window active for scrolling.
30632 Make the previous window active for scrolling.
30635 Make the source window active for scrolling.
30638 Make the assembly window active for scrolling.
30641 Make the register window active for scrolling.
30644 Make the command window active for scrolling.
30647 @kindex tui refresh
30651 Refresh the screen. This is similar to typing @kbd{C-L}.
30653 @item tui reg @var{group}
30655 Changes the register group displayed in the tui register window to
30656 @var{group}. If the register window is not currently displayed this
30657 command will cause the register window to be displayed. The list of
30658 register groups, as well as their order is target specific. The
30659 following groups are available on most targets:
30662 Repeatedly selecting this group will cause the display to cycle
30663 through all of the available register groups.
30666 Repeatedly selecting this group will cause the display to cycle
30667 through all of the available register groups in the reverse order to
30671 Display the general registers.
30673 Display the floating point registers.
30675 Display the system registers.
30677 Display the vector registers.
30679 Display all registers.
30684 Update the source window and the current execution point.
30686 @kindex tui window height
30688 @item tui window height @var{name} +@var{count}
30689 @itemx tui window height @var{name} -@var{count}
30690 @itemx winheight @var{name} +@var{count}
30691 @itemx winheight @var{name} -@var{count}
30692 Change the height of the window @var{name} by @var{count} lines.
30693 Positive counts increase the height, while negative counts decrease
30694 it. The @var{name} parameter can be the name of any currently visible
30695 window. The names of the currently visible windows can be discovered
30696 using @kbd{info win} (@pxref{info_win_command,,info win}).
30698 The set of currently visible windows must always fill the terminal,
30699 and so, it is only possible to resize on window if there are other
30700 visible windows that can either give or receive the extra terminal
30703 @kindex tui window width
30705 @item tui window width @var{name} +@var{count}
30706 @itemx tui window width @var{name} -@var{count}
30707 @itemx winwidth @var{name} +@var{count}
30708 @itemx winwidth @var{name} -@var{count}
30709 Change the width of the window @var{name} by @var{count} columns.
30710 Positive counts increase the width, while negative counts decrease it.
30711 The @var{name} parameter can be the name of any currently visible
30712 window. The names of the currently visible windows can be discovered
30713 using @code{info win} (@pxref{info_win_command,,info win}).
30715 The set of currently visible windows must always fill the terminal,
30716 and so, it is only possible to resize on window if there are other
30717 visible windows that can either give or receive the extra terminal
30721 @node TUI Configuration
30722 @section TUI Configuration Variables
30723 @cindex TUI configuration variables
30725 Several configuration variables control the appearance of TUI windows.
30728 @item set tui border-kind @var{kind}
30729 @kindex set tui border-kind
30730 Select the border appearance for the source, assembly and register windows.
30731 The possible values are the following:
30734 Use a space character to draw the border.
30737 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
30740 Use the Alternate Character Set to draw the border. The border is
30741 drawn using character line graphics if the terminal supports them.
30744 @item set tui border-mode @var{mode}
30745 @kindex set tui border-mode
30746 @itemx set tui active-border-mode @var{mode}
30747 @kindex set tui active-border-mode
30748 Select the display attributes for the borders of the inactive windows
30749 or the active window. The @var{mode} can be one of the following:
30752 Use normal attributes to display the border.
30758 Use reverse video mode.
30761 Use half bright mode.
30763 @item half-standout
30764 Use half bright and standout mode.
30767 Use extra bright or bold mode.
30769 @item bold-standout
30770 Use extra bright or bold and standout mode.
30773 @item set tui tab-width @var{nchars}
30774 @kindex set tui tab-width
30776 Set the width of tab stops to be @var{nchars} characters. This
30777 setting affects the display of TAB characters in the source and
30780 @item set tui compact-source @r{[}on@r{|}off@r{]}
30781 @kindex set tui compact-source
30782 Set whether the TUI source window is displayed in ``compact'' form.
30783 The default display uses more space for line numbers; the compact
30784 display uses only as much space as is needed for the line numbers in
30787 @anchor{tui-mouse-events}
30788 @item set tui mouse-events @r{[}on@r{|}off@r{]}
30789 @kindex set tui mouse-events
30790 When on (default), mouse clicks control the TUI (@pxref{TUI Mouse Support}).
30791 When off, mouse clicks are handled by the terminal, enabling terminal-native
30794 @kindex set debug tui
30795 @item set debug tui @r{[}on|off@r{]}
30796 Turn on or off display of @value{GDBN} internal debug messages relating
30799 @kindex show debug tui
30800 @item show debug tui
30801 Show the current status of displaying @value{GDBN} internal debug
30802 messages relating to the TUI.
30806 Note that the colors of the TUI borders can be controlled using the
30807 appropriate @code{set style} commands. @xref{Output Styling}.
30810 @chapter Using @value{GDBN} under @sc{gnu} Emacs
30813 @cindex @sc{gnu} Emacs
30814 A special interface allows you to use @sc{gnu} Emacs to view (and
30815 edit) the source files for the program you are debugging with
30818 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
30819 executable file you want to debug as an argument. This command starts
30820 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
30821 created Emacs buffer.
30822 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
30824 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
30829 All ``terminal'' input and output goes through an Emacs buffer, called
30832 This applies both to @value{GDBN} commands and their output, and to the input
30833 and output done by the program you are debugging.
30835 This is useful because it means that you can copy the text of previous
30836 commands and input them again; you can even use parts of the output
30839 All the facilities of Emacs' Shell mode are available for interacting
30840 with your program. In particular, you can send signals the usual
30841 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
30845 @value{GDBN} displays source code through Emacs.
30847 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
30848 source file for that frame and puts an arrow (@samp{=>}) at the
30849 left margin of the current line. Emacs uses a separate buffer for
30850 source display, and splits the screen to show both your @value{GDBN} session
30853 Explicit @value{GDBN} @code{list} or search commands still produce output as
30854 usual, but you probably have no reason to use them from Emacs.
30857 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
30858 a graphical mode, enabled by default, which provides further buffers
30859 that can control the execution and describe the state of your program.
30860 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
30862 If you specify an absolute file name when prompted for the @kbd{M-x
30863 gdb} argument, then Emacs sets your current working directory to where
30864 your program resides. If you only specify the file name, then Emacs
30865 sets your current working directory to the directory associated
30866 with the previous buffer. In this case, @value{GDBN} may find your
30867 program by searching your environment's @env{PATH} variable, but on
30868 some operating systems it might not find the source. So, although the
30869 @value{GDBN} input and output session proceeds normally, the auxiliary
30870 buffer does not display the current source and line of execution.
30872 The initial working directory of @value{GDBN} is printed on the top
30873 line of the GUD buffer and this serves as a default for the commands
30874 that specify files for @value{GDBN} to operate on. @xref{Files,
30875 ,Commands to Specify Files}.
30877 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
30878 need to call @value{GDBN} by a different name (for example, if you
30879 keep several configurations around, with different names) you can
30880 customize the Emacs variable @code{gud-gdb-command-name} to run the
30883 In the GUD buffer, you can use these special Emacs commands in
30884 addition to the standard Shell mode commands:
30888 Describe the features of Emacs' GUD Mode.
30891 Execute to another source line, like the @value{GDBN} @code{step} command; also
30892 update the display window to show the current file and location.
30895 Execute to next source line in this function, skipping all function
30896 calls, like the @value{GDBN} @code{next} command. Then update the display window
30897 to show the current file and location.
30900 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
30901 display window accordingly.
30904 Execute until exit from the selected stack frame, like the @value{GDBN}
30905 @code{finish} command.
30908 Continue execution of your program, like the @value{GDBN} @code{continue}
30912 Go up the number of frames indicated by the numeric argument
30913 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
30914 like the @value{GDBN} @code{up} command.
30917 Go down the number of frames indicated by the numeric argument, like the
30918 @value{GDBN} @code{down} command.
30921 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
30922 tells @value{GDBN} to set a breakpoint on the source line point is on.
30924 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
30925 separate frame which shows a backtrace when the GUD buffer is current.
30926 Move point to any frame in the stack and type @key{RET} to make it
30927 become the current frame and display the associated source in the
30928 source buffer. Alternatively, click @kbd{Mouse-2} to make the
30929 selected frame become the current one. In graphical mode, the
30930 speedbar displays watch expressions.
30932 If you accidentally delete the source-display buffer, an easy way to get
30933 it back is to type the command @code{f} in the @value{GDBN} buffer, to
30934 request a frame display; when you run under Emacs, this recreates
30935 the source buffer if necessary to show you the context of the current
30938 The source files displayed in Emacs are in ordinary Emacs buffers
30939 which are visiting the source files in the usual way. You can edit
30940 the files with these buffers if you wish; but keep in mind that @value{GDBN}
30941 communicates with Emacs in terms of line numbers. If you add or
30942 delete lines from the text, the line numbers that @value{GDBN} knows cease
30943 to correspond properly with the code.
30945 A more detailed description of Emacs' interaction with @value{GDBN} is
30946 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
30950 @chapter The @sc{gdb/mi} Interface
30952 @unnumberedsec Function and Purpose
30954 @cindex @sc{gdb/mi}, its purpose
30955 @sc{gdb/mi} is a line based machine oriented text interface to
30956 @value{GDBN} and is activated by specifying using the
30957 @option{--interpreter} command line option (@pxref{Mode Options}). It
30958 is specifically intended to support the development of systems which
30959 use the debugger as just one small component of a larger system.
30961 This chapter is a specification of the @sc{gdb/mi} interface. It is written
30962 in the form of a reference manual.
30964 Note that @sc{gdb/mi} is still under construction, so some of the
30965 features described below are incomplete and subject to change
30966 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
30968 @unnumberedsec Notation and Terminology
30970 @cindex notational conventions, for @sc{gdb/mi}
30971 This chapter uses the following notation:
30975 @code{|} separates two alternatives.
30978 @code{[ @var{something} ]} indicates that @var{something} is optional:
30979 it may or may not be given.
30982 @code{( @var{group} )*} means that @var{group} inside the parentheses
30983 may repeat zero or more times.
30986 @code{( @var{group} )+} means that @var{group} inside the parentheses
30987 may repeat one or more times.
30990 @code{( @var{group} )} means that @var{group} inside the parentheses
30991 occurs exactly once.
30994 @code{"@var{string}"} means a literal @var{string}.
30998 @heading Dependencies
31002 * GDB/MI General Design::
31003 * GDB/MI Command Syntax::
31004 * GDB/MI Compatibility with CLI::
31005 * GDB/MI Development and Front Ends::
31006 * GDB/MI Output Records::
31007 * GDB/MI Simple Examples::
31008 * GDB/MI Command Description Format::
31009 * GDB/MI Breakpoint Commands::
31010 * GDB/MI Catchpoint Commands::
31011 * GDB/MI Program Context::
31012 * GDB/MI Thread Commands::
31013 * GDB/MI Ada Tasking Commands::
31014 * GDB/MI Program Execution::
31015 * GDB/MI Stack Manipulation::
31016 * GDB/MI Variable Objects::
31017 * GDB/MI Data Manipulation::
31018 * GDB/MI Tracepoint Commands::
31019 * GDB/MI Symbol Query::
31020 * GDB/MI File Commands::
31022 * GDB/MI Kod Commands::
31023 * GDB/MI Memory Overlay Commands::
31024 * GDB/MI Signal Handling Commands::
31026 * GDB/MI Target Manipulation::
31027 * GDB/MI File Transfer Commands::
31028 * GDB/MI Ada Exceptions Commands::
31029 * GDB/MI Support Commands::
31030 * GDB/MI Miscellaneous Commands::
31033 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31034 @node GDB/MI General Design
31035 @section @sc{gdb/mi} General Design
31036 @cindex GDB/MI General Design
31038 Interaction of a @sc{gdb/mi} frontend with @value{GDBN} involves three
31039 parts---commands sent to @value{GDBN}, responses to those commands
31040 and notifications. Each command results in exactly one response,
31041 indicating either successful completion of the command, or an error.
31042 For the commands that do not resume the target, the response contains the
31043 requested information. For the commands that resume the target, the
31044 response only indicates whether the target was successfully resumed.
31045 Notifications is the mechanism for reporting changes in the state of the
31046 target, or in @value{GDBN} state, that cannot conveniently be associated with
31047 a command and reported as part of that command response.
31049 The important examples of notifications are:
31053 Exec notifications. These are used to report changes in
31054 target state---when a target is resumed, or stopped. It would not
31055 be feasible to include this information in response of resuming
31056 commands, because one resume commands can result in multiple events in
31057 different threads. Also, quite some time may pass before any event
31058 happens in the target, while a frontend needs to know whether the resuming
31059 command itself was successfully executed.
31062 Console output, and status notifications. Console output
31063 notifications are used to report output of CLI commands, as well as
31064 diagnostics for other commands. Status notifications are used to
31065 report the progress of a long-running operation. Naturally, including
31066 this information in command response would mean no output is produced
31067 until the command is finished, which is undesirable.
31070 General notifications. Commands may have various side effects on
31071 the @value{GDBN} or target state beyond their official purpose. For example,
31072 a command may change the selected thread. Although such changes can
31073 be included in command response, using notification allows for more
31074 orthogonal frontend design.
31078 There's no guarantee that whenever an MI command reports an error,
31079 @value{GDBN} or the target are in any specific state, and especially,
31080 the state is not reverted to the state before the MI command was
31081 processed. Therefore, whenever an MI command results in an error,
31082 we recommend that the frontend refreshes all the information shown in
31083 the user interface.
31087 * Context management::
31088 * Asynchronous and non-stop modes::
31092 @node Context management
31093 @subsection Context management
31095 @subsubsection Threads and Frames
31097 In most cases when @value{GDBN} accesses the target, this access is
31098 done in context of a specific thread and frame (@pxref{Frames}).
31099 Often, even when accessing global data, the target requires that a thread
31100 be specified. The CLI interface maintains the selected thread and frame,
31101 and supplies them to target on each command. This is convenient,
31102 because a command line user would not want to specify that information
31103 explicitly on each command, and because user interacts with
31104 @value{GDBN} via a single terminal, so no confusion is possible as
31105 to what thread and frame are the current ones.
31107 In the case of MI, the concept of selected thread and frame is less
31108 useful. First, a frontend can easily remember this information
31109 itself. Second, a graphical frontend can have more than one window,
31110 each one used for debugging a different thread, and the frontend might
31111 want to access additional threads for internal purposes. This
31112 increases the risk that by relying on implicitly selected thread, the
31113 frontend may be operating on a wrong one. Therefore, each MI command
31114 should explicitly specify which thread and frame to operate on. To
31115 make it possible, each MI command accepts the @samp{--thread} and
31116 @samp{--frame} options, the value to each is @value{GDBN} global
31117 identifier for thread and frame to operate on.
31119 Usually, each top-level window in a frontend allows the user to select
31120 a thread and a frame, and remembers the user selection for further
31121 operations. However, in some cases @value{GDBN} may suggest that the
31122 current thread or frame be changed. For example, when stopping on a
31123 breakpoint it is reasonable to switch to the thread where breakpoint is
31124 hit. For another example, if the user issues the CLI @samp{thread} or
31125 @samp{frame} commands via the frontend, it is desirable to change the
31126 frontend's selection to the one specified by user. @value{GDBN}
31127 communicates the suggestion to change current thread and frame using the
31128 @samp{=thread-selected} notification.
31130 Note that historically, MI shares the selected thread with CLI, so
31131 frontends used the @code{-thread-select} to execute commands in the
31132 right context. However, getting this to work right is cumbersome. The
31133 simplest way is for frontend to emit @code{-thread-select} command
31134 before every command. This doubles the number of commands that need
31135 to be sent. The alternative approach is to suppress @code{-thread-select}
31136 if the selected thread in @value{GDBN} is supposed to be identical to the
31137 thread the frontend wants to operate on. However, getting this
31138 optimization right can be tricky. In particular, if the frontend
31139 sends several commands to @value{GDBN}, and one of the commands changes the
31140 selected thread, then the behaviour of subsequent commands will
31141 change. So, a frontend should either wait for response from such
31142 problematic commands, or explicitly add @code{-thread-select} for
31143 all subsequent commands. No frontend is known to do this exactly
31144 right, so it is suggested to just always pass the @samp{--thread} and
31145 @samp{--frame} options.
31147 @subsubsection Language
31149 The execution of several commands depends on which language is selected.
31150 By default, the current language (@pxref{show language}) is used.
31151 But for commands known to be language-sensitive, it is recommended
31152 to use the @samp{--language} option. This option takes one argument,
31153 which is the name of the language to use while executing the command.
31157 -data-evaluate-expression --language c "sizeof (void*)"
31162 The valid language names are the same names accepted by the
31163 @samp{set language} command (@pxref{Manually}), excluding @samp{auto},
31164 @samp{local} or @samp{unknown}.
31166 @node Asynchronous and non-stop modes
31167 @subsection Asynchronous command execution and non-stop mode
31169 On some targets, @value{GDBN} is capable of processing MI commands
31170 even while the target is running. This is called @dfn{asynchronous
31171 command execution} (@pxref{Background Execution}). The frontend may
31172 specify a preference for asynchronous execution using the
31173 @code{-gdb-set mi-async 1} command, which should be emitted before
31174 either running the executable or attaching to the target. After the
31175 frontend has started the executable or attached to the target, it can
31176 find if asynchronous execution is enabled using the
31177 @code{-list-target-features} command.
31180 @cindex foreground execution
31181 @cindex background execution
31182 @cindex asynchronous execution
31183 @cindex execution, foreground, background and asynchronous
31184 @kindex set mi-async
31185 @item -gdb-set mi-async @r{[}on@r{|}off@r{]}
31186 Set whether MI is in asynchronous mode.
31188 When @code{off}, which is the default, MI execution commands (e.g.,
31189 @code{-exec-continue}) are foreground commands, and @value{GDBN} waits
31190 for the program to stop before processing further commands.
31192 When @code{on}, MI execution commands are background execution
31193 commands (e.g., @code{-exec-continue} becomes the equivalent of the
31194 @code{c&} CLI command), and so @value{GDBN} is capable of processing
31195 MI commands even while the target is running.
31197 @kindex show mi-async
31198 @item -gdb-show mi-async
31199 Show whether MI asynchronous mode is enabled.
31202 Note: In @value{GDBN} version 7.7 and earlier, this option was called
31203 @code{target-async} instead of @code{mi-async}, and it had the effect
31204 of both putting MI in asynchronous mode and making CLI background
31205 commands possible. CLI background commands are now always possible
31206 ``out of the box'' if the target supports them. The old spelling is
31207 kept as a deprecated alias for backwards compatibility.
31209 Even if @value{GDBN} can accept a command while target is running,
31210 many commands that access the target do not work when the target is
31211 running. Therefore, asynchronous command execution is most useful
31212 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
31213 it is possible to examine the state of one thread, while other threads
31216 When a given thread is running, MI commands that try to access the
31217 target in the context of that thread may not work, or may work only on
31218 some targets. In particular, commands that try to operate on thread's
31219 stack will not work, on any target. Commands that read memory, or
31220 modify breakpoints, may work or not work, depending on the target. Note
31221 that even commands that operate on global state, such as @code{print},
31222 @code{set}, and breakpoint commands, still access the target in the
31223 context of a specific thread, so frontend should try to find a
31224 stopped thread and perform the operation on that thread (using the
31225 @samp{--thread} option).
31227 Which commands will work in the context of a running thread is
31228 highly target dependent. However, the two commands
31229 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
31230 to find the state of a thread, will always work.
31232 @node Thread groups
31233 @subsection Thread groups
31234 @value{GDBN} may be used to debug several processes at the same time.
31235 On some platforms, @value{GDBN} may support debugging of several
31236 hardware systems, each one having several cores with several different
31237 processes running on each core. This section describes the MI
31238 mechanism to support such debugging scenarios.
31240 The key observation is that regardless of the structure of the
31241 target, MI can have a global list of threads, because most commands that
31242 accept the @samp{--thread} option do not need to know what process that
31243 thread belongs to. Therefore, it is not necessary to introduce
31244 neither additional @samp{--process} option, nor an notion of the
31245 current process in the MI interface. The only strictly new feature
31246 that is required is the ability to find how the threads are grouped
31249 To allow the user to discover such grouping, and to support arbitrary
31250 hierarchy of machines/cores/processes, MI introduces the concept of a
31251 @dfn{thread group}. Thread group is a collection of threads and other
31252 thread groups. A thread group always has a string identifier, a type,
31253 and may have additional attributes specific to the type. A new
31254 command, @code{-list-thread-groups}, returns the list of top-level
31255 thread groups, which correspond to processes that @value{GDBN} is
31256 debugging at the moment. By passing an identifier of a thread group
31257 to the @code{-list-thread-groups} command, it is possible to obtain
31258 the members of specific thread group.
31260 To allow the user to easily discover processes, and other objects, he
31261 wishes to debug, a concept of @dfn{available thread group} is
31262 introduced. Available thread group is an thread group that
31263 @value{GDBN} is not debugging, but that can be attached to, using the
31264 @code{-target-attach} command. The list of available top-level thread
31265 groups can be obtained using @samp{-list-thread-groups --available}.
31266 In general, the content of a thread group may be only retrieved only
31267 after attaching to that thread group.
31269 Thread groups are related to inferiors (@pxref{Inferiors Connections and
31270 Programs}). Each inferior corresponds to a thread group of a special
31271 type @samp{process}, and some additional operations are permitted on
31272 such thread groups.
31274 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31275 @node GDB/MI Command Syntax
31276 @section @sc{gdb/mi} Command Syntax
31279 * GDB/MI Input Syntax::
31280 * GDB/MI Output Syntax::
31283 @node GDB/MI Input Syntax
31284 @subsection @sc{gdb/mi} Input Syntax
31286 @cindex input syntax for @sc{gdb/mi}
31287 @cindex @sc{gdb/mi}, input syntax
31289 @item @var{command} @expansion{}
31290 @code{@var{cli-command} | @var{mi-command}}
31292 @item @var{cli-command} @expansion{}
31293 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
31294 @var{cli-command} is any existing @value{GDBN} CLI command.
31296 @item @var{mi-command} @expansion{}
31297 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
31298 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
31300 @item @var{token} @expansion{}
31301 "any sequence of digits"
31303 @item @var{option} @expansion{}
31304 @code{"-" @var{parameter} [ " " @var{parameter} ]}
31306 @item @var{parameter} @expansion{}
31307 @code{@var{non-blank-sequence} | @var{c-string}}
31309 @item @var{operation} @expansion{}
31310 @emph{any of the operations described in this chapter}
31312 @item @var{non-blank-sequence} @expansion{}
31313 @emph{anything, provided it doesn't contain special characters such as
31314 "-", @var{nl}, """ and of course " "}
31316 @item @var{c-string} @expansion{}
31317 @code{""" @var{seven-bit-iso-c-string-content} """}
31319 @item @var{nl} @expansion{}
31328 The CLI commands are still handled by the @sc{mi} interpreter; their
31329 output is described below.
31332 The @code{@var{token}}, when present, is passed back when the command
31336 Some @sc{mi} commands accept optional arguments as part of the parameter
31337 list. Each option is identified by a leading @samp{-} (dash) and may be
31338 followed by an optional argument parameter. Options occur first in the
31339 parameter list and can be delimited from normal parameters using
31340 @samp{--} (this is useful when some parameters begin with a dash).
31347 We want easy access to the existing CLI syntax (for debugging).
31350 We want it to be easy to spot a @sc{mi} operation.
31353 @node GDB/MI Output Syntax
31354 @subsection @sc{gdb/mi} Output Syntax
31356 @cindex output syntax of @sc{gdb/mi}
31357 @cindex @sc{gdb/mi}, output syntax
31358 The output from @sc{gdb/mi} consists of zero or more out-of-band records
31359 followed, optionally, by a single result record. This result record
31360 is for the most recent command. The sequence of output records is
31361 terminated by @samp{(gdb)}.
31363 If an input command was prefixed with a @code{@var{token}} then the
31364 corresponding output for that command will also be prefixed by that same
31368 @item @var{output} @expansion{}
31369 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
31371 @item @var{result-record} @expansion{}
31372 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
31374 @item @var{out-of-band-record} @expansion{}
31375 @code{@var{async-record} | @var{stream-record}}
31377 @item @var{async-record} @expansion{}
31378 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
31380 @item @var{exec-async-output} @expansion{}
31381 @code{[ @var{token} ] "*" @var{async-output nl}}
31383 @item @var{status-async-output} @expansion{}
31384 @code{[ @var{token} ] "+" @var{async-output nl}}
31386 @item @var{notify-async-output} @expansion{}
31387 @code{[ @var{token} ] "=" @var{async-output nl}}
31389 @item @var{async-output} @expansion{}
31390 @code{@var{async-class} ( "," @var{result} )*}
31392 @item @var{result-class} @expansion{}
31393 @code{"done" | "running" | "connected" | "error" | "exit"}
31395 @item @var{async-class} @expansion{}
31396 @code{"stopped" | @var{others}} (where @var{others} will be added
31397 depending on the needs---this is still in development).
31399 @item @var{result} @expansion{}
31400 @code{ @var{variable} "=" @var{value}}
31402 @item @var{variable} @expansion{}
31403 @code{ @var{string} }
31405 @item @var{value} @expansion{}
31406 @code{ @var{const} | @var{tuple} | @var{list} }
31408 @item @var{const} @expansion{}
31409 @code{@var{c-string}}
31411 @item @var{tuple} @expansion{}
31412 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
31414 @item @var{list} @expansion{}
31415 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
31416 @var{result} ( "," @var{result} )* "]" }
31418 @item @var{stream-record} @expansion{}
31419 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
31421 @item @var{console-stream-output} @expansion{}
31422 @code{"~" @var{c-string nl}}
31424 @item @var{target-stream-output} @expansion{}
31425 @code{"@@" @var{c-string nl}}
31427 @item @var{log-stream-output} @expansion{}
31428 @code{"&" @var{c-string nl}}
31430 @item @var{nl} @expansion{}
31433 @item @var{token} @expansion{}
31434 @emph{any sequence of digits}.
31442 All output sequences end in a single line containing a period.
31445 The @code{@var{token}} is from the corresponding request. Note that
31446 for all async output, while the token is allowed by the grammar and
31447 may be output by future versions of @value{GDBN} for select async
31448 output messages, it is generally omitted. Frontends should treat
31449 all async output as reporting general changes in the state of the
31450 target and there should be no need to associate async output to any
31454 @cindex status output in @sc{gdb/mi}
31455 @var{status-async-output} contains on-going status information about the
31456 progress of a slow operation. It can be discarded. All status output is
31457 prefixed by @samp{+}.
31460 @cindex async output in @sc{gdb/mi}
31461 @var{exec-async-output} contains asynchronous state change on the target
31462 (stopped, started, disappeared). All async output is prefixed by
31466 @cindex notify output in @sc{gdb/mi}
31467 @var{notify-async-output} contains supplementary information that the
31468 client should handle (e.g., a new breakpoint information). All notify
31469 output is prefixed by @samp{=}.
31472 @cindex console output in @sc{gdb/mi}
31473 @var{console-stream-output} is output that should be displayed as is in the
31474 console. It is the textual response to a CLI command. All the console
31475 output is prefixed by @samp{~}.
31478 @cindex target output in @sc{gdb/mi}
31479 @var{target-stream-output} is the output produced by the target program.
31480 All the target output is prefixed by @samp{@@}.
31483 @cindex log output in @sc{gdb/mi}
31484 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
31485 instance messages that should be displayed as part of an error log. All
31486 the log output is prefixed by @samp{&}.
31489 @cindex list output in @sc{gdb/mi}
31490 New @sc{gdb/mi} commands should only output @var{lists} containing
31496 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
31497 details about the various output records.
31499 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31500 @node GDB/MI Compatibility with CLI
31501 @section @sc{gdb/mi} Compatibility with CLI
31503 @cindex compatibility, @sc{gdb/mi} and CLI
31504 @cindex @sc{gdb/mi}, compatibility with CLI
31506 For the developers convenience CLI commands can be entered directly,
31507 but there may be some unexpected behaviour. For example, commands
31508 that query the user will behave as if the user replied yes, breakpoint
31509 command lists are not executed and some CLI commands, such as
31510 @code{if}, @code{when} and @code{define}, prompt for further input with
31511 @samp{>}, which is not valid MI output.
31513 This feature may be removed at some stage in the future and it is
31514 recommended that front ends use the @code{-interpreter-exec} command
31515 (@pxref{-interpreter-exec}).
31517 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31518 @node GDB/MI Development and Front Ends
31519 @section @sc{gdb/mi} Development and Front Ends
31520 @cindex @sc{gdb/mi} development
31522 The application which takes the MI output and presents the state of the
31523 program being debugged to the user is called a @dfn{front end}.
31525 Since @sc{gdb/mi} is used by a variety of front ends to @value{GDBN}, changes
31526 to the MI interface may break existing usage. This section describes how the
31527 protocol changes and how to request previous version of the protocol when it
31530 Some changes in MI need not break a carefully designed front end, and
31531 for these the MI version will remain unchanged. The following is a
31532 list of changes that may occur within one level, so front ends should
31533 parse MI output in a way that can handle them:
31537 New MI commands may be added.
31540 New fields may be added to the output of any MI command.
31543 The range of values for fields with specified values, e.g.,
31544 @code{in_scope} (@pxref{-var-update}) may be extended.
31546 @c The format of field's content e.g type prefix, may change so parse it
31547 @c at your own risk. Yes, in general?
31549 @c The order of fields may change? Shouldn't really matter but it might
31550 @c resolve inconsistencies.
31553 If the changes are likely to break front ends, the MI version level
31554 will be increased by one. The new versions of the MI protocol are not compatible
31555 with the old versions. Old versions of MI remain available, allowing front ends
31556 to keep using them until they are modified to use the latest MI version.
31558 Since @code{--interpreter=mi} always points to the latest MI version, it is
31559 recommended that front ends request a specific version of MI when launching
31560 @value{GDBN} (e.g.@: @code{--interpreter=mi2}) to make sure they get an
31561 interpreter with the MI version they expect.
31563 The following table gives a summary of the released versions of the MI
31564 interface: the version number, the version of GDB in which it first appeared
31565 and the breaking changes compared to the previous version.
31567 @multitable @columnfractions .1 .1 .8
31568 @headitem MI version @tab GDB version @tab Breaking changes
31585 The @code{-environment-pwd}, @code{-environment-directory} and
31586 @code{-environment-path} commands now returns values using the MI output
31587 syntax, rather than CLI output syntax.
31590 @code{-var-list-children}'s @code{children} result field is now a list, rather
31594 @code{-var-update}'s @code{changelist} result field is now a list, rather than
31606 The output of information about multi-location breakpoints has changed in the
31607 responses to the @code{-break-insert} and @code{-break-info} commands, as well
31608 as in the @code{=breakpoint-created} and @code{=breakpoint-modified} events.
31609 The multiple locations are now placed in a @code{locations} field, whose value
31621 The syntax of the "script" field in breakpoint output has changed in the
31622 responses to the @code{-break-insert} and @code{-break-info} commands, as
31623 well as the @code{=breakpoint-created} and @code{=breakpoint-modified}
31624 events. The previous output was syntactically invalid. The new output is
31630 If your front end cannot yet migrate to a more recent version of the
31631 MI protocol, you can nevertheless selectively enable specific features
31632 available in those recent MI versions, using the following commands:
31636 @item -fix-multi-location-breakpoint-output
31637 Use the output for multi-location breakpoints which was introduced by
31638 MI 3, even when using MI versions below 3. This command has no
31639 effect when using MI version 3 or later.
31641 @item -fix-breakpoint-script-output
31642 Use the output for the breakpoint "script" field which was introduced by
31643 MI 4, even when using MI versions below 4. This command has no effect when
31644 using MI version 4 or later.
31648 The best way to avoid unexpected changes in MI that might break your front
31649 end is to make your project known to @value{GDBN} developers and
31650 follow development on @email{gdb@@sourceware.org} and
31651 @email{gdb-patches@@sourceware.org}.
31652 @cindex mailing lists
31654 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31655 @node GDB/MI Output Records
31656 @section @sc{gdb/mi} Output Records
31659 * GDB/MI Result Records::
31660 * GDB/MI Stream Records::
31661 * GDB/MI Async Records::
31662 * GDB/MI Breakpoint Information::
31663 * GDB/MI Frame Information::
31664 * GDB/MI Thread Information::
31665 * GDB/MI Ada Exception Information::
31668 @node GDB/MI Result Records
31669 @subsection @sc{gdb/mi} Result Records
31671 @cindex result records in @sc{gdb/mi}
31672 @cindex @sc{gdb/mi}, result records
31673 In addition to a number of out-of-band notifications, the response to a
31674 @sc{gdb/mi} command includes one of the following result indications:
31678 @item "^done" [ "," @var{results} ]
31679 The synchronous operation was successful, @code{@var{results}} are the return
31684 This result record is equivalent to @samp{^done}. Historically, it
31685 was output instead of @samp{^done} if the command has resumed the
31686 target. This behaviour is maintained for backward compatibility, but
31687 all frontends should treat @samp{^done} and @samp{^running}
31688 identically and rely on the @samp{*running} output record to determine
31689 which threads are resumed.
31693 @value{GDBN} has connected to a remote target.
31696 @item "^error" "," "msg=" @var{c-string} [ "," "code=" @var{c-string} ]
31697 The operation failed. The @code{msg=@var{c-string}} variable contains
31698 the corresponding error message.
31700 If present, the @code{code=@var{c-string}} variable provides an error
31701 code on which consumers can rely on to detect the corresponding
31702 error condition. At present, only one error code is defined:
31705 @item "undefined-command"
31706 Indicates that the command causing the error does not exist.
31711 @value{GDBN} has terminated.
31715 @node GDB/MI Stream Records
31716 @subsection @sc{gdb/mi} Stream Records
31718 @cindex @sc{gdb/mi}, stream records
31719 @cindex stream records in @sc{gdb/mi}
31720 @value{GDBN} internally maintains a number of output streams: the console, the
31721 target, and the log. The output intended for each of these streams is
31722 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
31724 Each stream record begins with a unique @dfn{prefix character} which
31725 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
31726 Syntax}). In addition to the prefix, each stream record contains a
31727 @code{@var{string-output}}. This is either raw text (with an implicit new
31728 line) or a quoted C string (which does not contain an implicit newline).
31731 @item "~" @var{string-output}
31732 The console output stream contains text that should be displayed in the
31733 CLI console window. It contains the textual responses to CLI commands.
31735 @item "@@" @var{string-output}
31736 The target output stream contains any textual output from the running
31737 target. This is only present when GDB's event loop is truly
31738 asynchronous, which is currently only the case for remote targets.
31740 @item "&" @var{string-output}
31741 The log stream contains debugging messages being produced by @value{GDBN}'s
31745 @node GDB/MI Async Records
31746 @subsection @sc{gdb/mi} Async Records
31748 @cindex async records in @sc{gdb/mi}
31749 @cindex @sc{gdb/mi}, async records
31750 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
31751 additional changes that have occurred. Those changes can either be a
31752 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
31753 target activity (e.g., target stopped).
31755 The following is the list of possible async records:
31759 @item *running,thread-id="@var{thread}"
31760 The target is now running. The @var{thread} field can be the global
31761 thread ID of the thread that is now running, and it can be
31762 @samp{all} if all threads are running. The frontend should assume
31763 that no interaction with a running thread is possible after this
31764 notification is produced. The frontend should not assume that this
31765 notification is output only once for any command. @value{GDBN} may
31766 emit this notification several times, either for different threads,
31767 because it cannot resume all threads together, or even for a single
31768 thread, if the thread must be stepped though some code before letting
31771 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
31772 The target has stopped. The @var{reason} field can have one of the
31776 @item breakpoint-hit
31777 A breakpoint was reached.
31778 @item watchpoint-trigger
31779 A watchpoint was triggered.
31780 @item read-watchpoint-trigger
31781 A read watchpoint was triggered.
31782 @item access-watchpoint-trigger
31783 An access watchpoint was triggered.
31784 @item function-finished
31785 An -exec-finish or similar CLI command was accomplished.
31786 @item location-reached
31787 An -exec-until or similar CLI command was accomplished.
31788 @item watchpoint-scope
31789 A watchpoint has gone out of scope.
31790 @item end-stepping-range
31791 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
31792 similar CLI command was accomplished.
31793 @item exited-signalled
31794 The inferior exited because of a signal.
31796 The inferior exited.
31797 @item exited-normally
31798 The inferior exited normally.
31799 @item signal-received
31800 A signal was received by the inferior.
31802 The inferior has stopped due to a library being loaded or unloaded.
31803 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
31804 set or when a @code{catch load} or @code{catch unload} catchpoint is
31805 in use (@pxref{Set Catchpoints}).
31807 The inferior has forked. This is reported when @code{catch fork}
31808 (@pxref{Set Catchpoints}) has been used.
31810 The inferior has vforked. This is reported in when @code{catch vfork}
31811 (@pxref{Set Catchpoints}) has been used.
31812 @item syscall-entry
31813 The inferior entered a system call. This is reported when @code{catch
31814 syscall} (@pxref{Set Catchpoints}) has been used.
31815 @item syscall-return
31816 The inferior returned from a system call. This is reported when
31817 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
31819 The inferior called @code{exec}. This is reported when @code{catch exec}
31820 (@pxref{Set Catchpoints}) has been used.
31822 There isn't enough history recorded to continue reverse execution.
31825 The @var{id} field identifies the global thread ID of the thread
31826 that directly caused the stop -- for example by hitting a breakpoint.
31827 Depending on whether all-stop
31828 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
31829 stop all threads, or only the thread that directly triggered the stop.
31830 If all threads are stopped, the @var{stopped} field will have the
31831 value of @code{"all"}. Otherwise, the value of the @var{stopped}
31832 field will be a list of thread identifiers. Presently, this list will
31833 always include a single thread, but frontend should be prepared to see
31834 several threads in the list. The @var{core} field reports the
31835 processor core on which the stop event has happened. This field may be absent
31836 if such information is not available.
31838 @item =thread-group-added,id="@var{id}"
31839 @itemx =thread-group-removed,id="@var{id}"
31840 A thread group was either added or removed. The @var{id} field
31841 contains the @value{GDBN} identifier of the thread group. When a thread
31842 group is added, it generally might not be associated with a running
31843 process. When a thread group is removed, its id becomes invalid and
31844 cannot be used in any way.
31846 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
31847 A thread group became associated with a running program,
31848 either because the program was just started or the thread group
31849 was attached to a program. The @var{id} field contains the
31850 @value{GDBN} identifier of the thread group. The @var{pid} field
31851 contains process identifier, specific to the operating system.
31853 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
31854 A thread group is no longer associated with a running program,
31855 either because the program has exited, or because it was detached
31856 from. The @var{id} field contains the @value{GDBN} identifier of the
31857 thread group. The @var{code} field is the exit code of the inferior; it exists
31858 only when the inferior exited with some code.
31860 @item =thread-created,id="@var{id}",group-id="@var{gid}"
31861 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
31862 A thread either was created, or has exited. The @var{id} field
31863 contains the global @value{GDBN} identifier of the thread. The @var{gid}
31864 field identifies the thread group this thread belongs to.
31866 @item =thread-selected,id="@var{id}"[,frame="@var{frame}"]
31867 Informs that the selected thread or frame were changed. This notification
31868 is not emitted as result of the @code{-thread-select} or
31869 @code{-stack-select-frame} commands, but is emitted whenever an MI command
31870 that is not documented to change the selected thread and frame actually
31871 changes them. In particular, invoking, directly or indirectly
31872 (via user-defined command), the CLI @code{thread} or @code{frame} commands,
31873 will generate this notification. Changing the thread or frame from another
31874 user interface (see @ref{Interpreters}) will also generate this notification.
31876 The @var{frame} field is only present if the newly selected thread is
31877 stopped. See @ref{GDB/MI Frame Information} for the format of its value.
31879 We suggest that in response to this notification, front ends
31880 highlight the selected thread and cause subsequent commands to apply to
31883 @item =library-loaded,...
31884 Reports that a new library file was loaded by the program. This
31885 notification has 5 fields---@var{id}, @var{target-name},
31886 @var{host-name}, @var{symbols-loaded} and @var{ranges}. The @var{id} field is an
31887 opaque identifier of the library. For remote debugging case,
31888 @var{target-name} and @var{host-name} fields give the name of the
31889 library file on the target, and on the host respectively. For native
31890 debugging, both those fields have the same value. The
31891 @var{symbols-loaded} field is emitted only for backward compatibility
31892 and should not be relied on to convey any useful information. The
31893 @var{thread-group} field, if present, specifies the id of the thread
31894 group in whose context the library was loaded. If the field is
31895 absent, it means the library was loaded in the context of all present
31896 thread groups. The @var{ranges} field specifies the ranges of addresses belonging
31899 @item =library-unloaded,...
31900 Reports that a library was unloaded by the program. This notification
31901 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
31902 the same meaning as for the @code{=library-loaded} notification.
31903 The @var{thread-group} field, if present, specifies the id of the
31904 thread group in whose context the library was unloaded. If the field is
31905 absent, it means the library was unloaded in the context of all present
31908 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
31909 @itemx =traceframe-changed,end
31910 Reports that the trace frame was changed and its new number is
31911 @var{tfnum}. The number of the tracepoint associated with this trace
31912 frame is @var{tpnum}.
31914 @item =tsv-created,name=@var{name},initial=@var{initial}
31915 Reports that the new trace state variable @var{name} is created with
31916 initial value @var{initial}.
31918 @item =tsv-deleted,name=@var{name}
31919 @itemx =tsv-deleted
31920 Reports that the trace state variable @var{name} is deleted or all
31921 trace state variables are deleted.
31923 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
31924 Reports that the trace state variable @var{name} is modified with
31925 the initial value @var{initial}. The current value @var{current} of
31926 trace state variable is optional and is reported if the current
31927 value of trace state variable is known.
31929 @item =breakpoint-created,bkpt=@{...@}
31930 @itemx =breakpoint-modified,bkpt=@{...@}
31931 @itemx =breakpoint-deleted,id=@var{number}
31932 Reports that a breakpoint was created, modified, or deleted,
31933 respectively. Only user-visible breakpoints are reported to the MI
31936 The @var{bkpt} argument is of the same form as returned by the various
31937 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
31938 @var{number} is the ordinal number of the breakpoint.
31940 Note that if a breakpoint is emitted in the result record of a
31941 command, then it will not also be emitted in an async record.
31943 @item =record-started,thread-group="@var{id}",method="@var{method}"[,format="@var{format}"]
31944 @itemx =record-stopped,thread-group="@var{id}"
31945 Execution log recording was either started or stopped on an
31946 inferior. The @var{id} is the @value{GDBN} identifier of the thread
31947 group corresponding to the affected inferior.
31949 The @var{method} field indicates the method used to record execution. If the
31950 method in use supports multiple recording formats, @var{format} will be present
31951 and contain the currently used format. @xref{Process Record and Replay},
31952 for existing method and format values.
31954 @item =cmd-param-changed,param=@var{param},value=@var{value}
31955 Reports that a parameter of the command @code{set @var{param}} is
31956 changed to @var{value}. In the multi-word @code{set} command,
31957 the @var{param} is the whole parameter list to @code{set} command.
31958 For example, In command @code{set check type on}, @var{param}
31959 is @code{check type} and @var{value} is @code{on}.
31961 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
31962 Reports that bytes from @var{addr} to @var{data} + @var{len} were
31963 written in an inferior. The @var{id} is the identifier of the
31964 thread group corresponding to the affected inferior. The optional
31965 @code{type="code"} part is reported if the memory written to holds
31969 @node GDB/MI Breakpoint Information
31970 @subsection @sc{gdb/mi} Breakpoint Information
31972 When @value{GDBN} reports information about a breakpoint, a
31973 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
31978 The breakpoint number.
31981 The type of the breakpoint. For ordinary breakpoints this will be
31982 @samp{breakpoint}, but many values are possible.
31985 If the type of the breakpoint is @samp{catchpoint}, then this
31986 indicates the exact type of catchpoint.
31989 This is the breakpoint disposition---either @samp{del}, meaning that
31990 the breakpoint will be deleted at the next stop, or @samp{keep},
31991 meaning that the breakpoint will not be deleted.
31994 This indicates whether the breakpoint is enabled, in which case the
31995 value is @samp{y}, or disabled, in which case the value is @samp{n}.
31996 Note that this is not the same as the field @code{enable}.
31999 The address of the breakpoint. This may be a hexidecimal number,
32000 giving the address; or the string @samp{<PENDING>}, for a pending
32001 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
32002 multiple locations. This field will not be present if no address can
32003 be determined. For example, a watchpoint does not have an address.
32006 Optional field containing any flags related to the address. These flags are
32007 architecture-dependent; see @ref{Architectures} for their meaning for a
32011 If known, the function in which the breakpoint appears.
32012 If not known, this field is not present.
32015 The name of the source file which contains this function, if known.
32016 If not known, this field is not present.
32019 The full file name of the source file which contains this function, if
32020 known. If not known, this field is not present.
32023 The line number at which this breakpoint appears, if known.
32024 If not known, this field is not present.
32027 If the source file is not known, this field may be provided. If
32028 provided, this holds the address of the breakpoint, possibly followed
32032 If this breakpoint is pending, this field is present and holds the
32033 text used to set the breakpoint, as entered by the user.
32036 Where this breakpoint's condition is evaluated, either @samp{host} or
32040 If this is a thread-specific breakpoint, then this identifies the
32041 thread in which the breakpoint can trigger.
32044 If this is an inferior-specific breakpoint, this this identifies the
32045 inferior in which the breakpoint can trigger.
32048 If this breakpoint is restricted to a particular Ada task, then this
32049 field will hold the task identifier.
32052 If the breakpoint is conditional, this is the condition expression.
32055 The ignore count of the breakpoint.
32058 The enable count of the breakpoint.
32060 @item traceframe-usage
32063 @item static-tracepoint-marker-string-id
32064 For a static tracepoint, the name of the static tracepoint marker.
32067 For a masked watchpoint, this is the mask.
32070 A tracepoint's pass count.
32072 @item original-location
32073 The location of the breakpoint as originally specified by the user.
32074 This field is optional.
32077 The number of times the breakpoint has been hit.
32080 This field is only given for tracepoints. This is either @samp{y},
32081 meaning that the tracepoint is installed, or @samp{n}, meaning that it
32085 Some extra data, the exact contents of which are type-dependent.
32088 This field is present if the breakpoint has multiple locations. It is also
32089 exceptionally present if the breakpoint is enabled and has a single, disabled
32092 The value is a list of locations. The format of a location is described below.
32096 A location in a multi-location breakpoint is represented as a tuple with the
32102 The location number as a dotted pair, like @samp{1.2}. The first digit is the
32103 number of the parent breakpoint. The second digit is the number of the
32104 location within that breakpoint.
32107 There are three possible values, with the following meanings:
32110 The location is enabled.
32112 The location is disabled by the user.
32114 The location is disabled because the breakpoint condition is invalid
32119 The address of this location as an hexidecimal number.
32122 Optional field containing any flags related to the address. These flags are
32123 architecture-dependent; see @ref{Architectures} for their meaning for a
32127 If known, the function in which the location appears.
32128 If not known, this field is not present.
32131 The name of the source file which contains this location, if known.
32132 If not known, this field is not present.
32135 The full file name of the source file which contains this location, if
32136 known. If not known, this field is not present.
32139 The line number at which this location appears, if known.
32140 If not known, this field is not present.
32142 @item thread-groups
32143 The thread groups this location is in.
32147 For example, here is what the output of @code{-break-insert}
32148 (@pxref{GDB/MI Breakpoint Commands}) might be:
32151 -> -break-insert main
32152 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
32153 enabled="y",addr="0x08048564",func="main",file="myprog.c",
32154 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
32159 @node GDB/MI Frame Information
32160 @subsection @sc{gdb/mi} Frame Information
32162 Response from many MI commands includes an information about stack
32163 frame. This information is a tuple that may have the following
32168 The level of the stack frame. The innermost frame has the level of
32169 zero. This field is always present.
32172 The name of the function corresponding to the frame. This field may
32173 be absent if @value{GDBN} is unable to determine the function name.
32176 The code address for the frame. This field is always present.
32179 Optional field containing any flags related to the address. These flags are
32180 architecture-dependent; see @ref{Architectures} for their meaning for a
32184 The name of the source files that correspond to the frame's code
32185 address. This field may be absent.
32188 The source line corresponding to the frames' code address. This field
32192 The name of the binary file (either executable or shared library) the
32193 corresponds to the frame's code address. This field may be absent.
32197 @node GDB/MI Thread Information
32198 @subsection @sc{gdb/mi} Thread Information
32200 Whenever @value{GDBN} has to report an information about a thread, it
32201 uses a tuple with the following fields. The fields are always present unless
32206 The global numeric id assigned to the thread by @value{GDBN}.
32209 The target-specific string identifying the thread.
32212 Additional information about the thread provided by the target.
32213 It is supposed to be human-readable and not interpreted by the
32214 frontend. This field is optional.
32217 The name of the thread. If the user specified a name using the
32218 @code{thread name} command, then this name is given. Otherwise, if
32219 @value{GDBN} can extract the thread name from the target, then that
32220 name is given. If @value{GDBN} cannot find the thread name, then this
32224 The execution state of the thread, either @samp{stopped} or @samp{running},
32225 depending on whether the thread is presently running.
32228 The stack frame currently executing in the thread. This field is only present
32229 if the thread is stopped. Its format is documented in
32230 @ref{GDB/MI Frame Information}.
32233 The value of this field is an integer number of the processor core the
32234 thread was last seen on. This field is optional.
32237 @node GDB/MI Ada Exception Information
32238 @subsection @sc{gdb/mi} Ada Exception Information
32240 Whenever a @code{*stopped} record is emitted because the program
32241 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
32242 @value{GDBN} provides the name of the exception that was raised via
32243 the @code{exception-name} field. Also, for exceptions that were raised
32244 with an exception message, @value{GDBN} provides that message via
32245 the @code{exception-message} field.
32247 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32248 @node GDB/MI Simple Examples
32249 @section Simple Examples of @sc{gdb/mi} Interaction
32250 @cindex @sc{gdb/mi}, simple examples
32252 This subsection presents several simple examples of interaction using
32253 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
32254 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
32255 the output received from @sc{gdb/mi}.
32257 Note the line breaks shown in the examples are here only for
32258 readability, they don't appear in the real output.
32260 @subheading Setting a Breakpoint
32262 Setting a breakpoint generates synchronous output which contains detailed
32263 information of the breakpoint.
32266 -> -break-insert main
32267 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
32268 enabled="y",addr="0x08048564",func="main",file="myprog.c",
32269 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
32274 @subheading Program Execution
32276 Program execution generates asynchronous records and MI gives the
32277 reason that execution stopped.
32283 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
32284 frame=@{addr="0x08048564",func="main",
32285 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
32286 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68",
32287 arch="i386:x86_64"@}
32292 <- *stopped,reason="exited-normally"
32296 @subheading Quitting @value{GDBN}
32298 Quitting @value{GDBN} just prints the result class @samp{^exit}.
32306 Please note that @samp{^exit} is printed immediately, but it might
32307 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
32308 performs necessary cleanups, including killing programs being debugged
32309 or disconnecting from debug hardware, so the frontend should wait till
32310 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
32311 fails to exit in reasonable time.
32313 @subheading A Bad Command
32315 Here's what happens if you pass a non-existent command:
32319 <- ^error,msg="Undefined MI command: rubbish"
32324 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32325 @node GDB/MI Command Description Format
32326 @section @sc{gdb/mi} Command Description Format
32328 The remaining sections describe blocks of commands. Each block of
32329 commands is laid out in a fashion similar to this section.
32331 @subheading Motivation
32333 The motivation for this collection of commands.
32335 @subheading Introduction
32337 A brief introduction to this collection of commands as a whole.
32339 @subheading Commands
32341 For each command in the block, the following is described:
32343 @subsubheading Synopsis
32346 -command @var{args}@dots{}
32349 @subsubheading Result
32351 @subsubheading @value{GDBN} Command
32353 The corresponding @value{GDBN} CLI command(s), if any.
32355 @subsubheading Example
32357 Example(s) formatted for readability. Some of the described commands have
32358 not been implemented yet and these are labeled N.A.@: (not available).
32361 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32362 @node GDB/MI Breakpoint Commands
32363 @section @sc{gdb/mi} Breakpoint Commands
32365 @cindex breakpoint commands for @sc{gdb/mi}
32366 @cindex @sc{gdb/mi}, breakpoint commands
32367 This section documents @sc{gdb/mi} commands for manipulating
32370 @findex -break-after
32371 @subheading The @code{-break-after} Command
32373 @subsubheading Synopsis
32376 -break-after @var{number} @var{count}
32379 The breakpoint number @var{number} is not in effect until it has been
32380 hit @var{count} times. To see how this is reflected in the output of
32381 the @samp{-break-list} command, see the description of the
32382 @samp{-break-list} command below.
32384 @subsubheading @value{GDBN} Command
32386 The corresponding @value{GDBN} command is @samp{ignore}.
32388 @subsubheading Example
32393 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
32394 enabled="y",addr="0x000100d0",func="main",file="hello.c",
32395 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
32403 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
32404 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
32405 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
32406 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
32407 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
32408 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
32409 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
32410 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
32411 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
32412 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
32417 @findex -break-catch
32418 @subheading The @code{-break-catch} Command
32421 @findex -break-commands
32422 @subheading The @code{-break-commands} Command
32424 @subsubheading Synopsis
32427 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
32430 Specifies the CLI commands that should be executed when breakpoint
32431 @var{number} is hit. The parameters @var{command1} to @var{commandN}
32432 are the commands. If no command is specified, any previously-set
32433 commands are cleared. @xref{Break Commands}. Typical use of this
32434 functionality is tracing a program, that is, printing of values of
32435 some variables whenever breakpoint is hit and then continuing.
32437 @subsubheading @value{GDBN} Command
32439 The corresponding @value{GDBN} command is @samp{commands}.
32441 @subsubheading Example
32446 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
32447 enabled="y",addr="0x000100d0",func="main",file="hello.c",
32448 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
32451 -break-commands 1 "print v" "continue"
32456 @findex -break-condition
32457 @subheading The @code{-break-condition} Command
32459 @subsubheading Synopsis
32462 -break-condition [ --force ] @var{number} [ @var{expr} ]
32465 Breakpoint @var{number} will stop the program only if the condition in
32466 @var{expr} is true. The condition becomes part of the
32467 @samp{-break-list} output (see the description of the @samp{-break-list}
32468 command below). If the @samp{--force} flag is passed, the condition
32469 is forcibly defined even when it is invalid for all locations of
32470 breakpoint @var{number}. If the @var{expr} argument is omitted,
32471 breakpoint @var{number} becomes unconditional.
32473 @subsubheading @value{GDBN} Command
32475 The corresponding @value{GDBN} command is @samp{condition}.
32477 @subsubheading Example
32481 -break-condition 1 1
32485 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
32486 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
32487 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
32488 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
32489 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
32490 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
32491 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
32492 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
32493 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
32494 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
32498 @findex -break-delete
32499 @subheading The @code{-break-delete} Command
32501 @subsubheading Synopsis
32504 -break-delete ( @var{breakpoint} )+
32507 Delete the breakpoint(s) whose number(s) are specified in the argument
32508 list. This is obviously reflected in the breakpoint list.
32510 @subsubheading @value{GDBN} Command
32512 The corresponding @value{GDBN} command is @samp{delete}.
32514 @subsubheading Example
32522 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
32523 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
32524 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
32525 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
32526 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
32527 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
32528 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
32533 @findex -break-disable
32534 @subheading The @code{-break-disable} Command
32536 @subsubheading Synopsis
32539 -break-disable ( @var{breakpoint} )+
32542 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
32543 break list is now set to @samp{n} for the named @var{breakpoint}(s).
32545 @subsubheading @value{GDBN} Command
32547 The corresponding @value{GDBN} command is @samp{disable}.
32549 @subsubheading Example
32557 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
32558 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
32559 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
32560 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
32561 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
32562 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
32563 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
32564 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
32565 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
32566 line="5",thread-groups=["i1"],times="0"@}]@}
32570 @findex -break-enable
32571 @subheading The @code{-break-enable} Command
32573 @subsubheading Synopsis
32576 -break-enable ( @var{breakpoint} )+
32579 Enable (previously disabled) @var{breakpoint}(s).
32581 @subsubheading @value{GDBN} Command
32583 The corresponding @value{GDBN} command is @samp{enable}.
32585 @subsubheading Example
32593 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
32594 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
32595 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
32596 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
32597 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
32598 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
32599 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
32600 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
32601 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
32602 line="5",thread-groups=["i1"],times="0"@}]@}
32606 @findex -break-info
32607 @subheading The @code{-break-info} Command
32609 @subsubheading Synopsis
32612 -break-info @var{breakpoint}
32616 Get information about a single breakpoint.
32618 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
32619 Information}, for details on the format of each breakpoint in the
32622 @subsubheading @value{GDBN} Command
32624 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
32626 @subsubheading Example
32629 @findex -break-insert
32630 @anchor{-break-insert}
32631 @subheading The @code{-break-insert} Command
32633 @subsubheading Synopsis
32636 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ] [ --qualified ]
32637 [ -c @var{condition} ] [ --force-condition ] [ -i @var{ignore-count} ]
32638 [ -p @var{thread-id} ] [ -g @var{thread-group-id} ] [ @var{locspec} ]
32642 If specified, @var{locspec}, can be one of:
32645 @item linespec location
32646 A linespec location. @xref{Linespec Locations}.
32648 @item explicit location
32649 An explicit location. @sc{gdb/mi} explicit locations are
32650 analogous to the CLI's explicit locations using the option names
32651 listed below. @xref{Explicit Locations}.
32654 @item --source @var{filename}
32655 The source file name of the location. This option requires the use
32656 of either @samp{--function} or @samp{--line}.
32658 @item --function @var{function}
32659 The name of a function or method.
32661 @item --label @var{label}
32662 The name of a label.
32664 @item --line @var{lineoffset}
32665 An absolute or relative line offset from the start of the location.
32668 @item address location
32669 An address location, *@var{address}. @xref{Address Locations}.
32673 The possible optional parameters of this command are:
32677 Insert a temporary breakpoint.
32679 Insert a hardware breakpoint.
32681 If @var{locspec} cannot be resolved (for example if it
32682 refers to unknown files or functions), create a pending
32683 breakpoint. Without this flag, @value{GDBN} will report
32684 an error, and won't create a breakpoint, if @var{locspec}
32687 Create a disabled breakpoint.
32689 Create a tracepoint. @xref{Tracepoints}. When this parameter
32690 is used together with @samp{-h}, a fast tracepoint is created.
32691 @item -c @var{condition}
32692 Make the breakpoint conditional on @var{condition}.
32693 @item --force-condition
32694 Forcibly define the breakpoint even if the condition is invalid at
32695 all of the breakpoint locations.
32696 @item -i @var{ignore-count}
32697 Initialize the @var{ignore-count}.
32698 @item -p @var{thread-id}
32699 Restrict the breakpoint to the thread with the specified global
32700 @var{thread-id}. @var{thread-id} must be a valid thread-id at the
32701 time the breakpoint is requested. Breakpoints created with a
32702 @var{thread-id} will automatically be deleted when the corresponding
32704 @item -g @var{thread-group-id}
32705 Restrict the breakpoint to the thread group with the specified
32706 @var{thread-group-id}.
32708 This option makes @value{GDBN} interpret a function name specified as
32709 a complete fully-qualified name.
32712 @subsubheading Result
32714 @xref{GDB/MI Breakpoint Information}, for details on the format of the
32715 resulting breakpoint.
32717 Note: this format is open to change.
32718 @c An out-of-band breakpoint instead of part of the result?
32720 @subsubheading @value{GDBN} Command
32722 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
32723 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
32725 @subsubheading Example
32730 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
32731 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
32734 -break-insert -t foo
32735 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
32736 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
32740 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
32741 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
32742 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
32743 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
32744 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
32745 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
32746 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
32747 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
32748 addr="0x0001072c", func="main",file="recursive2.c",
32749 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
32751 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
32752 addr="0x00010774",func="foo",file="recursive2.c",
32753 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
32758 @findex -dprintf-insert
32759 @subheading The @code{-dprintf-insert} Command
32761 @subsubheading Synopsis
32764 -dprintf-insert [ -t ] [ -f ] [ -d ] [ --qualified ]
32765 [ -c @var{condition} ] [--force-condition] [ -i @var{ignore-count} ]
32766 [ -p @var{thread-id} ] [ @var{locspec} ] @var{format}
32767 [ @var{argument}@dots{} ]
32771 Insert a new dynamic print breakpoint at the given location.
32772 @xref{Dynamic Printf}. @var{format} is the format to use, and any
32773 remaining arguments are passed as expressions to substitute.
32776 If supplied, @var{locspec} and @code{--qualified} may be specified
32777 the same way as for the @code{-break-insert} command.
32778 @xref{-break-insert}.
32780 The possible optional parameters of this command are:
32784 Insert a temporary breakpoint.
32786 If @var{locspec} cannot be parsed (for example, if it
32787 refers to unknown files or functions), create a pending
32788 breakpoint. Without this flag, @value{GDBN} will report
32789 an error, and won't create a breakpoint, if @var{locspec}
32792 Create a disabled breakpoint.
32793 @item -c @var{condition}
32794 Make the breakpoint conditional on @var{condition}.
32795 @item --force-condition
32796 Forcibly define the breakpoint even if the condition is invalid at
32797 all of the breakpoint locations.
32798 @item -i @var{ignore-count}
32799 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
32800 to @var{ignore-count}.
32801 @item -p @var{thread-id}
32802 Restrict the breakpoint to the thread with the specified global
32806 @subsubheading Result
32808 @xref{GDB/MI Breakpoint Information}, for details on the format of the
32809 resulting breakpoint.
32811 @c An out-of-band breakpoint instead of part of the result?
32813 @subsubheading @value{GDBN} Command
32815 The corresponding @value{GDBN} command is @samp{dprintf}.
32817 @subsubheading Example
32821 4-dprintf-insert foo "At foo entry\n"
32822 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
32823 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
32824 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
32825 times="0",script=["printf \"At foo entry\\n\"","continue"],
32826 original-location="foo"@}
32828 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
32829 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
32830 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
32831 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
32832 times="0",script=["printf \"arg=%d, g=%d\\n\", arg, g","continue"],
32833 original-location="mi-dprintf.c:26"@}
32837 @findex -break-list
32838 @subheading The @code{-break-list} Command
32840 @subsubheading Synopsis
32846 Displays the list of inserted breakpoints, showing the following fields:
32850 number of the breakpoint
32852 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
32854 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
32857 is the breakpoint enabled or no: @samp{y} or @samp{n}
32859 memory location at which the breakpoint is set
32861 logical location of the breakpoint, expressed by function name, file
32863 @item Thread-groups
32864 list of thread groups to which this breakpoint applies
32866 number of times the breakpoint has been hit
32869 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
32870 @code{body} field is an empty list.
32872 @subsubheading @value{GDBN} Command
32874 The corresponding @value{GDBN} command is @samp{info break}.
32876 @subsubheading Example
32881 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
32882 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
32883 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
32884 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
32885 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
32886 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
32887 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
32888 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
32889 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
32891 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
32892 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
32893 line="13",thread-groups=["i1"],times="0"@}]@}
32897 Here's an example of the result when there are no breakpoints:
32902 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
32903 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
32904 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
32905 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
32906 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
32907 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
32908 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
32913 @findex -break-passcount
32914 @subheading The @code{-break-passcount} Command
32916 @subsubheading Synopsis
32919 -break-passcount @var{tracepoint-number} @var{passcount}
32922 Set the passcount for tracepoint @var{tracepoint-number} to
32923 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
32924 is not a tracepoint, error is emitted. This corresponds to CLI
32925 command @samp{passcount}.
32927 @findex -break-watch
32928 @subheading The @code{-break-watch} Command
32930 @subsubheading Synopsis
32933 -break-watch [ -a | -r ]
32936 Create a watchpoint. With the @samp{-a} option it will create an
32937 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
32938 read from or on a write to the memory location. With the @samp{-r}
32939 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
32940 trigger only when the memory location is accessed for reading. Without
32941 either of the options, the watchpoint created is a regular watchpoint,
32942 i.e., it will trigger when the memory location is accessed for writing.
32943 @xref{Set Watchpoints, , Setting Watchpoints}.
32945 Note that @samp{-break-list} will report a single list of watchpoints and
32946 breakpoints inserted.
32948 @subsubheading @value{GDBN} Command
32950 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
32953 @subsubheading Example
32955 Setting a watchpoint on a variable in the @code{main} function:
32960 ^done,wpt=@{number="2",exp="x"@}
32965 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
32966 value=@{old="-268439212",new="55"@},
32967 frame=@{func="main",args=[],file="recursive2.c",
32968 fullname="/home/foo/bar/recursive2.c",line="5",arch="i386:x86_64"@}
32972 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
32973 the program execution twice: first for the variable changing value, then
32974 for the watchpoint going out of scope.
32979 ^done,wpt=@{number="5",exp="C"@}
32984 *stopped,reason="watchpoint-trigger",
32985 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
32986 frame=@{func="callee4",args=[],
32987 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
32988 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
32989 arch="i386:x86_64"@}
32994 *stopped,reason="watchpoint-scope",wpnum="5",
32995 frame=@{func="callee3",args=[@{name="strarg",
32996 value="0x11940 \"A string argument.\""@}],
32997 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
32998 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
32999 arch="i386:x86_64"@}
33003 Listing breakpoints and watchpoints, at different points in the program
33004 execution. Note that once the watchpoint goes out of scope, it is
33010 ^done,wpt=@{number="2",exp="C"@}
33013 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
33014 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
33015 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
33016 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
33017 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
33018 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
33019 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
33020 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
33021 addr="0x00010734",func="callee4",
33022 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33023 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
33025 bkpt=@{number="2",type="watchpoint",disp="keep",
33026 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
33031 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
33032 value=@{old="-276895068",new="3"@},
33033 frame=@{func="callee4",args=[],
33034 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33035 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
33036 arch="i386:x86_64"@}
33039 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
33040 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
33041 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
33042 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
33043 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
33044 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
33045 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
33046 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
33047 addr="0x00010734",func="callee4",
33048 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33049 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
33051 bkpt=@{number="2",type="watchpoint",disp="keep",
33052 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
33056 ^done,reason="watchpoint-scope",wpnum="2",
33057 frame=@{func="callee3",args=[@{name="strarg",
33058 value="0x11940 \"A string argument.\""@}],
33059 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33060 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
33061 arch="i386:x86_64"@}
33064 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
33065 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
33066 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
33067 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
33068 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
33069 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
33070 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
33071 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
33072 addr="0x00010734",func="callee4",
33073 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33074 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
33075 thread-groups=["i1"],times="1"@}]@}
33080 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33081 @node GDB/MI Catchpoint Commands
33082 @section @sc{gdb/mi} Catchpoint Commands
33084 This section documents @sc{gdb/mi} commands for manipulating
33088 * Shared Library GDB/MI Catchpoint Commands::
33089 * Ada Exception GDB/MI Catchpoint Commands::
33090 * C++ Exception GDB/MI Catchpoint Commands::
33093 @node Shared Library GDB/MI Catchpoint Commands
33094 @subsection Shared Library @sc{gdb/mi} Catchpoints
33096 @findex -catch-load
33097 @subheading The @code{-catch-load} Command
33099 @subsubheading Synopsis
33102 -catch-load [ -t ] [ -d ] @var{regexp}
33105 Add a catchpoint for library load events. If the @samp{-t} option is used,
33106 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
33107 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
33108 in a disabled state. The @samp{regexp} argument is a regular
33109 expression used to match the name of the loaded library.
33112 @subsubheading @value{GDBN} Command
33114 The corresponding @value{GDBN} command is @samp{catch load}.
33116 @subsubheading Example
33119 -catch-load -t foo.so
33120 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
33121 what="load of library matching foo.so",catch-type="load",times="0"@}
33126 @findex -catch-unload
33127 @subheading The @code{-catch-unload} Command
33129 @subsubheading Synopsis
33132 -catch-unload [ -t ] [ -d ] @var{regexp}
33135 Add a catchpoint for library unload events. If the @samp{-t} option is
33136 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
33137 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
33138 created in a disabled state. The @samp{regexp} argument is a regular
33139 expression used to match the name of the unloaded library.
33141 @subsubheading @value{GDBN} Command
33143 The corresponding @value{GDBN} command is @samp{catch unload}.
33145 @subsubheading Example
33148 -catch-unload -d bar.so
33149 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
33150 what="load of library matching bar.so",catch-type="unload",times="0"@}
33154 @node Ada Exception GDB/MI Catchpoint Commands
33155 @subsection Ada Exception @sc{gdb/mi} Catchpoints
33157 The following @sc{gdb/mi} commands can be used to create catchpoints
33158 that stop the execution when Ada exceptions are being raised.
33160 @findex -catch-assert
33161 @subheading The @code{-catch-assert} Command
33163 @subsubheading Synopsis
33166 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
33169 Add a catchpoint for failed Ada assertions.
33171 The possible optional parameters for this command are:
33174 @item -c @var{condition}
33175 Make the catchpoint conditional on @var{condition}.
33177 Create a disabled catchpoint.
33179 Create a temporary catchpoint.
33182 @subsubheading @value{GDBN} Command
33184 The corresponding @value{GDBN} command is @samp{catch assert}.
33186 @subsubheading Example
33190 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
33191 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
33192 thread-groups=["i1"],times="0",
33193 original-location="__gnat_debug_raise_assert_failure"@}
33197 @findex -catch-exception
33198 @subheading The @code{-catch-exception} Command
33200 @subsubheading Synopsis
33203 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
33207 Add a catchpoint stopping when Ada exceptions are raised.
33208 By default, the command stops the program when any Ada exception
33209 gets raised. But it is also possible, by using some of the
33210 optional parameters described below, to create more selective
33213 The possible optional parameters for this command are:
33216 @item -c @var{condition}
33217 Make the catchpoint conditional on @var{condition}.
33219 Create a disabled catchpoint.
33220 @item -e @var{exception-name}
33221 Only stop when @var{exception-name} is raised. This option cannot
33222 be used combined with @samp{-u}.
33224 Create a temporary catchpoint.
33226 Stop only when an unhandled exception gets raised. This option
33227 cannot be used combined with @samp{-e}.
33230 @subsubheading @value{GDBN} Command
33232 The corresponding @value{GDBN} commands are @samp{catch exception}
33233 and @samp{catch exception unhandled}.
33235 @subsubheading Example
33238 -catch-exception -e Program_Error
33239 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
33240 enabled="y",addr="0x0000000000404874",
33241 what="`Program_Error' Ada exception", thread-groups=["i1"],
33242 times="0",original-location="__gnat_debug_raise_exception"@}
33246 @findex -catch-handlers
33247 @subheading The @code{-catch-handlers} Command
33249 @subsubheading Synopsis
33252 -catch-handlers [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
33256 Add a catchpoint stopping when Ada exceptions are handled.
33257 By default, the command stops the program when any Ada exception
33258 gets handled. But it is also possible, by using some of the
33259 optional parameters described below, to create more selective
33262 The possible optional parameters for this command are:
33265 @item -c @var{condition}
33266 Make the catchpoint conditional on @var{condition}.
33268 Create a disabled catchpoint.
33269 @item -e @var{exception-name}
33270 Only stop when @var{exception-name} is handled.
33272 Create a temporary catchpoint.
33275 @subsubheading @value{GDBN} Command
33277 The corresponding @value{GDBN} command is @samp{catch handlers}.
33279 @subsubheading Example
33282 -catch-handlers -e Constraint_Error
33283 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
33284 enabled="y",addr="0x0000000000402f68",
33285 what="`Constraint_Error' Ada exception handlers",thread-groups=["i1"],
33286 times="0",original-location="__gnat_begin_handler"@}
33290 @node C++ Exception GDB/MI Catchpoint Commands
33291 @subsection C@t{++} Exception @sc{gdb/mi} Catchpoints
33293 The following @sc{gdb/mi} commands can be used to create catchpoints
33294 that stop the execution when C@t{++} exceptions are being throw, rethrown,
33297 @findex -catch-throw
33298 @subheading The @code{-catch-throw} Command
33300 @subsubheading Synopsis
33303 -catch-throw [ -t ] [ -r @var{regexp}]
33306 Stop when the debuggee throws a C@t{++} exception. If @var{regexp} is
33307 given, then only exceptions whose type matches the regular expression
33310 If @samp{-t} is given, then the catchpoint is enabled only for one
33311 stop, the catchpoint is automatically deleted after stopping once for
33314 @subsubheading @value{GDBN} Command
33316 The corresponding @value{GDBN} commands are @samp{catch throw}
33317 and @samp{tcatch throw} (@pxref{Set Catchpoints}).
33319 @subsubheading Example
33322 -catch-throw -r exception_type
33323 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
33324 what="exception throw",catch-type="throw",
33325 thread-groups=["i1"],
33326 regexp="exception_type",times="0"@}
33332 ~"Catchpoint 1 (exception thrown), 0x00007ffff7ae00ed
33333 in __cxa_throw () from /lib64/libstdc++.so.6\n"
33334 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
33335 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_throw",
33336 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
33337 thread-id="1",stopped-threads="all",core="6"
33341 @findex -catch-rethrow
33342 @subheading The @code{-catch-rethrow} Command
33344 @subsubheading Synopsis
33347 -catch-rethrow [ -t ] [ -r @var{regexp}]
33350 Stop when a C@t{++} exception is re-thrown. If @var{regexp} is given,
33351 then only exceptions whose type matches the regular expression will be
33354 If @samp{-t} is given, then the catchpoint is enabled only for one
33355 stop, the catchpoint is automatically deleted after the first event is
33358 @subsubheading @value{GDBN} Command
33360 The corresponding @value{GDBN} commands are @samp{catch rethrow}
33361 and @samp{tcatch rethrow} (@pxref{Set Catchpoints}).
33363 @subsubheading Example
33366 -catch-rethrow -r exception_type
33367 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
33368 what="exception rethrow",catch-type="rethrow",
33369 thread-groups=["i1"],
33370 regexp="exception_type",times="0"@}
33376 ~"Catchpoint 1 (exception rethrown), 0x00007ffff7ae00ed
33377 in __cxa_rethrow () from /lib64/libstdc++.so.6\n"
33378 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
33379 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_rethrow",
33380 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
33381 thread-id="1",stopped-threads="all",core="6"
33385 @findex -catch-catch
33386 @subheading The @code{-catch-catch} Command
33388 @subsubheading Synopsis
33391 -catch-catch [ -t ] [ -r @var{regexp}]
33394 Stop when the debuggee catches a C@t{++} exception. If @var{regexp}
33395 is given, then only exceptions whose type matches the regular
33396 expression will be caught.
33398 If @samp{-t} is given, then the catchpoint is enabled only for one
33399 stop, the catchpoint is automatically deleted after the first event is
33402 @subsubheading @value{GDBN} Command
33404 The corresponding @value{GDBN} commands are @samp{catch catch}
33405 and @samp{tcatch catch} (@pxref{Set Catchpoints}).
33407 @subsubheading Example
33410 -catch-catch -r exception_type
33411 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
33412 what="exception catch",catch-type="catch",
33413 thread-groups=["i1"],
33414 regexp="exception_type",times="0"@}
33420 ~"Catchpoint 1 (exception caught), 0x00007ffff7ae00ed
33421 in __cxa_begin_catch () from /lib64/libstdc++.so.6\n"
33422 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
33423 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_begin_catch",
33424 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
33425 thread-id="1",stopped-threads="all",core="6"
33429 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33430 @node GDB/MI Program Context
33431 @section @sc{gdb/mi} Program Context
33433 @findex -exec-arguments
33434 @subheading The @code{-exec-arguments} Command
33437 @subsubheading Synopsis
33440 -exec-arguments @var{args}
33443 Set the inferior program arguments, to be used in the next
33446 @subsubheading @value{GDBN} Command
33448 The corresponding @value{GDBN} command is @samp{set args}.
33450 @subsubheading Example
33454 -exec-arguments -v word
33461 @findex -exec-show-arguments
33462 @subheading The @code{-exec-show-arguments} Command
33464 @subsubheading Synopsis
33467 -exec-show-arguments
33470 Print the arguments of the program.
33472 @subsubheading @value{GDBN} Command
33474 The corresponding @value{GDBN} command is @samp{show args}.
33476 @subsubheading Example
33481 @findex -environment-cd
33482 @subheading The @code{-environment-cd} Command
33484 @subsubheading Synopsis
33487 -environment-cd @var{pathdir}
33490 Set @value{GDBN}'s working directory.
33492 @subsubheading @value{GDBN} Command
33494 The corresponding @value{GDBN} command is @samp{cd}.
33496 @subsubheading Example
33500 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
33506 @findex -environment-directory
33507 @subheading The @code{-environment-directory} Command
33509 @subsubheading Synopsis
33512 -environment-directory [ -r ] [ @var{pathdir} ]+
33515 Add directories @var{pathdir} to beginning of search path for source files.
33516 If the @samp{-r} option is used, the search path is reset to the default
33517 search path. If directories @var{pathdir} are supplied in addition to the
33518 @samp{-r} option, the search path is first reset and then addition
33520 Multiple directories may be specified, separated by blanks. Specifying
33521 multiple directories in a single command
33522 results in the directories added to the beginning of the
33523 search path in the same order they were presented in the command.
33524 If blanks are needed as
33525 part of a directory name, double-quotes should be used around
33526 the name. In the command output, the path will show up separated
33527 by the system directory-separator character. The directory-separator
33528 character must not be used
33529 in any directory name.
33530 If no directories are specified, the current search path is displayed.
33532 @subsubheading @value{GDBN} Command
33534 The corresponding @value{GDBN} command is @samp{dir}.
33536 @subsubheading Example
33540 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
33541 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
33543 -environment-directory ""
33544 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
33546 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
33547 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
33549 -environment-directory -r
33550 ^done,source-path="$cdir:$cwd"
33555 @findex -environment-path
33556 @subheading The @code{-environment-path} Command
33558 @subsubheading Synopsis
33561 -environment-path [ -r ] [ @var{pathdir} ]+
33564 Add directories @var{pathdir} to beginning of search path for object files.
33565 If the @samp{-r} option is used, the search path is reset to the original
33566 search path that existed at gdb start-up. If directories @var{pathdir} are
33567 supplied in addition to the
33568 @samp{-r} option, the search path is first reset and then addition
33570 Multiple directories may be specified, separated by blanks. Specifying
33571 multiple directories in a single command
33572 results in the directories added to the beginning of the
33573 search path in the same order they were presented in the command.
33574 If blanks are needed as
33575 part of a directory name, double-quotes should be used around
33576 the name. In the command output, the path will show up separated
33577 by the system directory-separator character. The directory-separator
33578 character must not be used
33579 in any directory name.
33580 If no directories are specified, the current path is displayed.
33583 @subsubheading @value{GDBN} Command
33585 The corresponding @value{GDBN} command is @samp{path}.
33587 @subsubheading Example
33592 ^done,path="/usr/bin"
33594 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
33595 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
33597 -environment-path -r /usr/local/bin
33598 ^done,path="/usr/local/bin:/usr/bin"
33603 @findex -environment-pwd
33604 @subheading The @code{-environment-pwd} Command
33606 @subsubheading Synopsis
33612 Show the current working directory.
33614 @subsubheading @value{GDBN} Command
33616 The corresponding @value{GDBN} command is @samp{pwd}.
33618 @subsubheading Example
33623 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
33627 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33628 @node GDB/MI Thread Commands
33629 @section @sc{gdb/mi} Thread Commands
33632 @findex -thread-info
33633 @subheading The @code{-thread-info} Command
33635 @subsubheading Synopsis
33638 -thread-info [ @var{thread-id} ]
33641 Reports information about either a specific thread, if the
33642 @var{thread-id} parameter is present, or about all threads.
33643 @var{thread-id} is the thread's global thread ID. When printing
33644 information about all threads, also reports the global ID of the
33647 @subsubheading @value{GDBN} Command
33649 The @samp{info thread} command prints the same information
33652 @subsubheading Result
33654 The result contains the following attributes:
33658 A list of threads. The format of the elements of the list is described in
33659 @ref{GDB/MI Thread Information}.
33661 @item current-thread-id
33662 The global id of the currently selected thread. This field is omitted if there
33663 is no selected thread (for example, when the selected inferior is not running,
33664 and therefore has no threads) or if a @var{thread-id} argument was passed to
33669 @subsubheading Example
33674 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
33675 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
33676 args=[]@},state="running"@},
33677 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
33678 frame=@{level="0",addr="0x0804891f",func="foo",
33679 args=[@{name="i",value="10"@}],
33680 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},
33681 state="running"@}],
33682 current-thread-id="1"
33686 @findex -thread-list-ids
33687 @subheading The @code{-thread-list-ids} Command
33689 @subsubheading Synopsis
33695 Produces a list of the currently known global @value{GDBN} thread ids.
33696 At the end of the list it also prints the total number of such
33699 This command is retained for historical reasons, the
33700 @code{-thread-info} command should be used instead.
33702 @subsubheading @value{GDBN} Command
33704 Part of @samp{info threads} supplies the same information.
33706 @subsubheading Example
33711 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
33712 current-thread-id="1",number-of-threads="3"
33717 @findex -thread-select
33718 @subheading The @code{-thread-select} Command
33720 @subsubheading Synopsis
33723 -thread-select @var{thread-id}
33726 Make thread with global thread number @var{thread-id} the current
33727 thread. It prints the number of the new current thread, and the
33728 topmost frame for that thread.
33730 This command is deprecated in favor of explicitly using the
33731 @samp{--thread} option to each command.
33733 @subsubheading @value{GDBN} Command
33735 The corresponding @value{GDBN} command is @samp{thread}.
33737 @subsubheading Example
33744 *stopped,reason="end-stepping-range",thread-id="2",line="187",
33745 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
33749 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
33750 number-of-threads="3"
33753 ^done,new-thread-id="3",
33754 frame=@{level="0",func="vprintf",
33755 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
33756 @{name="arg",value="0x2"@}],file="vprintf.c",line="31",arch="i386:x86_64"@}
33760 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33761 @node GDB/MI Ada Tasking Commands
33762 @section @sc{gdb/mi} Ada Tasking Commands
33764 @findex -ada-task-info
33765 @subheading The @code{-ada-task-info} Command
33767 @subsubheading Synopsis
33770 -ada-task-info [ @var{task-id} ]
33773 Reports information about either a specific Ada task, if the
33774 @var{task-id} parameter is present, or about all Ada tasks.
33776 @subsubheading @value{GDBN} Command
33778 The @samp{info tasks} command prints the same information
33779 about all Ada tasks (@pxref{Ada Tasks}).
33781 @subsubheading Result
33783 The result is a table of Ada tasks. The following columns are
33784 defined for each Ada task:
33788 This field exists only for the current thread. It has the value @samp{*}.
33791 The identifier that @value{GDBN} uses to refer to the Ada task.
33794 The identifier that the target uses to refer to the Ada task.
33797 The global thread identifier of the thread corresponding to the Ada
33800 This field should always exist, as Ada tasks are always implemented
33801 on top of a thread. But if @value{GDBN} cannot find this corresponding
33802 thread for any reason, the field is omitted.
33805 This field exists only when the task was created by another task.
33806 In this case, it provides the ID of the parent task.
33809 The base priority of the task.
33812 The current state of the task. For a detailed description of the
33813 possible states, see @ref{Ada Tasks}.
33816 The name of the task.
33820 @subsubheading Example
33824 ^done,tasks=@{nr_rows="3",nr_cols="8",
33825 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
33826 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
33827 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
33828 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
33829 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
33830 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
33831 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
33832 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
33833 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
33834 state="Child Termination Wait",name="main_task"@}]@}
33838 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33839 @node GDB/MI Program Execution
33840 @section @sc{gdb/mi} Program Execution
33842 These are the asynchronous commands which generate the out-of-band
33843 record @samp{*stopped}. Currently @value{GDBN} only really executes
33844 asynchronously with remote targets and this interaction is mimicked in
33847 @findex -exec-continue
33848 @subheading The @code{-exec-continue} Command
33850 @subsubheading Synopsis
33853 -exec-continue [--reverse] [--all|--thread-group N]
33856 Resumes the execution of the inferior program, which will continue
33857 to execute until it reaches a debugger stop event. If the
33858 @samp{--reverse} option is specified, execution resumes in reverse until
33859 it reaches a stop event. Stop events may include
33862 breakpoints or watchpoints
33864 signals or exceptions
33866 the end of the process (or its beginning under @samp{--reverse})
33868 the end or beginning of a replay log if one is being used.
33870 In all-stop mode (@pxref{All-Stop
33871 Mode}), may resume only one thread, or all threads, depending on the
33872 value of the @samp{scheduler-locking} variable. If @samp{--all} is
33873 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
33874 ignored in all-stop mode. If the @samp{--thread-group} options is
33875 specified, then all threads in that thread group are resumed.
33877 @subsubheading @value{GDBN} Command
33879 The corresponding @value{GDBN} corresponding is @samp{continue}.
33881 @subsubheading Example
33888 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
33889 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
33890 line="13",arch="i386:x86_64"@}
33894 For a @samp{breakpoint-hit} stopped reason, when the breakpoint
33895 encountered has multiple locations, the field @samp{bkptno} is
33896 followed by the field @samp{locno}.
33903 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",locno="3",frame=@{
33904 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
33905 line="13",arch="i386:x86_64"@}
33909 @findex -exec-finish
33910 @subheading The @code{-exec-finish} Command
33912 @subsubheading Synopsis
33915 -exec-finish [--reverse]
33918 Resumes the execution of the inferior program until the current
33919 function is exited. Displays the results returned by the function.
33920 If the @samp{--reverse} option is specified, resumes the reverse
33921 execution of the inferior program until the point where current
33922 function was called.
33924 @subsubheading @value{GDBN} Command
33926 The corresponding @value{GDBN} command is @samp{finish}.
33928 @subsubheading Example
33930 Function returning @code{void}.
33937 *stopped,reason="function-finished",frame=@{func="main",args=[],
33938 file="hello.c",fullname="/home/foo/bar/hello.c",line="7",arch="i386:x86_64"@}
33942 Function returning other than @code{void}. The name of the internal
33943 @value{GDBN} variable storing the result is printed, together with the
33950 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
33951 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
33952 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33953 arch="i386:x86_64"@},
33954 gdb-result-var="$1",return-value="0"
33959 @findex -exec-interrupt
33960 @subheading The @code{-exec-interrupt} Command
33962 @subsubheading Synopsis
33965 -exec-interrupt [--all|--thread-group N]
33968 Interrupts the background execution of the target. Note how the token
33969 associated with the stop message is the one for the execution command
33970 that has been interrupted. The token for the interrupt itself only
33971 appears in the @samp{^done} output. If the user is trying to
33972 interrupt a non-running program, an error message will be printed.
33974 Note that when asynchronous execution is enabled, this command is
33975 asynchronous just like other execution commands. That is, first the
33976 @samp{^done} response will be printed, and the target stop will be
33977 reported after that using the @samp{*stopped} notification.
33979 In non-stop mode, only the context thread is interrupted by default.
33980 All threads (in all inferiors) will be interrupted if the
33981 @samp{--all} option is specified. If the @samp{--thread-group}
33982 option is specified, all threads in that group will be interrupted.
33984 @subsubheading @value{GDBN} Command
33986 The corresponding @value{GDBN} command is @samp{interrupt}.
33988 @subsubheading Example
33999 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
34000 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
34001 fullname="/home/foo/bar/try.c",line="13",arch="i386:x86_64"@}
34006 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
34011 @subheading The @code{-exec-jump} Command
34013 @subsubheading Synopsis
34016 -exec-jump @var{locspec}
34019 Resumes execution of the inferior program at the address to
34020 which @var{locspec} resolves. @xref{Location Specifications},
34021 for a description of the different forms of @var{locspec}.
34023 @subsubheading @value{GDBN} Command
34025 The corresponding @value{GDBN} command is @samp{jump}.
34027 @subsubheading Example
34030 -exec-jump foo.c:10
34031 *running,thread-id="all"
34037 @subheading The @code{-exec-next} Command
34039 @subsubheading Synopsis
34042 -exec-next [--reverse]
34045 Resumes execution of the inferior program, stopping when the beginning
34046 of the next source line is reached.
34048 If the @samp{--reverse} option is specified, resumes reverse execution
34049 of the inferior program, stopping at the beginning of the previous
34050 source line. If you issue this command on the first line of a
34051 function, it will take you back to the caller of that function, to the
34052 source line where the function was called.
34055 @subsubheading @value{GDBN} Command
34057 The corresponding @value{GDBN} command is @samp{next}.
34059 @subsubheading Example
34065 *stopped,reason="end-stepping-range",line="8",file="hello.c"
34070 @findex -exec-next-instruction
34071 @subheading The @code{-exec-next-instruction} Command
34073 @subsubheading Synopsis
34076 -exec-next-instruction [--reverse]
34079 Executes one machine instruction. If the instruction is a function
34080 call, continues until the function returns. If the program stops at an
34081 instruction in the middle of a source line, the address will be
34084 If the @samp{--reverse} option is specified, resumes reverse execution
34085 of the inferior program, stopping at the previous instruction. If the
34086 previously executed instruction was a return from another function,
34087 it will continue to execute in reverse until the call to that function
34088 (from the current stack frame) is reached.
34090 @subsubheading @value{GDBN} Command
34092 The corresponding @value{GDBN} command is @samp{nexti}.
34094 @subsubheading Example
34098 -exec-next-instruction
34102 *stopped,reason="end-stepping-range",
34103 addr="0x000100d4",line="5",file="hello.c"
34108 @findex -exec-return
34109 @subheading The @code{-exec-return} Command
34111 @subsubheading Synopsis
34117 Makes current function return immediately. Doesn't execute the inferior.
34118 Displays the new current frame.
34120 @subsubheading @value{GDBN} Command
34122 The corresponding @value{GDBN} command is @samp{return}.
34124 @subsubheading Example
34128 200-break-insert callee4
34129 200^done,bkpt=@{number="1",addr="0x00010734",
34130 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
34135 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
34136 frame=@{func="callee4",args=[],
34137 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
34138 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
34139 arch="i386:x86_64"@}
34145 111^done,frame=@{level="0",func="callee3",
34146 args=[@{name="strarg",
34147 value="0x11940 \"A string argument.\""@}],
34148 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
34149 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
34150 arch="i386:x86_64"@}
34156 @subheading The @code{-exec-run} Command
34158 @subsubheading Synopsis
34161 -exec-run [ --all | --thread-group N ] [ --start ]
34164 Starts execution of the inferior from the beginning. The inferior
34165 executes until either a breakpoint is encountered or the program
34166 exits. In the latter case the output will include an exit code, if
34167 the program has exited exceptionally.
34169 When neither the @samp{--all} nor the @samp{--thread-group} option
34170 is specified, the current inferior is started. If the
34171 @samp{--thread-group} option is specified, it should refer to a thread
34172 group of type @samp{process}, and that thread group will be started.
34173 If the @samp{--all} option is specified, then all inferiors will be started.
34175 Using the @samp{--start} option instructs the debugger to stop
34176 the execution at the start of the inferior's main subprogram,
34177 following the same behavior as the @code{start} command
34178 (@pxref{Starting}).
34180 @subsubheading @value{GDBN} Command
34182 The corresponding @value{GDBN} command is @samp{run}.
34184 @subsubheading Examples
34189 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
34194 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
34195 frame=@{func="main",args=[],file="recursive2.c",
34196 fullname="/home/foo/bar/recursive2.c",line="4",arch="i386:x86_64"@}
34201 Program exited normally:
34209 *stopped,reason="exited-normally"
34214 Program exited exceptionally:
34222 *stopped,reason="exited",exit-code="01"
34226 Another way the program can terminate is if it receives a signal such as
34227 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
34231 *stopped,reason="exited-signalled",signal-name="SIGINT",
34232 signal-meaning="Interrupt"
34236 @c @subheading -exec-signal
34240 @subheading The @code{-exec-step} Command
34242 @subsubheading Synopsis
34245 -exec-step [--reverse]
34248 Resumes execution of the inferior program, stopping when the beginning
34249 of the next source line is reached, if the next source line is not a
34250 function call. If it is, stop at the first instruction of the called
34251 function. If the @samp{--reverse} option is specified, resumes reverse
34252 execution of the inferior program, stopping at the beginning of the
34253 previously executed source line.
34255 @subsubheading @value{GDBN} Command
34257 The corresponding @value{GDBN} command is @samp{step}.
34259 @subsubheading Example
34261 Stepping into a function:
34267 *stopped,reason="end-stepping-range",
34268 frame=@{func="foo",args=[@{name="a",value="10"@},
34269 @{name="b",value="0"@}],file="recursive2.c",
34270 fullname="/home/foo/bar/recursive2.c",line="11",arch="i386:x86_64"@}
34280 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
34285 @findex -exec-step-instruction
34286 @subheading The @code{-exec-step-instruction} Command
34288 @subsubheading Synopsis
34291 -exec-step-instruction [--reverse]
34294 Resumes the inferior which executes one machine instruction. If the
34295 @samp{--reverse} option is specified, resumes reverse execution of the
34296 inferior program, stopping at the previously executed instruction.
34297 The output, once @value{GDBN} has stopped, will vary depending on
34298 whether we have stopped in the middle of a source line or not. In the
34299 former case, the address at which the program stopped will be printed
34302 @subsubheading @value{GDBN} Command
34304 The corresponding @value{GDBN} command is @samp{stepi}.
34306 @subsubheading Example
34310 -exec-step-instruction
34314 *stopped,reason="end-stepping-range",
34315 frame=@{func="foo",args=[],file="try.c",
34316 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
34318 -exec-step-instruction
34322 *stopped,reason="end-stepping-range",
34323 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
34324 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
34329 @findex -exec-until
34330 @subheading The @code{-exec-until} Command
34332 @subsubheading Synopsis
34335 -exec-until [ @var{locspec} ]
34338 Executes the inferior until it reaches the address to which
34339 @var{locspec} resolves. If there is no argument, the inferior
34340 executes until it reaches a source line greater than the current one.
34341 The reason for stopping in this case will be @samp{location-reached}.
34343 @subsubheading @value{GDBN} Command
34345 The corresponding @value{GDBN} command is @samp{until}.
34347 @subsubheading Example
34351 -exec-until recursive2.c:6
34355 *stopped,reason="location-reached",frame=@{func="main",args=[],
34356 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6",
34357 arch="i386:x86_64"@}
34362 @subheading -file-clear
34363 Is this going away????
34366 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34367 @node GDB/MI Stack Manipulation
34368 @section @sc{gdb/mi} Stack Manipulation Commands
34370 @findex -enable-frame-filters
34371 @subheading The @code{-enable-frame-filters} Command
34374 -enable-frame-filters
34377 @value{GDBN} allows Python-based frame filters to affect the output of
34378 the MI commands relating to stack traces. As there is no way to
34379 implement this in a fully backward-compatible way, a front end must
34380 request that this functionality be enabled.
34382 Once enabled, this feature cannot be disabled.
34384 Note that if Python support has not been compiled into @value{GDBN},
34385 this command will still succeed (and do nothing).
34387 @findex -stack-info-frame
34388 @subheading The @code{-stack-info-frame} Command
34390 @subsubheading Synopsis
34396 Get info on the selected frame.
34398 @subsubheading @value{GDBN} Command
34400 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
34401 (without arguments).
34403 @subsubheading Example
34408 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
34409 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
34410 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
34411 arch="i386:x86_64"@}
34415 @findex -stack-info-depth
34416 @subheading The @code{-stack-info-depth} Command
34418 @subsubheading Synopsis
34421 -stack-info-depth [ @var{max-depth} ]
34424 Return the depth of the stack. If the integer argument @var{max-depth}
34425 is specified, do not count beyond @var{max-depth} frames.
34427 @subsubheading @value{GDBN} Command
34429 There's no equivalent @value{GDBN} command.
34431 @subsubheading Example
34433 For a stack with frame levels 0 through 11:
34440 -stack-info-depth 4
34443 -stack-info-depth 12
34446 -stack-info-depth 11
34449 -stack-info-depth 13
34454 @anchor{-stack-list-arguments}
34455 @findex -stack-list-arguments
34456 @subheading The @code{-stack-list-arguments} Command
34458 @subsubheading Synopsis
34461 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
34462 [ @var{low-frame} @var{high-frame} ]
34465 Display a list of the arguments for the frames between @var{low-frame}
34466 and @var{high-frame} (inclusive). If @var{low-frame} and
34467 @var{high-frame} are not provided, list the arguments for the whole
34468 call stack. If the two arguments are equal, show the single frame
34469 at the corresponding level. It is an error if @var{low-frame} is
34470 larger than the actual number of frames. On the other hand,
34471 @var{high-frame} may be larger than the actual number of frames, in
34472 which case only existing frames will be returned.
34474 If @var{print-values} is 0 or @code{--no-values}, print only the names of
34475 the variables; if it is 1 or @code{--all-values}, print also their
34476 values; and if it is 2 or @code{--simple-values}, print the name,
34477 type and value for simple data types, and the name and type for arrays,
34478 structures and unions. If the option @code{--no-frame-filters} is
34479 supplied, then Python frame filters will not be executed.
34481 If the @code{--skip-unavailable} option is specified, arguments that
34482 are not available are not listed. Partially available arguments
34483 are still displayed, however.
34485 Use of this command to obtain arguments in a single frame is
34486 deprecated in favor of the @samp{-stack-list-variables} command.
34488 @subsubheading @value{GDBN} Command
34490 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
34491 @samp{gdb_get_args} command which partially overlaps with the
34492 functionality of @samp{-stack-list-arguments}.
34494 @subsubheading Example
34501 frame=@{level="0",addr="0x00010734",func="callee4",
34502 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
34503 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
34504 arch="i386:x86_64"@},
34505 frame=@{level="1",addr="0x0001076c",func="callee3",
34506 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
34507 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
34508 arch="i386:x86_64"@},
34509 frame=@{level="2",addr="0x0001078c",func="callee2",
34510 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
34511 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22",
34512 arch="i386:x86_64"@},
34513 frame=@{level="3",addr="0x000107b4",func="callee1",
34514 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
34515 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27",
34516 arch="i386:x86_64"@},
34517 frame=@{level="4",addr="0x000107e0",func="main",
34518 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
34519 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32",
34520 arch="i386:x86_64"@}]
34522 -stack-list-arguments 0
34525 frame=@{level="0",args=[]@},
34526 frame=@{level="1",args=[name="strarg"]@},
34527 frame=@{level="2",args=[name="intarg",name="strarg"]@},
34528 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
34529 frame=@{level="4",args=[]@}]
34531 -stack-list-arguments 1
34534 frame=@{level="0",args=[]@},
34536 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
34537 frame=@{level="2",args=[
34538 @{name="intarg",value="2"@},
34539 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
34540 @{frame=@{level="3",args=[
34541 @{name="intarg",value="2"@},
34542 @{name="strarg",value="0x11940 \"A string argument.\""@},
34543 @{name="fltarg",value="3.5"@}]@},
34544 frame=@{level="4",args=[]@}]
34546 -stack-list-arguments 0 2 2
34547 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
34549 -stack-list-arguments 1 2 2
34550 ^done,stack-args=[frame=@{level="2",
34551 args=[@{name="intarg",value="2"@},
34552 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
34556 @c @subheading -stack-list-exception-handlers
34559 @anchor{-stack-list-frames}
34560 @findex -stack-list-frames
34561 @subheading The @code{-stack-list-frames} Command
34563 @subsubheading Synopsis
34566 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
34569 List the frames currently on the stack. For each frame it displays the
34574 The frame number, 0 being the topmost frame, i.e., the innermost function.
34576 The @code{$pc} value for that frame.
34580 File name of the source file where the function lives.
34581 @item @var{fullname}
34582 The full file name of the source file where the function lives.
34584 Line number corresponding to the @code{$pc}.
34586 The shared library where this function is defined. This is only given
34587 if the frame's function is not known.
34589 Frame's architecture.
34592 If invoked without arguments, this command prints a backtrace for the
34593 whole stack. If given two integer arguments, it shows the frames whose
34594 levels are between the two arguments (inclusive). If the two arguments
34595 are equal, it shows the single frame at the corresponding level. It is
34596 an error if @var{low-frame} is larger than the actual number of
34597 frames. On the other hand, @var{high-frame} may be larger than the
34598 actual number of frames, in which case only existing frames will be
34599 returned. If the option @code{--no-frame-filters} is supplied, then
34600 Python frame filters will not be executed.
34602 @subsubheading @value{GDBN} Command
34604 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
34606 @subsubheading Example
34608 Full stack backtrace:
34614 [frame=@{level="0",addr="0x0001076c",func="foo",
34615 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11",
34616 arch="i386:x86_64"@},
34617 frame=@{level="1",addr="0x000107a4",func="foo",
34618 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34619 arch="i386:x86_64"@},
34620 frame=@{level="2",addr="0x000107a4",func="foo",
34621 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34622 arch="i386:x86_64"@},
34623 frame=@{level="3",addr="0x000107a4",func="foo",
34624 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34625 arch="i386:x86_64"@},
34626 frame=@{level="4",addr="0x000107a4",func="foo",
34627 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34628 arch="i386:x86_64"@},
34629 frame=@{level="5",addr="0x000107a4",func="foo",
34630 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34631 arch="i386:x86_64"@},
34632 frame=@{level="6",addr="0x000107a4",func="foo",
34633 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34634 arch="i386:x86_64"@},
34635 frame=@{level="7",addr="0x000107a4",func="foo",
34636 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34637 arch="i386:x86_64"@},
34638 frame=@{level="8",addr="0x000107a4",func="foo",
34639 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34640 arch="i386:x86_64"@},
34641 frame=@{level="9",addr="0x000107a4",func="foo",
34642 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34643 arch="i386:x86_64"@},
34644 frame=@{level="10",addr="0x000107a4",func="foo",
34645 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34646 arch="i386:x86_64"@},
34647 frame=@{level="11",addr="0x00010738",func="main",
34648 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4",
34649 arch="i386:x86_64"@}]
34653 Show frames between @var{low_frame} and @var{high_frame}:
34657 -stack-list-frames 3 5
34659 [frame=@{level="3",addr="0x000107a4",func="foo",
34660 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34661 arch="i386:x86_64"@},
34662 frame=@{level="4",addr="0x000107a4",func="foo",
34663 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34664 arch="i386:x86_64"@},
34665 frame=@{level="5",addr="0x000107a4",func="foo",
34666 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34667 arch="i386:x86_64"@}]
34671 Show a single frame:
34675 -stack-list-frames 3 3
34677 [frame=@{level="3",addr="0x000107a4",func="foo",
34678 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
34679 arch="i386:x86_64"@}]
34684 @findex -stack-list-locals
34685 @anchor{-stack-list-locals}
34686 @subheading The @code{-stack-list-locals} Command
34688 @subsubheading Synopsis
34691 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
34694 Display the local variable names for the selected frame. If
34695 @var{print-values} is 0 or @code{--no-values}, print only the names of
34696 the variables; if it is 1 or @code{--all-values}, print also their
34697 values; and if it is 2 or @code{--simple-values}, print the name,
34698 type and value for simple data types, and the name and type for arrays,
34699 structures and unions. In this last case, a frontend can immediately
34700 display the value of simple data types and create variable objects for
34701 other data types when the user wishes to explore their values in
34702 more detail. If the option @code{--no-frame-filters} is supplied, then
34703 Python frame filters will not be executed.
34705 If the @code{--skip-unavailable} option is specified, local variables
34706 that are not available are not listed. Partially available local
34707 variables are still displayed, however.
34709 This command is deprecated in favor of the
34710 @samp{-stack-list-variables} command.
34712 @subsubheading @value{GDBN} Command
34714 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
34716 @subsubheading Example
34720 -stack-list-locals 0
34721 ^done,locals=[name="A",name="B",name="C"]
34723 -stack-list-locals --all-values
34724 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
34725 @{name="C",value="@{1, 2, 3@}"@}]
34726 -stack-list-locals --simple-values
34727 ^done,locals=[@{name="A",type="int",value="1"@},
34728 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
34732 @anchor{-stack-list-variables}
34733 @findex -stack-list-variables
34734 @subheading The @code{-stack-list-variables} Command
34736 @subsubheading Synopsis
34739 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
34742 Display the names of local variables and function arguments for the selected frame. If
34743 @var{print-values} is 0 or @code{--no-values}, print only the names of
34744 the variables; if it is 1 or @code{--all-values}, print also their
34745 values; and if it is 2 or @code{--simple-values}, print the name,
34746 type and value for simple data types, and the name and type for arrays,
34747 structures and unions. If the option @code{--no-frame-filters} is
34748 supplied, then Python frame filters will not be executed.
34750 If the @code{--skip-unavailable} option is specified, local variables
34751 and arguments that are not available are not listed. Partially
34752 available arguments and local variables are still displayed, however.
34754 @subsubheading Example
34758 -stack-list-variables --thread 1 --frame 0 --all-values
34759 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
34764 @findex -stack-select-frame
34765 @subheading The @code{-stack-select-frame} Command
34767 @subsubheading Synopsis
34770 -stack-select-frame @var{framenum}
34773 Change the selected frame. Select a different frame @var{framenum} on
34776 This command in deprecated in favor of passing the @samp{--frame}
34777 option to every command.
34779 @subsubheading @value{GDBN} Command
34781 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
34782 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
34784 @subsubheading Example
34788 -stack-select-frame 2
34793 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34794 @node GDB/MI Variable Objects
34795 @section @sc{gdb/mi} Variable Objects
34799 @subheading Motivation for Variable Objects in @sc{gdb/mi}
34801 For the implementation of a variable debugger window (locals, watched
34802 expressions, etc.), we are proposing the adaptation of the existing code
34803 used by @code{Insight}.
34805 The two main reasons for that are:
34809 It has been proven in practice (it is already on its second generation).
34812 It will shorten development time (needless to say how important it is
34816 The original interface was designed to be used by Tcl code, so it was
34817 slightly changed so it could be used through @sc{gdb/mi}. This section
34818 describes the @sc{gdb/mi} operations that will be available and gives some
34819 hints about their use.
34821 @emph{Note}: In addition to the set of operations described here, we
34822 expect the @sc{gui} implementation of a variable window to require, at
34823 least, the following operations:
34826 @item @code{-gdb-show} @code{output-radix}
34827 @item @code{-stack-list-arguments}
34828 @item @code{-stack-list-locals}
34829 @item @code{-stack-select-frame}
34834 @subheading Introduction to Variable Objects
34836 @cindex variable objects in @sc{gdb/mi}
34838 Variable objects are "object-oriented" MI interface for examining and
34839 changing values of expressions. Unlike some other MI interfaces that
34840 work with expressions, variable objects are specifically designed for
34841 simple and efficient presentation in the frontend. A variable object
34842 is identified by string name. When a variable object is created, the
34843 frontend specifies the expression for that variable object. The
34844 expression can be a simple variable, or it can be an arbitrary complex
34845 expression, and can even involve CPU registers. After creating a
34846 variable object, the frontend can invoke other variable object
34847 operations---for example to obtain or change the value of a variable
34848 object, or to change display format.
34850 Variable objects have hierarchical tree structure. Any variable object
34851 that corresponds to a composite type, such as structure in C, has
34852 a number of child variable objects, for example corresponding to each
34853 element of a structure. A child variable object can itself have
34854 children, recursively. Recursion ends when we reach
34855 leaf variable objects, which always have built-in types. Child variable
34856 objects are created only by explicit request, so if a frontend
34857 is not interested in the children of a particular variable object, no
34858 child will be created.
34860 For a leaf variable object it is possible to obtain its value as a
34861 string, or set the value from a string. String value can be also
34862 obtained for a non-leaf variable object, but it's generally a string
34863 that only indicates the type of the object, and does not list its
34864 contents. Assignment to a non-leaf variable object is not allowed.
34866 A frontend does not need to read the values of all variable objects each time
34867 the program stops. Instead, MI provides an update command that lists all
34868 variable objects whose values has changed since the last update
34869 operation. This considerably reduces the amount of data that must
34870 be transferred to the frontend. As noted above, children variable
34871 objects are created on demand, and only leaf variable objects have a
34872 real value. As result, gdb will read target memory only for leaf
34873 variables that frontend has created.
34875 The automatic update is not always desirable. For example, a frontend
34876 might want to keep a value of some expression for future reference,
34877 and never update it. For another example, fetching memory is
34878 relatively slow for embedded targets, so a frontend might want
34879 to disable automatic update for the variables that are either not
34880 visible on the screen, or ``closed''. This is possible using so
34881 called ``frozen variable objects''. Such variable objects are never
34882 implicitly updated.
34884 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
34885 fixed variable object, the expression is parsed when the variable
34886 object is created, including associating identifiers to specific
34887 variables. The meaning of expression never changes. For a floating
34888 variable object the values of variables whose names appear in the
34889 expressions are re-evaluated every time in the context of the current
34890 frame. Consider this example:
34895 struct work_state state;
34902 If a fixed variable object for the @code{state} variable is created in
34903 this function, and we enter the recursive call, the variable
34904 object will report the value of @code{state} in the top-level
34905 @code{do_work} invocation. On the other hand, a floating variable
34906 object will report the value of @code{state} in the current frame.
34908 If an expression specified when creating a fixed variable object
34909 refers to a local variable, the variable object becomes bound to the
34910 thread and frame in which the variable object is created. When such
34911 variable object is updated, @value{GDBN} makes sure that the
34912 thread/frame combination the variable object is bound to still exists,
34913 and re-evaluates the variable object in context of that thread/frame.
34915 The following is the complete set of @sc{gdb/mi} operations defined to
34916 access this functionality:
34918 @multitable @columnfractions .4 .6
34919 @item @strong{Operation}
34920 @tab @strong{Description}
34922 @item @code{-enable-pretty-printing}
34923 @tab enable Python-based pretty-printing
34924 @item @code{-var-create}
34925 @tab create a variable object
34926 @item @code{-var-delete}
34927 @tab delete the variable object and/or its children
34928 @item @code{-var-set-format}
34929 @tab set the display format of this variable
34930 @item @code{-var-show-format}
34931 @tab show the display format of this variable
34932 @item @code{-var-info-num-children}
34933 @tab tells how many children this object has
34934 @item @code{-var-list-children}
34935 @tab return a list of the object's children
34936 @item @code{-var-info-type}
34937 @tab show the type of this variable object
34938 @item @code{-var-info-expression}
34939 @tab print parent-relative expression that this variable object represents
34940 @item @code{-var-info-path-expression}
34941 @tab print full expression that this variable object represents
34942 @item @code{-var-show-attributes}
34943 @tab is this variable editable? does it exist here?
34944 @item @code{-var-evaluate-expression}
34945 @tab get the value of this variable
34946 @item @code{-var-assign}
34947 @tab set the value of this variable
34948 @item @code{-var-update}
34949 @tab update the variable and its children
34950 @item @code{-var-set-frozen}
34951 @tab set frozenness attribute
34952 @item @code{-var-set-update-range}
34953 @tab set range of children to display on update
34956 In the next subsection we describe each operation in detail and suggest
34957 how it can be used.
34959 @subheading Description And Use of Operations on Variable Objects
34961 @findex -enable-pretty-printing
34962 @subheading The @code{-enable-pretty-printing} Command
34965 -enable-pretty-printing
34968 @value{GDBN} allows Python-based visualizers to affect the output of the
34969 MI variable object commands. However, because there was no way to
34970 implement this in a fully backward-compatible way, a front end must
34971 request that this functionality be enabled.
34973 Once enabled, this feature cannot be disabled.
34975 Note that if Python support has not been compiled into @value{GDBN},
34976 this command will still succeed (and do nothing).
34978 @findex -var-create
34979 @subheading The @code{-var-create} Command
34981 @subsubheading Synopsis
34984 -var-create @{@var{name} | "-"@}
34985 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
34988 This operation creates a variable object, which allows the monitoring of
34989 a variable, the result of an expression, a memory cell or a CPU
34992 The @var{name} parameter is the string by which the object can be
34993 referenced. It must be unique. If @samp{-} is specified, the varobj
34994 system will generate a string ``varNNNNNN'' automatically. It will be
34995 unique provided that one does not specify @var{name} of that format.
34996 The command fails if a duplicate name is found.
34998 The frame under which the expression should be evaluated can be
34999 specified by @var{frame-addr}. A @samp{*} indicates that the current
35000 frame should be used. A @samp{@@} indicates that a floating variable
35001 object must be created.
35003 @var{expression} is any expression valid on the current language set (must not
35004 begin with a @samp{*}), or one of the following:
35008 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
35011 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
35014 @samp{$@var{regname}} --- a CPU register name
35017 @cindex dynamic varobj
35018 A varobj's contents may be provided by a Python-based pretty-printer. In this
35019 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
35020 have slightly different semantics in some cases. If the
35021 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
35022 will never create a dynamic varobj. This ensures backward
35023 compatibility for existing clients.
35025 @subsubheading Result
35027 This operation returns attributes of the newly-created varobj. These
35032 The name of the varobj.
35035 The number of children of the varobj. This number is not necessarily
35036 reliable for a dynamic varobj. Instead, you must examine the
35037 @samp{has_more} attribute.
35040 The varobj's scalar value. For a varobj whose type is some sort of
35041 aggregate (e.g., a @code{struct}), this value will not be interesting.
35042 For a dynamic varobj, this value comes directly from the Python
35043 pretty-printer object's @code{to_string} method.
35046 The varobj's type. This is a string representation of the type, as
35047 would be printed by the @value{GDBN} CLI. If @samp{print object}
35048 (@pxref{Print Settings, set print object}) is set to @code{on}, the
35049 @emph{actual} (derived) type of the object is shown rather than the
35050 @emph{declared} one.
35053 If a variable object is bound to a specific thread, then this is the
35054 thread's global identifier.
35057 For a dynamic varobj, this indicates whether there appear to be any
35058 children available. For a non-dynamic varobj, this will be 0.
35061 This attribute will be present and have the value @samp{1} if the
35062 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
35063 then this attribute will not be present.
35066 A dynamic varobj can supply a display hint to the front end. The
35067 value comes directly from the Python pretty-printer object's
35068 @code{display_hint} method. @xref{Pretty Printing API}.
35071 Typical output will look like this:
35074 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
35075 has_more="@var{has_more}"
35079 @findex -var-delete
35080 @subheading The @code{-var-delete} Command
35082 @subsubheading Synopsis
35085 -var-delete [ -c ] @var{name}
35088 Deletes a previously created variable object and all of its children.
35089 With the @samp{-c} option, just deletes the children.
35091 Returns an error if the object @var{name} is not found.
35094 @findex -var-set-format
35095 @subheading The @code{-var-set-format} Command
35097 @subsubheading Synopsis
35100 -var-set-format @var{name} @var{format-spec}
35103 Sets the output format for the value of the object @var{name} to be
35106 @anchor{-var-set-format}
35107 The syntax for the @var{format-spec} is as follows:
35110 @var{format-spec} @expansion{}
35111 @{binary | decimal | hexadecimal | octal | natural | zero-hexadecimal@}
35114 The natural format is the default format choosen automatically
35115 based on the variable type (like decimal for an @code{int}, hex
35116 for pointers, etc.).
35118 The zero-hexadecimal format has a representation similar to hexadecimal
35119 but with padding zeroes to the left of the value. For example, a 32-bit
35120 hexadecimal value of 0x1234 would be represented as 0x00001234 in the
35121 zero-hexadecimal format.
35123 For a variable with children, the format is set only on the
35124 variable itself, and the children are not affected.
35126 @findex -var-show-format
35127 @subheading The @code{-var-show-format} Command
35129 @subsubheading Synopsis
35132 -var-show-format @var{name}
35135 Returns the format used to display the value of the object @var{name}.
35138 @var{format} @expansion{}
35143 @findex -var-info-num-children
35144 @subheading The @code{-var-info-num-children} Command
35146 @subsubheading Synopsis
35149 -var-info-num-children @var{name}
35152 Returns the number of children of a variable object @var{name}:
35158 Note that this number is not completely reliable for a dynamic varobj.
35159 It will return the current number of children, but more children may
35163 @findex -var-list-children
35164 @subheading The @code{-var-list-children} Command
35166 @subsubheading Synopsis
35169 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
35171 @anchor{-var-list-children}
35173 Return a list of the children of the specified variable object and
35174 create variable objects for them, if they do not already exist. With
35175 a single argument or if @var{print-values} has a value of 0 or
35176 @code{--no-values}, print only the names of the variables; if
35177 @var{print-values} is 1 or @code{--all-values}, also print their
35178 values; and if it is 2 or @code{--simple-values} print the name and
35179 value for simple data types and just the name for arrays, structures
35182 @var{from} and @var{to}, if specified, indicate the range of children
35183 to report. If @var{from} or @var{to} is less than zero, the range is
35184 reset and all children will be reported. Otherwise, children starting
35185 at @var{from} (zero-based) and up to and excluding @var{to} will be
35188 If a child range is requested, it will only affect the current call to
35189 @code{-var-list-children}, but not future calls to @code{-var-update}.
35190 For this, you must instead use @code{-var-set-update-range}. The
35191 intent of this approach is to enable a front end to implement any
35192 update approach it likes; for example, scrolling a view may cause the
35193 front end to request more children with @code{-var-list-children}, and
35194 then the front end could call @code{-var-set-update-range} with a
35195 different range to ensure that future updates are restricted to just
35198 For each child the following results are returned:
35203 Name of the variable object created for this child.
35206 The expression to be shown to the user by the front end to designate this child.
35207 For example this may be the name of a structure member.
35209 For a dynamic varobj, this value cannot be used to form an
35210 expression. There is no way to do this at all with a dynamic varobj.
35212 For C/C@t{++} structures there are several pseudo children returned to
35213 designate access qualifiers. For these pseudo children @var{exp} is
35214 @samp{public}, @samp{private}, or @samp{protected}. In this case the
35215 type and value are not present.
35217 A dynamic varobj will not report the access qualifying
35218 pseudo-children, regardless of the language. This information is not
35219 available at all with a dynamic varobj.
35222 Number of children this child has. For a dynamic varobj, this will be
35226 The type of the child. If @samp{print object}
35227 (@pxref{Print Settings, set print object}) is set to @code{on}, the
35228 @emph{actual} (derived) type of the object is shown rather than the
35229 @emph{declared} one.
35232 If values were requested, this is the value.
35235 If this variable object is associated with a thread, this is the
35236 thread's global thread id. Otherwise this result is not present.
35239 If the variable object is frozen, this variable will be present with a value of 1.
35242 A dynamic varobj can supply a display hint to the front end. The
35243 value comes directly from the Python pretty-printer object's
35244 @code{display_hint} method. @xref{Pretty Printing API}.
35247 This attribute will be present and have the value @samp{1} if the
35248 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
35249 then this attribute will not be present.
35253 The result may have its own attributes:
35257 A dynamic varobj can supply a display hint to the front end. The
35258 value comes directly from the Python pretty-printer object's
35259 @code{display_hint} method. @xref{Pretty Printing API}.
35262 This is an integer attribute which is nonzero if there are children
35263 remaining after the end of the selected range.
35266 @subsubheading Example
35270 -var-list-children n
35271 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
35272 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
35274 -var-list-children --all-values n
35275 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
35276 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
35280 @findex -var-info-type
35281 @subheading The @code{-var-info-type} Command
35283 @subsubheading Synopsis
35286 -var-info-type @var{name}
35289 Returns the type of the specified variable @var{name}. The type is
35290 returned as a string in the same format as it is output by the
35294 type=@var{typename}
35298 @findex -var-info-expression
35299 @subheading The @code{-var-info-expression} Command
35301 @subsubheading Synopsis
35304 -var-info-expression @var{name}
35307 Returns a string that is suitable for presenting this
35308 variable object in user interface. The string is generally
35309 not valid expression in the current language, and cannot be evaluated.
35311 For example, if @code{a} is an array, and variable object
35312 @code{A} was created for @code{a}, then we'll get this output:
35315 (gdb) -var-info-expression A.1
35316 ^done,lang="C",exp="1"
35320 Here, the value of @code{lang} is the language name, which can be
35321 found in @ref{Supported Languages}.
35323 Note that the output of the @code{-var-list-children} command also
35324 includes those expressions, so the @code{-var-info-expression} command
35327 @findex -var-info-path-expression
35328 @subheading The @code{-var-info-path-expression} Command
35330 @subsubheading Synopsis
35333 -var-info-path-expression @var{name}
35336 Returns an expression that can be evaluated in the current
35337 context and will yield the same value that a variable object has.
35338 Compare this with the @code{-var-info-expression} command, which
35339 result can be used only for UI presentation. Typical use of
35340 the @code{-var-info-path-expression} command is creating a
35341 watchpoint from a variable object.
35343 This command is currently not valid for children of a dynamic varobj,
35344 and will give an error when invoked on one.
35346 For example, suppose @code{C} is a C@t{++} class, derived from class
35347 @code{Base}, and that the @code{Base} class has a member called
35348 @code{m_size}. Assume a variable @code{c} is has the type of
35349 @code{C} and a variable object @code{C} was created for variable
35350 @code{c}. Then, we'll get this output:
35352 (gdb) -var-info-path-expression C.Base.public.m_size
35353 ^done,path_expr=((Base)c).m_size)
35356 @findex -var-show-attributes
35357 @subheading The @code{-var-show-attributes} Command
35359 @subsubheading Synopsis
35362 -var-show-attributes @var{name}
35365 List attributes of the specified variable object @var{name}:
35368 status=@var{attr} [ ( ,@var{attr} )* ]
35372 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
35374 @findex -var-evaluate-expression
35375 @subheading The @code{-var-evaluate-expression} Command
35377 @subsubheading Synopsis
35380 -var-evaluate-expression [-f @var{format-spec}] @var{name}
35383 Evaluates the expression that is represented by the specified variable
35384 object and returns its value as a string. The format of the string
35385 can be specified with the @samp{-f} option. The possible values of
35386 this option are the same as for @code{-var-set-format}
35387 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
35388 the current display format will be used. The current display format
35389 can be changed using the @code{-var-set-format} command.
35395 Note that one must invoke @code{-var-list-children} for a variable
35396 before the value of a child variable can be evaluated.
35398 @findex -var-assign
35399 @subheading The @code{-var-assign} Command
35401 @subsubheading Synopsis
35404 -var-assign @var{name} @var{expression}
35407 Assigns the value of @var{expression} to the variable object specified
35408 by @var{name}. The object must be @samp{editable}. If the variable's
35409 value is altered by the assign, the variable will show up in any
35410 subsequent @code{-var-update} list.
35412 @subsubheading Example
35420 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
35424 @findex -var-update
35425 @subheading The @code{-var-update} Command
35427 @subsubheading Synopsis
35430 -var-update [@var{print-values}] @{@var{name} | "*"@}
35433 Reevaluate the expressions corresponding to the variable object
35434 @var{name} and all its direct and indirect children, and return the
35435 list of variable objects whose values have changed; @var{name} must
35436 be a root variable object. Here, ``changed'' means that the result of
35437 @code{-var-evaluate-expression} before and after the
35438 @code{-var-update} is different. If @samp{*} is used as the variable
35439 object names, all existing variable objects are updated, except
35440 for frozen ones (@pxref{-var-set-frozen}). The option
35441 @var{print-values} determines whether both names and values, or just
35442 names are printed. The possible values of this option are the same
35443 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
35444 recommended to use the @samp{--all-values} option, to reduce the
35445 number of MI commands needed on each program stop.
35447 With the @samp{*} parameter, if a variable object is bound to a
35448 currently running thread, it will not be updated, without any
35451 If @code{-var-set-update-range} was previously used on a varobj, then
35452 only the selected range of children will be reported.
35454 @code{-var-update} reports all the changed varobjs in a tuple named
35457 Each item in the change list is itself a tuple holding:
35461 The name of the varobj.
35464 If values were requested for this update, then this field will be
35465 present and will hold the value of the varobj.
35468 @anchor{-var-update}
35469 This field is a string which may take one of three values:
35473 The variable object's current value is valid.
35476 The variable object does not currently hold a valid value but it may
35477 hold one in the future if its associated expression comes back into
35481 The variable object no longer holds a valid value.
35482 This can occur when the executable file being debugged has changed,
35483 either through recompilation or by using the @value{GDBN} @code{file}
35484 command. The front end should normally choose to delete these variable
35488 In the future new values may be added to this list so the front should
35489 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{gdb/mi} Development and Front Ends}.
35492 This is only present if the varobj is still valid. If the type
35493 changed, then this will be the string @samp{true}; otherwise it will
35496 When a varobj's type changes, its children are also likely to have
35497 become incorrect. Therefore, the varobj's children are automatically
35498 deleted when this attribute is @samp{true}. Also, the varobj's update
35499 range, when set using the @code{-var-set-update-range} command, is
35503 If the varobj's type changed, then this field will be present and will
35506 @item new_num_children
35507 For a dynamic varobj, if the number of children changed, or if the
35508 type changed, this will be the new number of children.
35510 The @samp{numchild} field in other varobj responses is generally not
35511 valid for a dynamic varobj -- it will show the number of children that
35512 @value{GDBN} knows about, but because dynamic varobjs lazily
35513 instantiate their children, this will not reflect the number of
35514 children which may be available.
35516 The @samp{new_num_children} attribute only reports changes to the
35517 number of children known by @value{GDBN}. This is the only way to
35518 detect whether an update has removed children (which necessarily can
35519 only happen at the end of the update range).
35522 The display hint, if any.
35525 This is an integer value, which will be 1 if there are more children
35526 available outside the varobj's update range.
35529 This attribute will be present and have the value @samp{1} if the
35530 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
35531 then this attribute will not be present.
35534 If new children were added to a dynamic varobj within the selected
35535 update range (as set by @code{-var-set-update-range}), then they will
35536 be listed in this attribute.
35539 @subsubheading Example
35546 -var-update --all-values var1
35547 ^done,changelist=[@{name="var1",value="3",in_scope="true",
35548 type_changed="false"@}]
35552 @findex -var-set-frozen
35553 @anchor{-var-set-frozen}
35554 @subheading The @code{-var-set-frozen} Command
35556 @subsubheading Synopsis
35559 -var-set-frozen @var{name} @var{flag}
35562 Set the frozenness flag on the variable object @var{name}. The
35563 @var{flag} parameter should be either @samp{1} to make the variable
35564 frozen or @samp{0} to make it unfrozen. If a variable object is
35565 frozen, then neither itself, nor any of its children, are
35566 implicitly updated by @code{-var-update} of
35567 a parent variable or by @code{-var-update *}. Only
35568 @code{-var-update} of the variable itself will update its value and
35569 values of its children. After a variable object is unfrozen, it is
35570 implicitly updated by all subsequent @code{-var-update} operations.
35571 Unfreezing a variable does not update it, only subsequent
35572 @code{-var-update} does.
35574 @subsubheading Example
35578 -var-set-frozen V 1
35583 @findex -var-set-update-range
35584 @anchor{-var-set-update-range}
35585 @subheading The @code{-var-set-update-range} command
35587 @subsubheading Synopsis
35590 -var-set-update-range @var{name} @var{from} @var{to}
35593 Set the range of children to be returned by future invocations of
35594 @code{-var-update}.
35596 @var{from} and @var{to} indicate the range of children to report. If
35597 @var{from} or @var{to} is less than zero, the range is reset and all
35598 children will be reported. Otherwise, children starting at @var{from}
35599 (zero-based) and up to and excluding @var{to} will be reported.
35601 @subsubheading Example
35605 -var-set-update-range V 1 2
35609 @findex -var-set-visualizer
35610 @anchor{-var-set-visualizer}
35611 @subheading The @code{-var-set-visualizer} command
35613 @subsubheading Synopsis
35616 -var-set-visualizer @var{name} @var{visualizer}
35619 Set a visualizer for the variable object @var{name}.
35621 @var{visualizer} is the visualizer to use. The special value
35622 @samp{None} means to disable any visualizer in use.
35624 If not @samp{None}, @var{visualizer} must be a Python expression.
35625 This expression must evaluate to a callable object which accepts a
35626 single argument. @value{GDBN} will call this object with the value of
35627 the varobj @var{name} as an argument (this is done so that the same
35628 Python pretty-printing code can be used for both the CLI and MI).
35629 When called, this object must return an object which conforms to the
35630 pretty-printing interface (@pxref{Pretty Printing API}).
35632 The pre-defined function @code{gdb.default_visualizer} may be used to
35633 select a visualizer by following the built-in process
35634 (@pxref{Selecting Pretty-Printers}). This is done automatically when
35635 a varobj is created, and so ordinarily is not needed.
35637 This feature is only available if Python support is enabled. The MI
35638 command @code{-list-features} (@pxref{GDB/MI Support Commands})
35639 can be used to check this.
35641 @subsubheading Example
35643 Resetting the visualizer:
35647 -var-set-visualizer V None
35651 Reselecting the default (type-based) visualizer:
35655 -var-set-visualizer V gdb.default_visualizer
35659 Suppose @code{SomeClass} is a visualizer class. A lambda expression
35660 can be used to instantiate this class for a varobj:
35664 -var-set-visualizer V "lambda val: SomeClass()"
35668 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
35669 @node GDB/MI Data Manipulation
35670 @section @sc{gdb/mi} Data Manipulation
35672 @cindex data manipulation, in @sc{gdb/mi}
35673 @cindex @sc{gdb/mi}, data manipulation
35674 This section describes the @sc{gdb/mi} commands that manipulate data:
35675 examine memory and registers, evaluate expressions, etc.
35677 For details about what an addressable memory unit is,
35678 @pxref{addressable memory unit}.
35680 @c REMOVED FROM THE INTERFACE.
35681 @c @subheading -data-assign
35682 @c Change the value of a program variable. Plenty of side effects.
35683 @c @subsubheading GDB Command
35685 @c @subsubheading Example
35688 @findex -data-disassemble
35689 @subheading The @code{-data-disassemble} Command
35691 @subsubheading Synopsis
35695 ( -s @var{start-addr} -e @var{end-addr}
35697 | -f @var{filename} -l @var{linenum} [ -n @var{lines} ] )
35698 [ --opcodes @var{opcodes-mode} ]
35707 @item @var{start-addr}
35708 is the beginning address (or @code{$pc})
35709 @item @var{end-addr}
35712 is an address anywhere within (or the name of) the function to
35713 disassemble. If an address is specified, the whole function
35714 surrounding that address will be disassembled. If a name is
35715 specified, the whole function with that name will be disassembled.
35716 @item @var{filename}
35717 is the name of the file to disassemble
35718 @item @var{linenum}
35719 is the line number to disassemble around
35721 is the number of disassembly lines to be produced. If it is -1,
35722 the whole function will be disassembled, in case no @var{end-addr} is
35723 specified. If @var{end-addr} is specified as a non-zero value, and
35724 @var{lines} is lower than the number of disassembly lines between
35725 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
35726 displayed; if @var{lines} is higher than the number of lines between
35727 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
35729 @item @var{opcodes-mode}
35730 can only be used with @var{mode} 0, and should be one of the following:
35733 no opcode information will be included in the result.
35736 opcodes will be included in the result, the opcodes will be formatted
35737 as for @kbd{disassemble /b}.
35740 opcodes will be included in the result, the opcodes will be formatted
35741 as for @kbd{disassemble /r}.
35744 the use of @var{mode} is deprecated in favour of using the
35745 @code{--opcodes} and @code{--source} options. When no @var{mode} is
35746 given, @var{mode} 0 will be assumed. However, the @var{mode} is still
35747 available for backward compatibility. The @var{mode} should be one of:
35750 @emph{disassembly only}, this is the default mode if no mode is
35754 @emph{mixed source and disassembly (deprecated)}, it is not possible
35755 to recreate this mode using @code{--opcodes} and @code{--source}
35759 @emph{disassembly with raw opcodes}, this mode is equivalent to using
35760 @var{mode} 0 and passing @code{--opcodes bytes} to the command.
35763 @emph{mixed source and disassembly with raw opcodes (deprecated)}, it
35764 is not possible to recreate this mode using @code{--opcodes} and
35765 @code{--source} options.
35768 @emph{mixed source and disassembly}, this mode is equivalent to using
35769 @var{mode} 0 and passing @code{--source} to the command.
35772 @emph{mixed source and disassembly with raw opcodes}, this mode is
35773 equivalent to using @var{mode} 0 and passing @code{--opcodes bytes}
35774 and @code{--source} to the command.
35776 Modes 1 and 3 are deprecated. The output is ``source centric''
35777 which hasn't proved useful in practice.
35778 @xref{Machine Code}, for a discussion of the difference between
35779 @code{/m} and @code{/s} output of the @code{disassemble} command.
35782 The @code{--source} can only be used with @var{mode} 0. Passing this
35783 option will include the source code in the disassembly result as if
35784 @var{mode} 4 or 5 had been used.
35786 @subsubheading Result
35788 The result of the @code{-data-disassemble} command will be a list named
35789 @samp{asm_insns}, the contents of this list depend on the options used
35790 with the @code{-data-disassemble} command.
35792 For modes 0 and 2, and when the @code{--source} option is not used, the
35793 @samp{asm_insns} list contains tuples with the following fields:
35797 The address at which this instruction was disassembled.
35800 The name of the function this instruction is within.
35803 The decimal offset in bytes from the start of @samp{func-name}.
35806 The text disassembly for this @samp{address}.
35809 This field is only present for modes 2, 3 and 5, or when the
35810 @code{--opcodes} option @samp{bytes} or @samp{display} is used. This
35811 contains the raw opcode bytes for the @samp{inst} field.
35813 When the @samp{--opcodes} option is not passed to
35814 @code{-data-disassemble}, or the @samp{bytes} value is passed to
35815 @samp{--opcodes}, then the bytes are formatted as a series of single
35816 bytes, in hex, in ascending address order, with a single space between
35817 each byte. This format is equivalent to the @samp{/b} option being
35818 used with the @kbd{disassemble} command
35819 (@pxref{disassemble,,@kbd{disassemble}}).
35821 When @samp{--opcodes} is passed the value @samp{display} then the bytes
35822 are formatted in the natural instruction display order. This means
35823 multiple bytes can be grouped together, and the bytes might be
35824 byte-swapped. This format is equivalent to the @samp{/r} option being
35825 used with the @kbd{disassemble} command.
35828 For modes 1, 3, 4 and 5, or when the @code{--source} option is used, the
35829 @samp{asm_insns} list contains tuples named @samp{src_and_asm_line},
35830 each of which has the following fields:
35834 The line number within @samp{file}.
35837 The file name from the compilation unit. This might be an absolute
35838 file name or a relative file name depending on the compile command
35842 Absolute file name of @samp{file}. It is converted to a canonical form
35843 using the source file search path
35844 (@pxref{Source Path, ,Specifying Source Directories})
35845 and after resolving all the symbolic links.
35847 If the source file is not found this field will contain the path as
35848 present in the debug information.
35850 @item line_asm_insn
35851 This is a list of tuples containing the disassembly for @samp{line} in
35852 @samp{file}. The fields of each tuple are the same as for
35853 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
35854 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
35859 Note that whatever included in the @samp{inst} field, is not
35860 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
35863 @subsubheading @value{GDBN} Command
35865 The corresponding @value{GDBN} command is @samp{disassemble}.
35867 @subsubheading Example
35869 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
35873 -data-disassemble -s $pc -e "$pc + 20" -- 0
35876 @{address="0x000107c0",func-name="main",offset="4",
35877 inst="mov 2, %o0"@},
35878 @{address="0x000107c4",func-name="main",offset="8",
35879 inst="sethi %hi(0x11800), %o2"@},
35880 @{address="0x000107c8",func-name="main",offset="12",
35881 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
35882 @{address="0x000107cc",func-name="main",offset="16",
35883 inst="sethi %hi(0x11800), %o2"@},
35884 @{address="0x000107d0",func-name="main",offset="20",
35885 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
35889 Disassemble the whole @code{main} function. Line 32 is part of
35893 -data-disassemble -f basics.c -l 32 -- 0
35895 @{address="0x000107bc",func-name="main",offset="0",
35896 inst="save %sp, -112, %sp"@},
35897 @{address="0x000107c0",func-name="main",offset="4",
35898 inst="mov 2, %o0"@},
35899 @{address="0x000107c4",func-name="main",offset="8",
35900 inst="sethi %hi(0x11800), %o2"@},
35902 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
35903 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
35907 Disassemble 3 instructions from the start of @code{main}:
35911 -data-disassemble -f basics.c -l 32 -n 3 -- 0
35913 @{address="0x000107bc",func-name="main",offset="0",
35914 inst="save %sp, -112, %sp"@},
35915 @{address="0x000107c0",func-name="main",offset="4",
35916 inst="mov 2, %o0"@},
35917 @{address="0x000107c4",func-name="main",offset="8",
35918 inst="sethi %hi(0x11800), %o2"@}]
35922 Disassemble 3 instructions from the start of @code{main} in mixed mode:
35926 -data-disassemble -f basics.c -l 32 -n 3 -- 1
35928 src_and_asm_line=@{line="31",
35929 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
35930 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
35931 line_asm_insn=[@{address="0x000107bc",
35932 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
35933 src_and_asm_line=@{line="32",
35934 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
35935 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
35936 line_asm_insn=[@{address="0x000107c0",
35937 func-name="main",offset="4",inst="mov 2, %o0"@},
35938 @{address="0x000107c4",func-name="main",offset="8",
35939 inst="sethi %hi(0x11800), %o2"@}]@}]
35944 @findex -data-evaluate-expression
35945 @subheading The @code{-data-evaluate-expression} Command
35947 @subsubheading Synopsis
35950 -data-evaluate-expression @var{expr}
35953 Evaluate @var{expr} as an expression. The expression could contain an
35954 inferior function call. The function call will execute synchronously.
35955 If the expression contains spaces, it must be enclosed in double quotes.
35957 @subsubheading @value{GDBN} Command
35959 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
35960 @samp{call}. In @code{gdbtk} only, there's a corresponding
35961 @samp{gdb_eval} command.
35963 @subsubheading Example
35965 In the following example, the numbers that precede the commands are the
35966 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
35967 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
35971 211-data-evaluate-expression A
35974 311-data-evaluate-expression &A
35975 311^done,value="0xefffeb7c"
35977 411-data-evaluate-expression A+3
35980 511-data-evaluate-expression "A + 3"
35986 @findex -data-list-changed-registers
35987 @subheading The @code{-data-list-changed-registers} Command
35989 @subsubheading Synopsis
35992 -data-list-changed-registers
35995 Display a list of the registers that have changed.
35997 @subsubheading @value{GDBN} Command
35999 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
36000 has the corresponding command @samp{gdb_changed_register_list}.
36002 @subsubheading Example
36004 On a PPC MBX board:
36012 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
36013 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
36014 line="5",arch="powerpc"@}
36016 -data-list-changed-registers
36017 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
36018 "10","11","13","14","15","16","17","18","19","20","21","22","23",
36019 "24","25","26","27","28","30","31","64","65","66","67","69"]
36024 @findex -data-list-register-names
36025 @subheading The @code{-data-list-register-names} Command
36027 @subsubheading Synopsis
36030 -data-list-register-names [ ( @var{regno} )+ ]
36033 Show a list of register names for the current target. If no arguments
36034 are given, it shows a list of the names of all the registers. If
36035 integer numbers are given as arguments, it will print a list of the
36036 names of the registers corresponding to the arguments. To ensure
36037 consistency between a register name and its number, the output list may
36038 include empty register names.
36040 @subsubheading @value{GDBN} Command
36042 @value{GDBN} does not have a command which corresponds to
36043 @samp{-data-list-register-names}. In @code{gdbtk} there is a
36044 corresponding command @samp{gdb_regnames}.
36046 @subsubheading Example
36048 For the PPC MBX board:
36051 -data-list-register-names
36052 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
36053 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
36054 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
36055 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
36056 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
36057 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
36058 "", "pc","ps","cr","lr","ctr","xer"]
36060 -data-list-register-names 1 2 3
36061 ^done,register-names=["r1","r2","r3"]
36065 @findex -data-list-register-values
36066 @subheading The @code{-data-list-register-values} Command
36068 @subsubheading Synopsis
36071 -data-list-register-values
36072 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
36075 Display the registers' contents. The format according to which the
36076 registers' contents are to be returned is given by @var{fmt}, followed
36077 by an optional list of numbers specifying the registers to display. A
36078 missing list of numbers indicates that the contents of all the
36079 registers must be returned. The @code{--skip-unavailable} option
36080 indicates that only the available registers are to be returned.
36082 Allowed formats for @var{fmt} are:
36099 @subsubheading @value{GDBN} Command
36101 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
36102 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
36104 @subsubheading Example
36106 For a PPC MBX board (note: line breaks are for readability only, they
36107 don't appear in the actual output):
36111 -data-list-register-values r 64 65
36112 ^done,register-values=[@{number="64",value="0xfe00a300"@},
36113 @{number="65",value="0x00029002"@}]
36115 -data-list-register-values x
36116 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
36117 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
36118 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
36119 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
36120 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
36121 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
36122 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
36123 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
36124 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
36125 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
36126 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
36127 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
36128 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
36129 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
36130 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
36131 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
36132 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
36133 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
36134 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
36135 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
36136 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
36137 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
36138 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
36139 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
36140 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
36141 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
36142 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
36143 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
36144 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
36145 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
36146 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
36147 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
36148 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
36149 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
36150 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
36151 @{number="69",value="0x20002b03"@}]
36156 @findex -data-read-memory
36157 @subheading The @code{-data-read-memory} Command
36159 This command is deprecated, use @code{-data-read-memory-bytes} instead.
36161 @subsubheading Synopsis
36164 -data-read-memory [ -o @var{byte-offset} ]
36165 @var{address} @var{word-format} @var{word-size}
36166 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
36173 @item @var{address}
36174 An expression specifying the address of the first memory word to be
36175 read. Complex expressions containing embedded white space should be
36176 quoted using the C convention.
36178 @item @var{word-format}
36179 The format to be used to print the memory words. The notation is the
36180 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
36183 @item @var{word-size}
36184 The size of each memory word in bytes.
36186 @item @var{nr-rows}
36187 The number of rows in the output table.
36189 @item @var{nr-cols}
36190 The number of columns in the output table.
36193 If present, indicates that each row should include an @sc{ascii} dump. The
36194 value of @var{aschar} is used as a padding character when a byte is not a
36195 member of the printable @sc{ascii} character set (printable @sc{ascii}
36196 characters are those whose code is between 32 and 126, inclusively).
36198 @item @var{byte-offset}
36199 An offset to add to the @var{address} before fetching memory.
36202 This command displays memory contents as a table of @var{nr-rows} by
36203 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
36204 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
36205 (returned as @samp{total-bytes}). Should less than the requested number
36206 of bytes be returned by the target, the missing words are identified
36207 using @samp{N/A}. The number of bytes read from the target is returned
36208 in @samp{nr-bytes} and the starting address used to read memory in
36211 The address of the next/previous row or page is available in
36212 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
36215 @subsubheading @value{GDBN} Command
36217 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
36218 @samp{gdb_get_mem} memory read command.
36220 @subsubheading Example
36222 Read six bytes of memory starting at @code{bytes+6} but then offset by
36223 @code{-6} bytes. Format as three rows of two columns. One byte per
36224 word. Display each word in hex.
36228 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
36229 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
36230 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
36231 prev-page="0x0000138a",memory=[
36232 @{addr="0x00001390",data=["0x00","0x01"]@},
36233 @{addr="0x00001392",data=["0x02","0x03"]@},
36234 @{addr="0x00001394",data=["0x04","0x05"]@}]
36238 Read two bytes of memory starting at address @code{shorts + 64} and
36239 display as a single word formatted in decimal.
36243 5-data-read-memory shorts+64 d 2 1 1
36244 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
36245 next-row="0x00001512",prev-row="0x0000150e",
36246 next-page="0x00001512",prev-page="0x0000150e",memory=[
36247 @{addr="0x00001510",data=["128"]@}]
36251 Read thirty two bytes of memory starting at @code{bytes+16} and format
36252 as eight rows of four columns. Include a string encoding with @samp{x}
36253 used as the non-printable character.
36257 4-data-read-memory bytes+16 x 1 8 4 x
36258 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
36259 next-row="0x000013c0",prev-row="0x0000139c",
36260 next-page="0x000013c0",prev-page="0x00001380",memory=[
36261 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
36262 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
36263 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
36264 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
36265 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
36266 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
36267 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
36268 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
36272 @findex -data-read-memory-bytes
36273 @subheading The @code{-data-read-memory-bytes} Command
36275 @subsubheading Synopsis
36278 -data-read-memory-bytes [ -o @var{offset} ]
36279 @var{address} @var{count}
36286 @item @var{address}
36287 An expression specifying the address of the first addressable memory unit
36288 to be read. Complex expressions containing embedded white space should be
36289 quoted using the C convention.
36292 The number of addressable memory units to read. This should be an integer
36296 The offset relative to @var{address} at which to start reading. This
36297 should be an integer literal. This option is provided so that a frontend
36298 is not required to first evaluate address and then perform address
36299 arithmetics itself.
36303 This command attempts to read all accessible memory regions in the
36304 specified range. First, all regions marked as unreadable in the memory
36305 map (if one is defined) will be skipped. @xref{Memory Region
36306 Attributes}. Second, @value{GDBN} will attempt to read the remaining
36307 regions. For each one, if reading full region results in an errors,
36308 @value{GDBN} will try to read a subset of the region.
36310 In general, every single memory unit in the region may be readable or not,
36311 and the only way to read every readable unit is to try a read at
36312 every address, which is not practical. Therefore, @value{GDBN} will
36313 attempt to read all accessible memory units at either beginning or the end
36314 of the region, using a binary division scheme. This heuristic works
36315 well for reading across a memory map boundary. Note that if a region
36316 has a readable range that is neither at the beginning or the end,
36317 @value{GDBN} will not read it.
36319 The result record (@pxref{GDB/MI Result Records}) that is output of
36320 the command includes a field named @samp{memory} whose content is a
36321 list of tuples. Each tuple represent a successfully read memory block
36322 and has the following fields:
36326 The start address of the memory block, as hexadecimal literal.
36329 The end address of the memory block, as hexadecimal literal.
36332 The offset of the memory block, as hexadecimal literal, relative to
36333 the start address passed to @code{-data-read-memory-bytes}.
36336 The contents of the memory block, in hex.
36342 @subsubheading @value{GDBN} Command
36344 The corresponding @value{GDBN} command is @samp{x}.
36346 @subsubheading Example
36350 -data-read-memory-bytes &a 10
36351 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
36353 contents="01000000020000000300"@}]
36358 @findex -data-write-memory-bytes
36359 @subheading The @code{-data-write-memory-bytes} Command
36361 @subsubheading Synopsis
36364 -data-write-memory-bytes @var{address} @var{contents}
36365 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
36372 @item @var{address}
36373 An expression specifying the address of the first addressable memory unit
36374 to be written. Complex expressions containing embedded white space should
36375 be quoted using the C convention.
36377 @item @var{contents}
36378 The hex-encoded data to write. It is an error if @var{contents} does
36379 not represent an integral number of addressable memory units.
36382 Optional argument indicating the number of addressable memory units to be
36383 written. If @var{count} is greater than @var{contents}' length,
36384 @value{GDBN} will repeatedly write @var{contents} until it fills
36385 @var{count} memory units.
36389 @subsubheading @value{GDBN} Command
36391 There's no corresponding @value{GDBN} command.
36393 @subsubheading Example
36397 -data-write-memory-bytes &a "aabbccdd"
36404 -data-write-memory-bytes &a "aabbccdd" 16e
36409 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
36410 @node GDB/MI Tracepoint Commands
36411 @section @sc{gdb/mi} Tracepoint Commands
36413 The commands defined in this section implement MI support for
36414 tracepoints. For detailed introduction, see @ref{Tracepoints}.
36416 @findex -trace-find
36417 @subheading The @code{-trace-find} Command
36419 @subsubheading Synopsis
36422 -trace-find @var{mode} [@var{parameters}@dots{}]
36425 Find a trace frame using criteria defined by @var{mode} and
36426 @var{parameters}. The following table lists permissible
36427 modes and their parameters. For details of operation, see @ref{tfind}.
36432 No parameters are required. Stops examining trace frames.
36435 An integer is required as parameter. Selects tracepoint frame with
36438 @item tracepoint-number
36439 An integer is required as parameter. Finds next
36440 trace frame that corresponds to tracepoint with the specified number.
36443 An address is required as parameter. Finds
36444 next trace frame that corresponds to any tracepoint at the specified
36447 @item pc-inside-range
36448 Two addresses are required as parameters. Finds next trace
36449 frame that corresponds to a tracepoint at an address inside the
36450 specified range. Both bounds are considered to be inside the range.
36452 @item pc-outside-range
36453 Two addresses are required as parameters. Finds
36454 next trace frame that corresponds to a tracepoint at an address outside
36455 the specified range. Both bounds are considered to be inside the range.
36458 Location specification is required as parameter. @xref{Location Specifications}.
36459 Finds next trace frame that corresponds to a tracepoint at
36460 the specified location.
36464 If @samp{none} was passed as @var{mode}, the response does not
36465 have fields. Otherwise, the response may have the following fields:
36469 This field has either @samp{0} or @samp{1} as the value, depending
36470 on whether a matching tracepoint was found.
36473 The index of the found traceframe. This field is present iff
36474 the @samp{found} field has value of @samp{1}.
36477 The index of the found tracepoint. This field is present iff
36478 the @samp{found} field has value of @samp{1}.
36481 The information about the frame corresponding to the found trace
36482 frame. This field is present only if a trace frame was found.
36483 @xref{GDB/MI Frame Information}, for description of this field.
36487 @subsubheading @value{GDBN} Command
36489 The corresponding @value{GDBN} command is @samp{tfind}.
36491 @findex -trace-define-variable
36492 @subheading The @code{-trace-define-variable} Command
36494 @subsubheading Synopsis
36497 -trace-define-variable @var{name} [ @var{value} ]
36500 Create trace variable @var{name} if it does not exist. If
36501 @var{value} is specified, sets the initial value of the specified
36502 trace variable to that value. Note that the @var{name} should start
36503 with the @samp{$} character.
36505 @subsubheading @value{GDBN} Command
36507 The corresponding @value{GDBN} command is @samp{tvariable}.
36509 @findex -trace-frame-collected
36510 @subheading The @code{-trace-frame-collected} Command
36512 @subsubheading Synopsis
36515 -trace-frame-collected
36516 [--var-print-values @var{var_pval}]
36517 [--comp-print-values @var{comp_pval}]
36518 [--registers-format @var{regformat}]
36519 [--memory-contents]
36522 This command returns the set of collected objects, register names,
36523 trace state variable names, memory ranges and computed expressions
36524 that have been collected at a particular trace frame. The optional
36525 parameters to the command affect the output format in different ways.
36526 See the output description table below for more details.
36528 The reported names can be used in the normal manner to create
36529 varobjs and inspect the objects themselves. The items returned by
36530 this command are categorized so that it is clear which is a variable,
36531 which is a register, which is a trace state variable, which is a
36532 memory range and which is a computed expression.
36534 For instance, if the actions were
36536 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
36537 collect *(int*)0xaf02bef0@@40
36541 the object collected in its entirety would be @code{myVar}. The
36542 object @code{myArray} would be partially collected, because only the
36543 element at index @code{myIndex} would be collected. The remaining
36544 objects would be computed expressions.
36546 An example output would be:
36550 -trace-frame-collected
36552 explicit-variables=[@{name="myVar",value="1"@}],
36553 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
36554 @{name="myObj.field",value="0"@},
36555 @{name="myPtr->field",value="1"@},
36556 @{name="myCount + 2",value="3"@},
36557 @{name="$tvar1 + 1",value="43970027"@}],
36558 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
36559 @{number="1",value="0x0"@},
36560 @{number="2",value="0x4"@},
36562 @{number="125",value="0x0"@}],
36563 tvars=[@{name="$tvar1",current="43970026"@}],
36564 memory=[@{address="0x0000000000602264",length="4"@},
36565 @{address="0x0000000000615bc0",length="4"@}]
36572 @item explicit-variables
36573 The set of objects that have been collected in their entirety (as
36574 opposed to collecting just a few elements of an array or a few struct
36575 members). For each object, its name and value are printed.
36576 The @code{--var-print-values} option affects how or whether the value
36577 field is output. If @var{var_pval} is 0, then print only the names;
36578 if it is 1, print also their values; and if it is 2, print the name,
36579 type and value for simple data types, and the name and type for
36580 arrays, structures and unions.
36582 @item computed-expressions
36583 The set of computed expressions that have been collected at the
36584 current trace frame. The @code{--comp-print-values} option affects
36585 this set like the @code{--var-print-values} option affects the
36586 @code{explicit-variables} set. See above.
36589 The registers that have been collected at the current trace frame.
36590 For each register collected, the name and current value are returned.
36591 The value is formatted according to the @code{--registers-format}
36592 option. See the @command{-data-list-register-values} command for a
36593 list of the allowed formats. The default is @samp{x}.
36596 The trace state variables that have been collected at the current
36597 trace frame. For each trace state variable collected, the name and
36598 current value are returned.
36601 The set of memory ranges that have been collected at the current trace
36602 frame. Its content is a list of tuples. Each tuple represents a
36603 collected memory range and has the following fields:
36607 The start address of the memory range, as hexadecimal literal.
36610 The length of the memory range, as decimal literal.
36613 The contents of the memory block, in hex. This field is only present
36614 if the @code{--memory-contents} option is specified.
36620 @subsubheading @value{GDBN} Command
36622 There is no corresponding @value{GDBN} command.
36624 @subsubheading Example
36626 @findex -trace-list-variables
36627 @subheading The @code{-trace-list-variables} Command
36629 @subsubheading Synopsis
36632 -trace-list-variables
36635 Return a table of all defined trace variables. Each element of the
36636 table has the following fields:
36640 The name of the trace variable. This field is always present.
36643 The initial value. This is a 64-bit signed integer. This
36644 field is always present.
36647 The value the trace variable has at the moment. This is a 64-bit
36648 signed integer. This field is absent iff current value is
36649 not defined, for example if the trace was never run, or is
36654 @subsubheading @value{GDBN} Command
36656 The corresponding @value{GDBN} command is @samp{tvariables}.
36658 @subsubheading Example
36662 -trace-list-variables
36663 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
36664 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
36665 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
36666 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
36667 body=[variable=@{name="$trace_timestamp",initial="0"@}
36668 variable=@{name="$foo",initial="10",current="15"@}]@}
36672 @findex -trace-save
36673 @subheading The @code{-trace-save} Command
36675 @subsubheading Synopsis
36678 -trace-save [ -r ] [ -ctf ] @var{filename}
36681 Saves the collected trace data to @var{filename}. Without the
36682 @samp{-r} option, the data is downloaded from the target and saved
36683 in a local file. With the @samp{-r} option the target is asked
36684 to perform the save.
36686 By default, this command will save the trace in the tfile format. You can
36687 supply the optional @samp{-ctf} argument to save it the CTF format. See
36688 @ref{Trace Files} for more information about CTF.
36690 @subsubheading @value{GDBN} Command
36692 The corresponding @value{GDBN} command is @samp{tsave}.
36695 @findex -trace-start
36696 @subheading The @code{-trace-start} Command
36698 @subsubheading Synopsis
36704 Starts a tracing experiment. The result of this command does not
36707 @subsubheading @value{GDBN} Command
36709 The corresponding @value{GDBN} command is @samp{tstart}.
36711 @findex -trace-status
36712 @subheading The @code{-trace-status} Command
36714 @subsubheading Synopsis
36720 Obtains the status of a tracing experiment. The result may include
36721 the following fields:
36726 May have a value of either @samp{0}, when no tracing operations are
36727 supported, @samp{1}, when all tracing operations are supported, or
36728 @samp{file} when examining trace file. In the latter case, examining
36729 of trace frame is possible but new tracing experiement cannot be
36730 started. This field is always present.
36733 May have a value of either @samp{0} or @samp{1} depending on whether
36734 tracing experiement is in progress on target. This field is present
36735 if @samp{supported} field is not @samp{0}.
36738 Report the reason why the tracing was stopped last time. This field
36739 may be absent iff tracing was never stopped on target yet. The
36740 value of @samp{request} means the tracing was stopped as result of
36741 the @code{-trace-stop} command. The value of @samp{overflow} means
36742 the tracing buffer is full. The value of @samp{disconnection} means
36743 tracing was automatically stopped when @value{GDBN} has disconnected.
36744 The value of @samp{passcount} means tracing was stopped when a
36745 tracepoint was passed a maximal number of times for that tracepoint.
36746 This field is present if @samp{supported} field is not @samp{0}.
36748 @item stopping-tracepoint
36749 The number of tracepoint whose passcount as exceeded. This field is
36750 present iff the @samp{stop-reason} field has the value of
36754 @itemx frames-created
36755 The @samp{frames} field is a count of the total number of trace frames
36756 in the trace buffer, while @samp{frames-created} is the total created
36757 during the run, including ones that were discarded, such as when a
36758 circular trace buffer filled up. Both fields are optional.
36762 These fields tell the current size of the tracing buffer and the
36763 remaining space. These fields are optional.
36766 The value of the circular trace buffer flag. @code{1} means that the
36767 trace buffer is circular and old trace frames will be discarded if
36768 necessary to make room, @code{0} means that the trace buffer is linear
36772 The value of the disconnected tracing flag. @code{1} means that
36773 tracing will continue after @value{GDBN} disconnects, @code{0} means
36774 that the trace run will stop.
36777 The filename of the trace file being examined. This field is
36778 optional, and only present when examining a trace file.
36782 @subsubheading @value{GDBN} Command
36784 The corresponding @value{GDBN} command is @samp{tstatus}.
36786 @findex -trace-stop
36787 @subheading The @code{-trace-stop} Command
36789 @subsubheading Synopsis
36795 Stops a tracing experiment. The result of this command has the same
36796 fields as @code{-trace-status}, except that the @samp{supported} and
36797 @samp{running} fields are not output.
36799 @subsubheading @value{GDBN} Command
36801 The corresponding @value{GDBN} command is @samp{tstop}.
36804 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
36805 @node GDB/MI Symbol Query
36806 @section @sc{gdb/mi} Symbol Query Commands
36810 @findex -symbol-info-address
36811 @subheading The @code{-symbol-info-address} Command
36813 @subsubheading Synopsis
36816 -symbol-info-address @var{symbol}
36819 Describe where @var{symbol} is stored.
36821 @subsubheading @value{GDBN} Command
36823 The corresponding @value{GDBN} command is @samp{info address}.
36825 @subsubheading Example
36829 @findex -symbol-info-file
36830 @subheading The @code{-symbol-info-file} Command
36832 @subsubheading Synopsis
36838 Show the file for the symbol.
36840 @subsubheading @value{GDBN} Command
36842 There's no equivalent @value{GDBN} command. @code{gdbtk} has
36843 @samp{gdb_find_file}.
36845 @subsubheading Example
36849 @findex -symbol-info-functions
36850 @anchor{-symbol-info-functions}
36851 @subheading The @code{-symbol-info-functions} Command
36853 @subsubheading Synopsis
36856 -symbol-info-functions [--include-nondebug]
36857 [--type @var{type_regexp}]
36858 [--name @var{name_regexp}]
36859 [--max-results @var{limit}]
36863 Return a list containing the names and types for all global functions
36864 taken from the debug information. The functions are grouped by source
36865 file, and shown with the line number on which each function is
36868 The @code{--include-nondebug} option causes the output to include
36869 code symbols from the symbol table.
36871 The options @code{--type} and @code{--name} allow the symbols returned
36872 to be filtered based on either the name of the function, or the type
36873 signature of the function.
36875 The option @code{--max-results} restricts the command to return no
36876 more than @var{limit} results. If exactly @var{limit} results are
36877 returned then there might be additional results available if a higher
36880 @subsubheading @value{GDBN} Command
36882 The corresponding @value{GDBN} command is @samp{info functions}.
36884 @subsubheading Example
36888 -symbol-info-functions
36891 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36892 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36893 symbols=[@{line="36", name="f4", type="void (int *)",
36894 description="void f4(int *);"@},
36895 @{line="42", name="main", type="int ()",
36896 description="int main();"@},
36897 @{line="30", name="f1", type="my_int_t (int, int)",
36898 description="static my_int_t f1(int, int);"@}]@},
36899 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36900 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36901 symbols=[@{line="33", name="f2", type="float (another_float_t)",
36902 description="float f2(another_float_t);"@},
36903 @{line="39", name="f3", type="int (another_int_t)",
36904 description="int f3(another_int_t);"@},
36905 @{line="27", name="f1", type="another_float_t (int)",
36906 description="static another_float_t f1(int);"@}]@}]@}
36910 -symbol-info-functions --name f1
36913 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36914 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36915 symbols=[@{line="30", name="f1", type="my_int_t (int, int)",
36916 description="static my_int_t f1(int, int);"@}]@},
36917 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36918 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36919 symbols=[@{line="27", name="f1", type="another_float_t (int)",
36920 description="static another_float_t f1(int);"@}]@}]@}
36924 -symbol-info-functions --type void
36927 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36928 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36929 symbols=[@{line="36", name="f4", type="void (int *)",
36930 description="void f4(int *);"@}]@}]@}
36934 -symbol-info-functions --include-nondebug
36937 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36938 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36939 symbols=[@{line="36", name="f4", type="void (int *)",
36940 description="void f4(int *);"@},
36941 @{line="42", name="main", type="int ()",
36942 description="int main();"@},
36943 @{line="30", name="f1", type="my_int_t (int, int)",
36944 description="static my_int_t f1(int, int);"@}]@},
36945 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36946 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36947 symbols=[@{line="33", name="f2", type="float (another_float_t)",
36948 description="float f2(another_float_t);"@},
36949 @{line="39", name="f3", type="int (another_int_t)",
36950 description="int f3(another_int_t);"@},
36951 @{line="27", name="f1", type="another_float_t (int)",
36952 description="static another_float_t f1(int);"@}]@}],
36954 [@{address="0x0000000000400398",name="_init"@},
36955 @{address="0x00000000004003b0",name="_start"@},
36961 @findex -symbol-info-module-functions
36962 @anchor{-symbol-info-module-functions}
36963 @subheading The @code{-symbol-info-module-functions} Command
36965 @subsubheading Synopsis
36968 -symbol-info-module-functions [--module @var{module_regexp}]
36969 [--name @var{name_regexp}]
36970 [--type @var{type_regexp}]
36974 Return a list containing the names of all known functions within all
36975 know Fortran modules. The functions are grouped by source file and
36976 containing module, and shown with the line number on which each
36977 function is defined.
36979 The option @code{--module} only returns results for modules matching
36980 @var{module_regexp}. The option @code{--name} only returns functions
36981 whose name matches @var{name_regexp}, and @code{--type} only returns
36982 functions whose type matches @var{type_regexp}.
36984 @subsubheading @value{GDBN} Command
36986 The corresponding @value{GDBN} command is @samp{info module functions}.
36988 @subsubheading Example
36993 -symbol-info-module-functions
36996 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
36997 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
36998 symbols=[@{line="21",name="mod1::check_all",type="void (void)",
36999 description="void mod1::check_all(void);"@}]@}]@},
37001 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
37002 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
37003 symbols=[@{line="30",name="mod2::check_var_i",type="void (void)",
37004 description="void mod2::check_var_i(void);"@}]@}]@},
37006 files=[@{filename="/projec/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37007 fullname="/projec/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37008 symbols=[@{line="21",name="mod3::check_all",type="void (void)",
37009 description="void mod3::check_all(void);"@},
37010 @{line="27",name="mod3::check_mod2",type="void (void)",
37011 description="void mod3::check_mod2(void);"@}]@}]@},
37012 @{module="modmany",
37013 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37014 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37015 symbols=[@{line="35",name="modmany::check_some",type="void (void)",
37016 description="void modmany::check_some(void);"@}]@}]@},
37018 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37019 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37020 symbols=[@{line="44",name="moduse::check_all",type="void (void)",
37021 description="void moduse::check_all(void);"@},
37022 @{line="49",name="moduse::check_var_x",type="void (void)",
37023 description="void moduse::check_var_x(void);"@}]@}]@}]
37027 @findex -symbol-info-module-variables
37028 @anchor{-symbol-info-module-variables}
37029 @subheading The @code{-symbol-info-module-variables} Command
37031 @subsubheading Synopsis
37034 -symbol-info-module-variables [--module @var{module_regexp}]
37035 [--name @var{name_regexp}]
37036 [--type @var{type_regexp}]
37040 Return a list containing the names of all known variables within all
37041 know Fortran modules. The variables are grouped by source file and
37042 containing module, and shown with the line number on which each
37043 variable is defined.
37045 The option @code{--module} only returns results for modules matching
37046 @var{module_regexp}. The option @code{--name} only returns variables
37047 whose name matches @var{name_regexp}, and @code{--type} only returns
37048 variables whose type matches @var{type_regexp}.
37050 @subsubheading @value{GDBN} Command
37052 The corresponding @value{GDBN} command is @samp{info module variables}.
37054 @subsubheading Example
37059 -symbol-info-module-variables
37062 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
37063 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
37064 symbols=[@{line="18",name="mod1::var_const",type="integer(kind=4)",
37065 description="integer(kind=4) mod1::var_const;"@},
37066 @{line="17",name="mod1::var_i",type="integer(kind=4)",
37067 description="integer(kind=4) mod1::var_i;"@}]@}]@},
37069 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
37070 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
37071 symbols=[@{line="28",name="mod2::var_i",type="integer(kind=4)",
37072 description="integer(kind=4) mod2::var_i;"@}]@}]@},
37074 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37075 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37076 symbols=[@{line="18",name="mod3::mod1",type="integer(kind=4)",
37077 description="integer(kind=4) mod3::mod1;"@},
37078 @{line="17",name="mod3::mod2",type="integer(kind=4)",
37079 description="integer(kind=4) mod3::mod2;"@},
37080 @{line="19",name="mod3::var_i",type="integer(kind=4)",
37081 description="integer(kind=4) mod3::var_i;"@}]@}]@},
37082 @{module="modmany",
37083 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37084 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37085 symbols=[@{line="33",name="modmany::var_a",type="integer(kind=4)",
37086 description="integer(kind=4) modmany::var_a;"@},
37087 @{line="33",name="modmany::var_b",type="integer(kind=4)",
37088 description="integer(kind=4) modmany::var_b;"@},
37089 @{line="33",name="modmany::var_c",type="integer(kind=4)",
37090 description="integer(kind=4) modmany::var_c;"@},
37091 @{line="33",name="modmany::var_i",type="integer(kind=4)",
37092 description="integer(kind=4) modmany::var_i;"@}]@}]@},
37094 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37095 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37096 symbols=[@{line="42",name="moduse::var_x",type="integer(kind=4)",
37097 description="integer(kind=4) moduse::var_x;"@},
37098 @{line="42",name="moduse::var_y",type="integer(kind=4)",
37099 description="integer(kind=4) moduse::var_y;"@}]@}]@}]
37103 @findex -symbol-info-modules
37104 @anchor{-symbol-info-modules}
37105 @subheading The @code{-symbol-info-modules} Command
37107 @subsubheading Synopsis
37110 -symbol-info-modules [--name @var{name_regexp}]
37111 [--max-results @var{limit}]
37116 Return a list containing the names of all known Fortran modules. The
37117 modules are grouped by source file, and shown with the line number on
37118 which each modules is defined.
37120 The option @code{--name} allows the modules returned to be filtered
37121 based the name of the module.
37123 The option @code{--max-results} restricts the command to return no
37124 more than @var{limit} results. If exactly @var{limit} results are
37125 returned then there might be additional results available if a higher
37128 @subsubheading @value{GDBN} Command
37130 The corresponding @value{GDBN} command is @samp{info modules}.
37132 @subsubheading Example
37136 -symbol-info-modules
37139 [@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
37140 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
37141 symbols=[@{line="16",name="mod1"@},
37142 @{line="22",name="mod2"@}]@},
37143 @{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37144 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37145 symbols=[@{line="16",name="mod3"@},
37146 @{line="22",name="modmany"@},
37147 @{line="26",name="moduse"@}]@}]@}
37151 -symbol-info-modules --name mod[123]
37154 [@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
37155 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
37156 symbols=[@{line="16",name="mod1"@},
37157 @{line="22",name="mod2"@}]@},
37158 @{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37159 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
37160 symbols=[@{line="16",name="mod3"@}]@}]@}
37164 @findex -symbol-info-types
37165 @anchor{-symbol-info-types}
37166 @subheading The @code{-symbol-info-types} Command
37168 @subsubheading Synopsis
37171 -symbol-info-types [--name @var{name_regexp}]
37172 [--max-results @var{limit}]
37177 Return a list of all defined types. The types are grouped by source
37178 file, and shown with the line number on which each user defined type
37179 is defined. Some base types are not defined in the source code but
37180 are added to the debug information by the compiler, for example
37181 @code{int}, @code{float}, etc.; these types do not have an associated
37184 The option @code{--name} allows the list of types returned to be
37187 The option @code{--max-results} restricts the command to return no
37188 more than @var{limit} results. If exactly @var{limit} results are
37189 returned then there might be additional results available if a higher
37192 @subsubheading @value{GDBN} Command
37194 The corresponding @value{GDBN} command is @samp{info types}.
37196 @subsubheading Example
37203 [@{filename="gdb.mi/mi-sym-info-1.c",
37204 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
37205 symbols=[@{name="float"@},
37207 @{line="27",name="typedef int my_int_t;"@}]@},
37208 @{filename="gdb.mi/mi-sym-info-2.c",
37209 fullname="/project/gdb.mi/mi-sym-info-2.c",
37210 symbols=[@{line="24",name="typedef float another_float_t;"@},
37211 @{line="23",name="typedef int another_int_t;"@},
37213 @{name="int"@}]@}]@}
37217 -symbol-info-types --name _int_
37220 [@{filename="gdb.mi/mi-sym-info-1.c",
37221 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
37222 symbols=[@{line="27",name="typedef int my_int_t;"@}]@},
37223 @{filename="gdb.mi/mi-sym-info-2.c",
37224 fullname="/project/gdb.mi/mi-sym-info-2.c",
37225 symbols=[@{line="23",name="typedef int another_int_t;"@}]@}]@}
37229 @findex -symbol-info-variables
37230 @anchor{-symbol-info-variables}
37231 @subheading The @code{-symbol-info-variables} Command
37233 @subsubheading Synopsis
37236 -symbol-info-variables [--include-nondebug]
37237 [--type @var{type_regexp}]
37238 [--name @var{name_regexp}]
37239 [--max-results @var{limit}]
37244 Return a list containing the names and types for all global variables
37245 taken from the debug information. The variables are grouped by source
37246 file, and shown with the line number on which each variable is
37249 The @code{--include-nondebug} option causes the output to include
37250 data symbols from the symbol table.
37252 The options @code{--type} and @code{--name} allow the symbols returned
37253 to be filtered based on either the name of the variable, or the type
37256 The option @code{--max-results} restricts the command to return no
37257 more than @var{limit} results. If exactly @var{limit} results are
37258 returned then there might be additional results available if a higher
37261 @subsubheading @value{GDBN} Command
37263 The corresponding @value{GDBN} command is @samp{info variables}.
37265 @subsubheading Example
37269 -symbol-info-variables
37272 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
37273 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
37274 symbols=[@{line="25",name="global_f1",type="float",
37275 description="static float global_f1;"@},
37276 @{line="24",name="global_i1",type="int",
37277 description="static int global_i1;"@}]@},
37278 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
37279 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
37280 symbols=[@{line="21",name="global_f2",type="int",
37281 description="int global_f2;"@},
37282 @{line="20",name="global_i2",type="int",
37283 description="int global_i2;"@},
37284 @{line="19",name="global_f1",type="float",
37285 description="static float global_f1;"@},
37286 @{line="18",name="global_i1",type="int",
37287 description="static int global_i1;"@}]@}]@}
37291 -symbol-info-variables --name f1
37294 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
37295 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
37296 symbols=[@{line="25",name="global_f1",type="float",
37297 description="static float global_f1;"@}]@},
37298 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
37299 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
37300 symbols=[@{line="19",name="global_f1",type="float",
37301 description="static float global_f1;"@}]@}]@}
37305 -symbol-info-variables --type float
37308 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
37309 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
37310 symbols=[@{line="25",name="global_f1",type="float",
37311 description="static float global_f1;"@}]@},
37312 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
37313 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
37314 symbols=[@{line="19",name="global_f1",type="float",
37315 description="static float global_f1;"@}]@}]@}
37319 -symbol-info-variables --include-nondebug
37322 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
37323 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
37324 symbols=[@{line="25",name="global_f1",type="float",
37325 description="static float global_f1;"@},
37326 @{line="24",name="global_i1",type="int",
37327 description="static int global_i1;"@}]@},
37328 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
37329 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
37330 symbols=[@{line="21",name="global_f2",type="int",
37331 description="int global_f2;"@},
37332 @{line="20",name="global_i2",type="int",
37333 description="int global_i2;"@},
37334 @{line="19",name="global_f1",type="float",
37335 description="static float global_f1;"@},
37336 @{line="18",name="global_i1",type="int",
37337 description="static int global_i1;"@}]@}],
37339 [@{address="0x00000000004005d0",name="_IO_stdin_used"@},
37340 @{address="0x00000000004005d8",name="__dso_handle"@}
37347 @findex -symbol-info-line
37348 @subheading The @code{-symbol-info-line} Command
37350 @subsubheading Synopsis
37356 Show the core addresses of the code for a source line.
37358 @subsubheading @value{GDBN} Command
37360 The corresponding @value{GDBN} command is @samp{info line}.
37361 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
37363 @subsubheading Example
37367 @findex -symbol-info-symbol
37368 @subheading The @code{-symbol-info-symbol} Command
37370 @subsubheading Synopsis
37373 -symbol-info-symbol @var{addr}
37376 Describe what symbol is at location @var{addr}.
37378 @subsubheading @value{GDBN} Command
37380 The corresponding @value{GDBN} command is @samp{info symbol}.
37382 @subsubheading Example
37386 @findex -symbol-list-functions
37387 @subheading The @code{-symbol-list-functions} Command
37389 @subsubheading Synopsis
37392 -symbol-list-functions
37395 List the functions in the executable.
37397 @subsubheading @value{GDBN} Command
37399 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
37400 @samp{gdb_search} in @code{gdbtk}.
37402 @subsubheading Example
37407 @findex -symbol-list-lines
37408 @subheading The @code{-symbol-list-lines} Command
37410 @subsubheading Synopsis
37413 -symbol-list-lines @var{filename}
37416 Print the list of lines that contain code and their associated program
37417 addresses for the given source filename. The entries are sorted in
37418 ascending PC order.
37420 @subsubheading @value{GDBN} Command
37422 There is no corresponding @value{GDBN} command.
37424 @subsubheading Example
37427 -symbol-list-lines basics.c
37428 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
37434 @findex -symbol-list-types
37435 @subheading The @code{-symbol-list-types} Command
37437 @subsubheading Synopsis
37443 List all the type names.
37445 @subsubheading @value{GDBN} Command
37447 The corresponding commands are @samp{info types} in @value{GDBN},
37448 @samp{gdb_search} in @code{gdbtk}.
37450 @subsubheading Example
37454 @findex -symbol-list-variables
37455 @subheading The @code{-symbol-list-variables} Command
37457 @subsubheading Synopsis
37460 -symbol-list-variables
37463 List all the global and static variable names.
37465 @subsubheading @value{GDBN} Command
37467 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
37469 @subsubheading Example
37473 @findex -symbol-locate
37474 @subheading The @code{-symbol-locate} Command
37476 @subsubheading Synopsis
37482 @subsubheading @value{GDBN} Command
37484 @samp{gdb_loc} in @code{gdbtk}.
37486 @subsubheading Example
37490 @findex -symbol-type
37491 @subheading The @code{-symbol-type} Command
37493 @subsubheading Synopsis
37496 -symbol-type @var{variable}
37499 Show type of @var{variable}.
37501 @subsubheading @value{GDBN} Command
37503 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
37504 @samp{gdb_obj_variable}.
37506 @subsubheading Example
37511 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
37512 @node GDB/MI File Commands
37513 @section @sc{gdb/mi} File Commands
37515 This section describes the GDB/MI commands to specify executable file names
37516 and to read in and obtain symbol table information.
37518 @findex -file-exec-and-symbols
37519 @subheading The @code{-file-exec-and-symbols} Command
37521 @subsubheading Synopsis
37524 -file-exec-and-symbols @var{file}
37527 Specify the executable file to be debugged. This file is the one from
37528 which the symbol table is also read. If no file is specified, the
37529 command clears the executable and symbol information. If breakpoints
37530 are set when using this command with no arguments, @value{GDBN} will produce
37531 error messages. Otherwise, no output is produced, except a completion
37534 @subsubheading @value{GDBN} Command
37536 The corresponding @value{GDBN} command is @samp{file}.
37538 @subsubheading Example
37542 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
37548 @findex -file-exec-file
37549 @subheading The @code{-file-exec-file} Command
37551 @subsubheading Synopsis
37554 -file-exec-file @var{file}
37557 Specify the executable file to be debugged. Unlike
37558 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
37559 from this file. If used without argument, @value{GDBN} clears the information
37560 about the executable file. No output is produced, except a completion
37563 @subsubheading @value{GDBN} Command
37565 The corresponding @value{GDBN} command is @samp{exec-file}.
37567 @subsubheading Example
37571 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
37578 @findex -file-list-exec-sections
37579 @subheading The @code{-file-list-exec-sections} Command
37581 @subsubheading Synopsis
37584 -file-list-exec-sections
37587 List the sections of the current executable file.
37589 @subsubheading @value{GDBN} Command
37591 The @value{GDBN} command @samp{info file} shows, among the rest, the same
37592 information as this command. @code{gdbtk} has a corresponding command
37593 @samp{gdb_load_info}.
37595 @subsubheading Example
37600 @findex -file-list-exec-source-file
37601 @subheading The @code{-file-list-exec-source-file} Command
37603 @subsubheading Synopsis
37606 -file-list-exec-source-file
37609 List the line number, the current source file, and the absolute path
37610 to the current source file for the current executable. The macro
37611 information field has a value of @samp{1} or @samp{0} depending on
37612 whether or not the file includes preprocessor macro information.
37614 @subsubheading @value{GDBN} Command
37616 The @value{GDBN} equivalent is @samp{info source}
37618 @subsubheading Example
37622 123-file-list-exec-source-file
37623 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
37628 @findex -file-list-exec-source-files
37629 @subheading The @code{-file-list-exec-source-files} Command
37630 @kindex info sources
37632 @subsubheading Synopsis
37635 -file-list-exec-source-files @r{[} @var{--group-by-objfile} @r{]}
37636 @r{[} @var{--dirname} @r{|} @var{--basename} @r{]}
37638 @r{[} @var{regexp} @r{]}
37641 This command returns information about the source files @value{GDBN}
37642 knows about, it will output both the filename and fullname (absolute
37643 file name) of a source file, though the fullname can be elided if this
37644 information is not known to @value{GDBN}.
37646 With no arguments this command returns a list of source files. Each
37647 source file is represented by a tuple with the fields; @var{file},
37648 @var{fullname}, and @var{debug-fully-read}. The @var{file} is the
37649 display name for the file, while @var{fullname} is the absolute name
37650 of the file. The @var{fullname} field can be elided if the absolute
37651 name of the source file can't be computed. The field
37652 @var{debug-fully-read} will be a string, either @code{true} or
37653 @code{false}. When @code{true}, this indicates the full debug
37654 information for the compilation unit describing this file has been
37655 read in. When @code{false}, the full debug information has not yet
37656 been read in. While reading in the full debug information it is
37657 possible that @value{GDBN} could become aware of additional source
37660 The optional @var{regexp} can be used to filter the list of source
37661 files returned. The @var{regexp} will be matched against the full
37662 source file name. The matching is case-sensitive, except on operating
37663 systems that have case-insensitive filesystem (e.g.,
37664 MS-Windows). @samp{--} can be used before @var{regexp} to prevent
37665 @value{GDBN} interpreting @var{regexp} as a command option (e.g.@: if
37666 @var{regexp} starts with @samp{-}).
37668 If @code{--dirname} is provided, then @var{regexp} is matched only
37669 against the directory name of each source file. If @code{--basename}
37670 is provided, then @var{regexp} is matched against the basename of each
37671 source file. Only one of @code{--dirname} or @code{--basename} may be
37672 given, and if either is given then @var{regexp} is required.
37674 If @code{--group-by-objfile} is used then the format of the results is
37675 changed. The results will now be a list of tuples, with each tuple
37676 representing an object file (executable or shared library) loaded into
37677 @value{GDBN}. The fields of these tuples are; @var{filename},
37678 @var{debug-info}, and @var{sources}. The @var{filename} is the
37679 absolute name of the object file, @var{debug-info} is a string with
37680 one of the following values:
37684 This object file has no debug information.
37685 @item partially-read
37686 This object file has debug information, but it is not fully read in
37687 yet. When it is read in later, GDB might become aware of additional
37690 This object file has debug information, and this information is fully
37691 read into GDB. The list of source files is complete.
37694 The @var{sources} is a list or tuples, with each tuple describing a
37695 single source file with the same fields as described previously. The
37696 @var{sources} list can be empty for object files that have no debug
37699 @subsubheading @value{GDBN} Command
37701 The @value{GDBN} equivalent is @samp{info sources}.
37702 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
37704 @subsubheading Example
37707 -file-list-exec-source-files
37708 ^done,files=[@{file="foo.c",fullname="/home/foo.c",debug-fully-read="true"@},
37709 @{file="/home/bar.c",fullname="/home/bar.c",debug-fully-read="true"@},
37710 @{file="gdb_could_not_find_fullpath.c",debug-fully-read="true"@}]
37712 -file-list-exec-source-files
37713 ^done,files=[@{file="test.c",
37714 fullname="/tmp/info-sources/test.c",
37715 debug-fully-read="true"@},
37716 @{file="/usr/include/stdc-predef.h",
37717 fullname="/usr/include/stdc-predef.h",
37718 debug-fully-read="true"@},
37720 fullname="/tmp/info-sources/header.h",
37721 debug-fully-read="true"@},
37723 fullname="/tmp/info-sources/helper.c",
37724 debug-fully-read="true"@}]
37726 -file-list-exec-source-files -- \\.c
37727 ^done,files=[@{file="test.c",
37728 fullname="/tmp/info-sources/test.c",
37729 debug-fully-read="true"@},
37731 fullname="/tmp/info-sources/helper.c",
37732 debug-fully-read="true"@}]
37734 -file-list-exec-source-files --group-by-objfile
37735 ^done,files=[@{filename="/tmp/info-sources/test.x",
37736 debug-info="fully-read",
37737 sources=[@{file="test.c",
37738 fullname="/tmp/info-sources/test.c",
37739 debug-fully-read="true"@},
37740 @{file="/usr/include/stdc-predef.h",
37741 fullname="/usr/include/stdc-predef.h",
37742 debug-fully-read="true"@},
37744 fullname="/tmp/info-sources/header.h",
37745 debug-fully-read="true"@}]@},
37746 @{filename="/lib64/ld-linux-x86-64.so.2",
37749 @{filename="system-supplied DSO at 0x7ffff7fcf000",
37752 @{filename="/tmp/info-sources/libhelper.so",
37753 debug-info="fully-read",
37754 sources=[@{file="helper.c",
37755 fullname="/tmp/info-sources/helper.c",
37756 debug-fully-read="true"@},
37757 @{file="/usr/include/stdc-predef.h",
37758 fullname="/usr/include/stdc-predef.h",
37759 debug-fully-read="true"@},
37761 fullname="/tmp/info-sources/header.h",
37762 debug-fully-read="true"@}]@},
37763 @{filename="/lib64/libc.so.6",
37768 @findex -file-list-shared-libraries
37769 @subheading The @code{-file-list-shared-libraries} Command
37771 @subsubheading Synopsis
37774 -file-list-shared-libraries [ @var{regexp} ]
37777 List the shared libraries in the program.
37778 With a regular expression @var{regexp}, only those libraries whose
37779 names match @var{regexp} are listed.
37781 @subsubheading @value{GDBN} Command
37783 The corresponding @value{GDBN} command is @samp{info shared}. The fields
37784 have a similar meaning to the @code{=library-loaded} notification.
37785 The @code{ranges} field specifies the multiple segments belonging to this
37786 library. Each range has the following fields:
37790 The address defining the inclusive lower bound of the segment.
37792 The address defining the exclusive upper bound of the segment.
37795 @subsubheading Example
37798 -file-list-exec-source-files
37799 ^done,shared-libraries=[
37800 @{id="/lib/libfoo.so",target-name="/lib/libfoo.so",host-name="/lib/libfoo.so",symbols-loaded="1",thread-group="i1",ranges=[@{from="0x72815989",to="0x728162c0"@}]@},
37801 @{id="/lib/libbar.so",target-name="/lib/libbar.so",host-name="/lib/libbar.so",symbols-loaded="1",thread-group="i1",ranges=[@{from="0x76ee48c0",to="0x76ee9160"@}]@}]
37807 @findex -file-list-symbol-files
37808 @subheading The @code{-file-list-symbol-files} Command
37810 @subsubheading Synopsis
37813 -file-list-symbol-files
37818 @subsubheading @value{GDBN} Command
37820 The corresponding @value{GDBN} command is @samp{info file} (part of it).
37822 @subsubheading Example
37827 @findex -file-symbol-file
37828 @subheading The @code{-file-symbol-file} Command
37830 @subsubheading Synopsis
37833 -file-symbol-file @var{file}
37836 Read symbol table info from the specified @var{file} argument. When
37837 used without arguments, clears @value{GDBN}'s symbol table info. No output is
37838 produced, except for a completion notification.
37840 @subsubheading @value{GDBN} Command
37842 The corresponding @value{GDBN} command is @samp{symbol-file}.
37844 @subsubheading Example
37848 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
37854 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
37855 @node GDB/MI Memory Overlay Commands
37856 @section @sc{gdb/mi} Memory Overlay Commands
37858 The memory overlay commands are not implemented.
37860 @c @subheading -overlay-auto
37862 @c @subheading -overlay-list-mapping-state
37864 @c @subheading -overlay-list-overlays
37866 @c @subheading -overlay-map
37868 @c @subheading -overlay-off
37870 @c @subheading -overlay-on
37872 @c @subheading -overlay-unmap
37874 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
37875 @node GDB/MI Signal Handling Commands
37876 @section @sc{gdb/mi} Signal Handling Commands
37878 Signal handling commands are not implemented.
37880 @c @subheading -signal-handle
37882 @c @subheading -signal-list-handle-actions
37884 @c @subheading -signal-list-signal-types
37888 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
37889 @node GDB/MI Target Manipulation
37890 @section @sc{gdb/mi} Target Manipulation Commands
37893 @findex -target-attach
37894 @subheading The @code{-target-attach} Command
37896 @subsubheading Synopsis
37899 -target-attach @var{pid} | @var{gid} | @var{file}
37902 Attach to a process @var{pid} or a file @var{file} outside of
37903 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
37904 group, the id previously returned by
37905 @samp{-list-thread-groups --available} must be used.
37907 @subsubheading @value{GDBN} Command
37909 The corresponding @value{GDBN} command is @samp{attach}.
37911 @subsubheading Example
37915 =thread-created,id="1"
37916 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
37922 @findex -target-compare-sections
37923 @subheading The @code{-target-compare-sections} Command
37925 @subsubheading Synopsis
37928 -target-compare-sections [ @var{section} ]
37931 Compare data of section @var{section} on target to the exec file.
37932 Without the argument, all sections are compared.
37934 @subsubheading @value{GDBN} Command
37936 The @value{GDBN} equivalent is @samp{compare-sections}.
37938 @subsubheading Example
37943 @findex -target-detach
37944 @subheading The @code{-target-detach} Command
37946 @subsubheading Synopsis
37949 -target-detach [ @var{pid} | @var{gid} ]
37952 Detach from the remote target which normally resumes its execution.
37953 If either @var{pid} or @var{gid} is specified, detaches from either
37954 the specified process, or specified thread group. There's no output.
37956 @subsubheading @value{GDBN} Command
37958 The corresponding @value{GDBN} command is @samp{detach}.
37960 @subsubheading Example
37970 @findex -target-disconnect
37971 @subheading The @code{-target-disconnect} Command
37973 @subsubheading Synopsis
37979 Disconnect from the remote target. There's no output and the target is
37980 generally not resumed.
37982 @subsubheading @value{GDBN} Command
37984 The corresponding @value{GDBN} command is @samp{disconnect}.
37986 @subsubheading Example
37996 @findex -target-download
37997 @subheading The @code{-target-download} Command
37999 @subsubheading Synopsis
38005 Loads the executable onto the remote target.
38006 It prints out an update message every half second, which includes the fields:
38010 The name of the section.
38012 The size of what has been sent so far for that section.
38014 The size of the section.
38016 The total size of what was sent so far (the current and the previous sections).
38018 The size of the overall executable to download.
38022 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
38023 @sc{gdb/mi} Output Syntax}).
38025 In addition, it prints the name and size of the sections, as they are
38026 downloaded. These messages include the following fields:
38030 The name of the section.
38032 The size of the section.
38034 The size of the overall executable to download.
38038 At the end, a summary is printed.
38040 @subsubheading @value{GDBN} Command
38042 The corresponding @value{GDBN} command is @samp{load}.
38044 @subsubheading Example
38046 Note: each status message appears on a single line. Here the messages
38047 have been broken down so that they can fit onto a page.
38052 +download,@{section=".text",section-size="6668",total-size="9880"@}
38053 +download,@{section=".text",section-sent="512",section-size="6668",
38054 total-sent="512",total-size="9880"@}
38055 +download,@{section=".text",section-sent="1024",section-size="6668",
38056 total-sent="1024",total-size="9880"@}
38057 +download,@{section=".text",section-sent="1536",section-size="6668",
38058 total-sent="1536",total-size="9880"@}
38059 +download,@{section=".text",section-sent="2048",section-size="6668",
38060 total-sent="2048",total-size="9880"@}
38061 +download,@{section=".text",section-sent="2560",section-size="6668",
38062 total-sent="2560",total-size="9880"@}
38063 +download,@{section=".text",section-sent="3072",section-size="6668",
38064 total-sent="3072",total-size="9880"@}
38065 +download,@{section=".text",section-sent="3584",section-size="6668",
38066 total-sent="3584",total-size="9880"@}
38067 +download,@{section=".text",section-sent="4096",section-size="6668",
38068 total-sent="4096",total-size="9880"@}
38069 +download,@{section=".text",section-sent="4608",section-size="6668",
38070 total-sent="4608",total-size="9880"@}
38071 +download,@{section=".text",section-sent="5120",section-size="6668",
38072 total-sent="5120",total-size="9880"@}
38073 +download,@{section=".text",section-sent="5632",section-size="6668",
38074 total-sent="5632",total-size="9880"@}
38075 +download,@{section=".text",section-sent="6144",section-size="6668",
38076 total-sent="6144",total-size="9880"@}
38077 +download,@{section=".text",section-sent="6656",section-size="6668",
38078 total-sent="6656",total-size="9880"@}
38079 +download,@{section=".init",section-size="28",total-size="9880"@}
38080 +download,@{section=".fini",section-size="28",total-size="9880"@}
38081 +download,@{section=".data",section-size="3156",total-size="9880"@}
38082 +download,@{section=".data",section-sent="512",section-size="3156",
38083 total-sent="7236",total-size="9880"@}
38084 +download,@{section=".data",section-sent="1024",section-size="3156",
38085 total-sent="7748",total-size="9880"@}
38086 +download,@{section=".data",section-sent="1536",section-size="3156",
38087 total-sent="8260",total-size="9880"@}
38088 +download,@{section=".data",section-sent="2048",section-size="3156",
38089 total-sent="8772",total-size="9880"@}
38090 +download,@{section=".data",section-sent="2560",section-size="3156",
38091 total-sent="9284",total-size="9880"@}
38092 +download,@{section=".data",section-sent="3072",section-size="3156",
38093 total-sent="9796",total-size="9880"@}
38094 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
38101 @findex -target-exec-status
38102 @subheading The @code{-target-exec-status} Command
38104 @subsubheading Synopsis
38107 -target-exec-status
38110 Provide information on the state of the target (whether it is running or
38111 not, for instance).
38113 @subsubheading @value{GDBN} Command
38115 There's no equivalent @value{GDBN} command.
38117 @subsubheading Example
38121 @findex -target-list-available-targets
38122 @subheading The @code{-target-list-available-targets} Command
38124 @subsubheading Synopsis
38127 -target-list-available-targets
38130 List the possible targets to connect to.
38132 @subsubheading @value{GDBN} Command
38134 The corresponding @value{GDBN} command is @samp{help target}.
38136 @subsubheading Example
38140 @findex -target-list-current-targets
38141 @subheading The @code{-target-list-current-targets} Command
38143 @subsubheading Synopsis
38146 -target-list-current-targets
38149 Describe the current target.
38151 @subsubheading @value{GDBN} Command
38153 The corresponding information is printed by @samp{info file} (among
38156 @subsubheading Example
38160 @findex -target-list-parameters
38161 @subheading The @code{-target-list-parameters} Command
38163 @subsubheading Synopsis
38166 -target-list-parameters
38172 @subsubheading @value{GDBN} Command
38176 @subsubheading Example
38179 @findex -target-flash-erase
38180 @subheading The @code{-target-flash-erase} Command
38182 @subsubheading Synopsis
38185 -target-flash-erase
38188 Erases all known flash memory regions on the target.
38190 The corresponding @value{GDBN} command is @samp{flash-erase}.
38192 The output is a list of flash regions that have been erased, with starting
38193 addresses and memory region sizes.
38197 -target-flash-erase
38198 ^done,erased-regions=@{address="0x0",size="0x40000"@}
38202 @findex -target-select
38203 @subheading The @code{-target-select} Command
38205 @subsubheading Synopsis
38208 -target-select @var{type} @var{parameters @dots{}}
38211 Connect @value{GDBN} to the remote target. This command takes two args:
38215 The type of target, for instance @samp{remote}, etc.
38216 @item @var{parameters}
38217 Device names, host names and the like. @xref{Target Commands, ,
38218 Commands for Managing Targets}, for more details.
38221 The output is a connection notification, followed by the address at
38222 which the target program is, in the following form:
38225 ^connected,addr="@var{address}",func="@var{function name}",
38226 args=[@var{arg list}]
38229 @subsubheading @value{GDBN} Command
38231 The corresponding @value{GDBN} command is @samp{target}.
38233 @subsubheading Example
38237 -target-select remote /dev/ttya
38238 ^connected,addr="0xfe00a300",func="??",args=[]
38242 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
38243 @node GDB/MI File Transfer Commands
38244 @section @sc{gdb/mi} File Transfer Commands
38247 @findex -target-file-put
38248 @subheading The @code{-target-file-put} Command
38250 @subsubheading Synopsis
38253 -target-file-put @var{hostfile} @var{targetfile}
38256 Copy file @var{hostfile} from the host system (the machine running
38257 @value{GDBN}) to @var{targetfile} on the target system.
38259 @subsubheading @value{GDBN} Command
38261 The corresponding @value{GDBN} command is @samp{remote put}.
38263 @subsubheading Example
38267 -target-file-put localfile remotefile
38273 @findex -target-file-get
38274 @subheading The @code{-target-file-get} Command
38276 @subsubheading Synopsis
38279 -target-file-get @var{targetfile} @var{hostfile}
38282 Copy file @var{targetfile} from the target system to @var{hostfile}
38283 on the host system.
38285 @subsubheading @value{GDBN} Command
38287 The corresponding @value{GDBN} command is @samp{remote get}.
38289 @subsubheading Example
38293 -target-file-get remotefile localfile
38299 @findex -target-file-delete
38300 @subheading The @code{-target-file-delete} Command
38302 @subsubheading Synopsis
38305 -target-file-delete @var{targetfile}
38308 Delete @var{targetfile} from the target system.
38310 @subsubheading @value{GDBN} Command
38312 The corresponding @value{GDBN} command is @samp{remote delete}.
38314 @subsubheading Example
38318 -target-file-delete remotefile
38324 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
38325 @node GDB/MI Ada Exceptions Commands
38326 @section Ada Exceptions @sc{gdb/mi} Commands
38328 @findex -info-ada-exceptions
38329 @subheading The @code{-info-ada-exceptions} Command
38331 @subsubheading Synopsis
38334 -info-ada-exceptions [ @var{regexp}]
38337 List all Ada exceptions defined within the program being debugged.
38338 With a regular expression @var{regexp}, only those exceptions whose
38339 names match @var{regexp} are listed.
38341 @subsubheading @value{GDBN} Command
38343 The corresponding @value{GDBN} command is @samp{info exceptions}.
38345 @subsubheading Result
38347 The result is a table of Ada exceptions. The following columns are
38348 defined for each exception:
38352 The name of the exception.
38355 The address of the exception.
38359 @subsubheading Example
38362 -info-ada-exceptions aint
38363 ^done,ada-exceptions=@{nr_rows="2",nr_cols="2",
38364 hdr=[@{width="1",alignment="-1",col_name="name",colhdr="Name"@},
38365 @{width="1",alignment="-1",col_name="address",colhdr="Address"@}],
38366 body=[@{name="constraint_error",address="0x0000000000613da0"@},
38367 @{name="const.aint_global_e",address="0x0000000000613b00"@}]@}
38370 @subheading Catching Ada Exceptions
38372 The commands describing how to ask @value{GDBN} to stop when a program
38373 raises an exception are described at @ref{Ada Exception GDB/MI
38374 Catchpoint Commands}.
38377 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
38378 @node GDB/MI Support Commands
38379 @section @sc{gdb/mi} Support Commands
38381 Since new commands and features get regularly added to @sc{gdb/mi},
38382 some commands are available to help front-ends query the debugger
38383 about support for these capabilities. Similarly, it is also possible
38384 to query @value{GDBN} about target support of certain features.
38386 @cindex @code{-info-gdb-mi-command}
38387 @findex -info-gdb-mi-command
38388 @subheading The @code{-info-gdb-mi-command} Command
38390 @subsubheading Synopsis
38393 -info-gdb-mi-command @var{cmd_name}
38396 Query support for the @sc{gdb/mi} command named @var{cmd_name}.
38398 Note that the dash (@code{-}) starting all @sc{gdb/mi} commands
38399 is technically not part of the command name (@pxref{GDB/MI Input
38400 Syntax}), and thus should be omitted in @var{cmd_name}. However,
38401 for ease of use, this command also accepts the form with the leading
38404 @subsubheading @value{GDBN} Command
38406 There is no corresponding @value{GDBN} command.
38408 @subsubheading Result
38410 The result is a tuple. There is currently only one field:
38414 This field is equal to @code{"true"} if the @sc{gdb/mi} command exists,
38415 @code{"false"} otherwise.
38419 @subsubheading Example
38421 Here is an example where the @sc{gdb/mi} command does not exist:
38424 -info-gdb-mi-command unsupported-command
38425 ^done,command=@{exists="false"@}
38429 And here is an example where the @sc{gdb/mi} command is known
38433 -info-gdb-mi-command symbol-list-lines
38434 ^done,command=@{exists="true"@}
38437 @findex -list-features
38438 @cindex supported @sc{gdb/mi} features, list
38439 @subheading The @code{-list-features} Command
38441 Returns a list of particular features of the MI protocol that
38442 this version of gdb implements. A feature can be a command,
38443 or a new field in an output of some command, or even an
38444 important bugfix. While a frontend can sometimes detect presence
38445 of a feature at runtime, it is easier to perform detection at debugger
38448 The command returns a list of strings, with each string naming an
38449 available feature. Each returned string is just a name, it does not
38450 have any internal structure. The list of possible feature names
38456 (gdb) -list-features
38457 ^done,result=["feature1","feature2"]
38460 The current list of features is:
38463 @item frozen-varobjs
38464 Indicates support for the @code{-var-set-frozen} command, as well
38465 as possible presence of the @code{frozen} field in the output
38466 of @code{-varobj-create}.
38467 @item pending-breakpoints
38468 Indicates support for the @option{-f} option to the @code{-break-insert}
38471 Indicates Python scripting support, Python-based
38472 pretty-printing commands, and possible presence of the
38473 @samp{display_hint} field in the output of @code{-var-list-children}
38475 Indicates support for the @code{-thread-info} command.
38476 @item data-read-memory-bytes
38477 Indicates support for the @code{-data-read-memory-bytes} and the
38478 @code{-data-write-memory-bytes} commands.
38479 @item breakpoint-notifications
38480 Indicates that changes to breakpoints and breakpoints created via the
38481 CLI will be announced via async records.
38482 @item ada-task-info
38483 Indicates support for the @code{-ada-task-info} command.
38484 @item language-option
38485 Indicates that all @sc{gdb/mi} commands accept the @option{--language}
38486 option (@pxref{Context management}).
38487 @item info-gdb-mi-command
38488 Indicates support for the @code{-info-gdb-mi-command} command.
38489 @item undefined-command-error-code
38490 Indicates support for the "undefined-command" error code in error result
38491 records, produced when trying to execute an undefined @sc{gdb/mi} command
38492 (@pxref{GDB/MI Result Records}).
38493 @item exec-run-start-option
38494 Indicates that the @code{-exec-run} command supports the @option{--start}
38495 option (@pxref{GDB/MI Program Execution}).
38496 @item data-disassemble-a-option
38497 Indicates that the @code{-data-disassemble} command supports the @option{-a}
38498 option (@pxref{GDB/MI Data Manipulation}).
38499 @item simple-values-ref-types
38500 Indicates that the @code{--simple-values} argument to the
38501 @code{-stack-list-arguments}, @code{-stack-list-locals},
38502 @code{-stack-list-variables}, and @code{-var-list-children} commands
38503 takes reference types into account: that is, a value is considered
38504 simple if it is neither an array, structure, or union, nor a reference
38505 to an array, structure, or union.
38508 @findex -list-target-features
38509 @subheading The @code{-list-target-features} Command
38511 Returns a list of particular features that are supported by the
38512 target. Those features affect the permitted MI commands, but
38513 unlike the features reported by the @code{-list-features} command, the
38514 features depend on which target GDB is using at the moment. Whenever
38515 a target can change, due to commands such as @code{-target-select},
38516 @code{-target-attach} or @code{-exec-run}, the list of target features
38517 may change, and the frontend should obtain it again.
38521 (gdb) -list-target-features
38522 ^done,result=["async"]
38525 The current list of features is:
38529 Indicates that the target is capable of asynchronous command
38530 execution, which means that @value{GDBN} will accept further commands
38531 while the target is running.
38534 Indicates that the target is capable of reverse execution.
38535 @xref{Reverse Execution}, for more information.
38539 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
38540 @node GDB/MI Miscellaneous Commands
38541 @section Miscellaneous @sc{gdb/mi} Commands
38543 @c @subheading -gdb-complete
38546 @subheading The @code{-gdb-exit} Command
38548 @subsubheading Synopsis
38554 Exit @value{GDBN} immediately.
38556 @subsubheading @value{GDBN} Command
38558 Approximately corresponds to @samp{quit}.
38560 @subsubheading Example
38570 @findex -exec-abort
38571 @subheading The @code{-exec-abort} Command
38573 @subsubheading Synopsis
38579 Kill the inferior running program.
38581 @subsubheading @value{GDBN} Command
38583 The corresponding @value{GDBN} command is @samp{kill}.
38585 @subsubheading Example
38591 @subheading The @code{-gdb-set} Command
38593 @subsubheading Synopsis
38599 Set an internal @value{GDBN} variable.
38600 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
38602 @subsubheading @value{GDBN} Command
38604 The corresponding @value{GDBN} command is @samp{set}.
38606 @subsubheading Example
38617 @subheading The @code{-gdb-show} Command
38619 @subsubheading Synopsis
38625 Show the current value of a @value{GDBN} variable.
38627 @subsubheading @value{GDBN} Command
38629 The corresponding @value{GDBN} command is @samp{show}.
38631 @subsubheading Example
38640 @c @subheading -gdb-source
38643 @findex -gdb-version
38644 @subheading The @code{-gdb-version} Command
38646 @subsubheading Synopsis
38652 Show version information for @value{GDBN}. Used mostly in testing.
38654 @subsubheading @value{GDBN} Command
38656 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
38657 default shows this information when you start an interactive session.
38659 @subsubheading Example
38661 @c This example modifies the actual output from GDB to avoid overfull
38667 ~Copyright 2000 Free Software Foundation, Inc.
38668 ~GDB is free software, covered by the GNU General Public License, and
38669 ~you are welcome to change it and/or distribute copies of it under
38670 ~ certain conditions.
38671 ~Type "show copying" to see the conditions.
38672 ~There is absolutely no warranty for GDB. Type "show warranty" for
38674 ~This GDB was configured as
38675 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
38680 @findex -list-thread-groups
38681 @subheading The @code{-list-thread-groups} Command
38683 @subsubheading Synopsis
38686 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
38689 Lists thread groups (@pxref{Thread groups}). When a single thread
38690 group is passed as the argument, lists the children of that group.
38691 When several thread group are passed, lists information about those
38692 thread groups. Without any parameters, lists information about all
38693 top-level thread groups.
38695 Normally, thread groups that are being debugged are reported.
38696 With the @samp{--available} option, @value{GDBN} reports thread groups
38697 available on the target.
38699 The output of this command may have either a @samp{threads} result or
38700 a @samp{groups} result. The @samp{thread} result has a list of tuples
38701 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
38702 Information}). The @samp{groups} result has a list of tuples as value,
38703 each tuple describing a thread group. If top-level groups are
38704 requested (that is, no parameter is passed), or when several groups
38705 are passed, the output always has a @samp{groups} result. The format
38706 of the @samp{group} result is described below.
38708 To reduce the number of roundtrips it's possible to list thread groups
38709 together with their children, by passing the @samp{--recurse} option
38710 and the recursion depth. Presently, only recursion depth of 1 is
38711 permitted. If this option is present, then every reported thread group
38712 will also include its children, either as @samp{group} or
38713 @samp{threads} field.
38715 In general, any combination of option and parameters is permitted, with
38716 the following caveats:
38720 When a single thread group is passed, the output will typically
38721 be the @samp{threads} result. Because threads may not contain
38722 anything, the @samp{recurse} option will be ignored.
38725 When the @samp{--available} option is passed, limited information may
38726 be available. In particular, the list of threads of a process might
38727 be inaccessible. Further, specifying specific thread groups might
38728 not give any performance advantage over listing all thread groups.
38729 The frontend should assume that @samp{-list-thread-groups --available}
38730 is always an expensive operation and cache the results.
38734 The @samp{groups} result is a list of tuples, where each tuple may
38735 have the following fields:
38739 Identifier of the thread group. This field is always present.
38740 The identifier is an opaque string; frontends should not try to
38741 convert it to an integer, even though it might look like one.
38744 The type of the thread group. At present, only @samp{process} is a
38748 The target-specific process identifier. This field is only present
38749 for thread groups of type @samp{process} and only if the process exists.
38752 The exit code of this group's last exited thread, formatted in octal.
38753 This field is only present for thread groups of type @samp{process} and
38754 only if the process is not running.
38757 The number of children this thread group has. This field may be
38758 absent for an available thread group.
38761 This field has a list of tuples as value, each tuple describing a
38762 thread. It may be present if the @samp{--recurse} option is
38763 specified, and it's actually possible to obtain the threads.
38766 This field is a list of integers, each identifying a core that one
38767 thread of the group is running on. This field may be absent if
38768 such information is not available.
38771 The name of the executable file that corresponds to this thread group.
38772 The field is only present for thread groups of type @samp{process},
38773 and only if there is a corresponding executable file.
38777 @subsubheading Example
38781 -list-thread-groups
38782 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
38783 -list-thread-groups 17
38784 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
38785 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
38786 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
38787 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
38788 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},state="running"@}]]
38789 -list-thread-groups --available
38790 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
38791 -list-thread-groups --available --recurse 1
38792 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
38793 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
38794 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
38795 -list-thread-groups --available --recurse 1 17 18
38796 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
38797 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
38798 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
38802 @subheading The @code{-info-os} Command
38804 @subsubheading Synopsis
38807 -info-os [ @var{type} ]
38810 If no argument is supplied, the command returns a table of available
38811 operating-system-specific information types. If one of these types is
38812 supplied as an argument @var{type}, then the command returns a table
38813 of data of that type.
38815 The types of information available depend on the target operating
38818 @subsubheading @value{GDBN} Command
38820 The corresponding @value{GDBN} command is @samp{info os}.
38822 @subsubheading Example
38824 When run on a @sc{gnu}/Linux system, the output will look something
38830 ^done,OSDataTable=@{nr_rows="10",nr_cols="3",
38831 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
38832 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
38833 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
38834 body=[item=@{col0="cpus",col1="Listing of all cpus/cores on the system",
38836 item=@{col0="files",col1="Listing of all file descriptors",
38837 col2="File descriptors"@},
38838 item=@{col0="modules",col1="Listing of all loaded kernel modules",
38839 col2="Kernel modules"@},
38840 item=@{col0="msg",col1="Listing of all message queues",
38841 col2="Message queues"@},
38842 item=@{col0="processes",col1="Listing of all processes",
38843 col2="Processes"@},
38844 item=@{col0="procgroups",col1="Listing of all process groups",
38845 col2="Process groups"@},
38846 item=@{col0="semaphores",col1="Listing of all semaphores",
38847 col2="Semaphores"@},
38848 item=@{col0="shm",col1="Listing of all shared-memory regions",
38849 col2="Shared-memory regions"@},
38850 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
38852 item=@{col0="threads",col1="Listing of all threads",
38856 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
38857 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
38858 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
38859 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
38860 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
38861 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
38862 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
38863 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
38865 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
38866 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
38870 (Note that the MI output here includes a @code{"Title"} column that
38871 does not appear in command-line @code{info os}; this column is useful
38872 for MI clients that want to enumerate the types of data, such as in a
38873 popup menu, but is needless clutter on the command line, and
38874 @code{info os} omits it.)
38876 @findex -add-inferior
38877 @subheading The @code{-add-inferior} Command
38879 @subsubheading Synopsis
38882 -add-inferior [ --no-connection ]
38885 Creates a new inferior (@pxref{Inferiors Connections and Programs}). The created
38886 inferior is not associated with any executable. Such association may
38887 be established with the @samp{-file-exec-and-symbols} command
38888 (@pxref{GDB/MI File Commands}).
38890 By default, the new inferior begins connected to the same target
38891 connection as the current inferior. For example, if the current
38892 inferior was connected to @code{gdbserver} with @code{target remote},
38893 then the new inferior will be connected to the same @code{gdbserver}
38894 instance. The @samp{--no-connection} option starts the new inferior
38895 with no connection yet. You can then for example use the
38896 @code{-target-select remote} command to connect to some other
38897 @code{gdbserver} instance, use @code{-exec-run} to spawn a local
38900 The command response always has a field, @var{inferior}, whose value
38901 is the identifier of the thread group corresponding to the new
38904 An additional section field, @var{connection}, is optional. This
38905 field will only exist if the new inferior has a target connection. If
38906 this field exists, then its value will be a tuple containing the
38911 The number of the connection used for the new inferior.
38914 The name of the connection type used for the new inferior.
38917 @subsubheading @value{GDBN} Command
38919 The corresponding @value{GDBN} command is @samp{add-inferior}
38920 (@pxref{add_inferior_cli,,@samp{add-inferior}}).
38922 @subsubheading Example
38927 ^done,inferior="i3"
38930 @findex -remove-inferior
38931 @subheading The @code{-remove-inferior} Command
38933 @subsubheading Synopsis
38936 -remove-inferior @var{inferior-id}
38939 Removes an inferior (@pxref{Inferiors Connections and Programs}).
38940 Only inferiors that have exited can be removed. The @var{inferior-id}
38941 is the inferior to be removed, and should be the same id string as
38942 returned by the @samp{-add-inferior} command.
38944 When an inferior is successfully removed a
38945 @code{=thread-group-removed} notification (@pxref{GDB/MI Async
38946 Records}) is emitted, the @var{id} field of which contains the
38947 @var{inferior-id} for the removed inferior.
38949 @subsubheading @value{GDBN} Command
38951 The corresponding @value{GDBN} command is @samp{remove-inferiors}
38952 (@pxref{remove_inferiors_cli,,@samp{remove-inferiors}}).
38954 @subsubheading Example
38958 -remove-inferior i3
38959 =thread-group-removed,id="i3"
38963 @findex -interpreter-exec
38964 @subheading The @code{-interpreter-exec} Command
38966 @subsubheading Synopsis
38969 -interpreter-exec @var{interpreter} @var{command}
38971 @anchor{-interpreter-exec}
38973 Execute the specified @var{command} in the given @var{interpreter}.
38975 @subsubheading @value{GDBN} Command
38977 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
38979 @subsubheading Example
38983 -interpreter-exec console "break main"
38984 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
38985 &"During symbol reading, bad structure-type format.\n"
38986 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
38991 @findex -inferior-tty-set
38992 @subheading The @code{-inferior-tty-set} Command
38994 @subsubheading Synopsis
38997 -inferior-tty-set /dev/pts/1
39000 Set terminal for future runs of the program being debugged.
39002 @subsubheading @value{GDBN} Command
39004 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
39006 @subsubheading Example
39010 -inferior-tty-set /dev/pts/1
39015 @findex -inferior-tty-show
39016 @subheading The @code{-inferior-tty-show} Command
39018 @subsubheading Synopsis
39024 Show terminal for future runs of program being debugged.
39026 @subsubheading @value{GDBN} Command
39028 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
39030 @subsubheading Example
39034 -inferior-tty-set /dev/pts/1
39038 ^done,inferior_tty_terminal="/dev/pts/1"
39042 @findex -enable-timings
39043 @subheading The @code{-enable-timings} Command
39045 @subsubheading Synopsis
39048 -enable-timings [yes | no]
39051 Toggle the printing of the wallclock, user and system times for an MI
39052 command as a field in its output. This command is to help frontend
39053 developers optimize the performance of their code. No argument is
39054 equivalent to @samp{yes}.
39056 @subsubheading @value{GDBN} Command
39060 @subsubheading Example
39068 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
39069 addr="0x080484ed",func="main",file="myprog.c",
39070 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
39072 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
39080 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
39081 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
39082 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
39083 fullname="/home/nickrob/myprog.c",line="73",arch="i386:x86_64"@}
39088 @subheading The @code{-complete} Command
39090 @subsubheading Synopsis
39093 -complete @var{command}
39096 Show a list of completions for partially typed CLI @var{command}.
39098 This command is intended for @sc{gdb/mi} frontends that cannot use two separate
39099 CLI and MI channels --- for example: because of lack of PTYs like on Windows or
39100 because @value{GDBN} is used remotely via a SSH connection.
39102 @subsubheading Result
39104 The result consists of two or three fields:
39108 This field contains the completed @var{command}. If @var{command}
39109 has no known completions, this field is omitted.
39112 This field contains a (possibly empty) array of matches. It is always present.
39114 @item max_completions_reached
39115 This field contains @code{1} if number of known completions is above
39116 @code{max-completions} limit (@pxref{Completion}), otherwise it contains
39117 @code{0}. It is always present.
39121 @subsubheading @value{GDBN} Command
39123 The corresponding @value{GDBN} command is @samp{complete}.
39125 @subsubheading Example
39130 ^done,completion="break",
39131 matches=["break","break-range"],
39132 max_completions_reached="0"
39135 ^done,completion="b ma",
39136 matches=["b madvise","b main"],max_completions_reached="0"
39138 -complete "b push_b"
39139 ^done,completion="b push_back(",
39141 "b A::push_back(void*)",
39142 "b std::string::push_back(char)",
39143 "b std::vector<int, std::allocator<int> >::push_back(int&&)"],
39144 max_completions_reached="0"
39146 -complete "nonexist"
39147 ^done,matches=[],max_completions_reached="0"
39153 @chapter @value{GDBN} Annotations
39155 This chapter describes annotations in @value{GDBN}. Annotations were
39156 designed to interface @value{GDBN} to graphical user interfaces or other
39157 similar programs which want to interact with @value{GDBN} at a
39158 relatively high level.
39160 The annotation mechanism has largely been superseded by @sc{gdb/mi}
39164 This is Edition @value{EDITION}, @value{DATE}.
39168 * Annotations Overview:: What annotations are; the general syntax.
39169 * Server Prefix:: Issuing a command without affecting user state.
39170 * Prompting:: Annotations marking @value{GDBN}'s need for input.
39171 * Errors:: Annotations for error messages.
39172 * Invalidation:: Some annotations describe things now invalid.
39173 * Annotations for Running::
39174 Whether the program is running, how it stopped, etc.
39175 * Source Annotations:: Annotations describing source code.
39178 @node Annotations Overview
39179 @section What is an Annotation?
39180 @cindex annotations
39182 Annotations start with a newline character, two @samp{control-z}
39183 characters, and the name of the annotation. If there is no additional
39184 information associated with this annotation, the name of the annotation
39185 is followed immediately by a newline. If there is additional
39186 information, the name of the annotation is followed by a space, the
39187 additional information, and a newline. The additional information
39188 cannot contain newline characters.
39190 Any output not beginning with a newline and two @samp{control-z}
39191 characters denotes literal output from @value{GDBN}. Currently there is
39192 no need for @value{GDBN} to output a newline followed by two
39193 @samp{control-z} characters, but if there was such a need, the
39194 annotations could be extended with an @samp{escape} annotation which
39195 means those three characters as output.
39197 The annotation @var{level}, which is specified using the
39198 @option{--annotate} command line option (@pxref{Mode Options}), controls
39199 how much information @value{GDBN} prints together with its prompt,
39200 values of expressions, source lines, and other types of output. Level 0
39201 is for no annotations, level 1 is for use when @value{GDBN} is run as a
39202 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
39203 for programs that control @value{GDBN}, and level 2 annotations have
39204 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
39205 Interface, annotate, GDB's Obsolete Annotations}).
39208 @kindex set annotate
39209 @item set annotate @var{level}
39210 The @value{GDBN} command @code{set annotate} sets the level of
39211 annotations to the specified @var{level}.
39213 @item show annotate
39214 @kindex show annotate
39215 Show the current annotation level.
39218 This chapter describes level 3 annotations.
39220 A simple example of starting up @value{GDBN} with annotations is:
39223 $ @kbd{gdb --annotate=3}
39225 Copyright 2003 Free Software Foundation, Inc.
39226 GDB is free software, covered by the GNU General Public License,
39227 and you are welcome to change it and/or distribute copies of it
39228 under certain conditions.
39229 Type "show copying" to see the conditions.
39230 There is absolutely no warranty for GDB. Type "show warranty"
39232 This GDB was configured as "i386-pc-linux-gnu"
39243 Here @samp{quit} is input to @value{GDBN}; the rest is output from
39244 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
39245 denotes a @samp{control-z} character) are annotations; the rest is
39246 output from @value{GDBN}.
39248 @node Server Prefix
39249 @section The Server Prefix
39250 @cindex server prefix
39252 If you prefix a command with @samp{server } then it will not affect
39253 the command history, nor will it affect @value{GDBN}'s notion of which
39254 command to repeat if @key{RET} is pressed on a line by itself. This
39255 means that commands can be run behind a user's back by a front-end in
39256 a transparent manner.
39258 The @code{server } prefix does not affect the recording of values into
39259 the value history; to print a value without recording it into the
39260 value history, use the @code{output} command instead of the
39261 @code{print} command.
39263 Using this prefix also disables confirmation requests
39264 (@pxref{confirmation requests}).
39267 @section Annotation for @value{GDBN} Input
39269 @cindex annotations for prompts
39270 When @value{GDBN} prompts for input, it annotates this fact so it is possible
39271 to know when to send output, when the output from a given command is
39274 Different kinds of input each have a different @dfn{input type}. Each
39275 input type has three annotations: a @code{pre-} annotation, which
39276 denotes the beginning of any prompt which is being output, a plain
39277 annotation, which denotes the end of the prompt, and then a @code{post-}
39278 annotation which denotes the end of any echo which may (or may not) be
39279 associated with the input. For example, the @code{prompt} input type
39280 features the following annotations:
39288 The input types are
39291 @findex pre-prompt annotation
39292 @findex prompt annotation
39293 @findex post-prompt annotation
39295 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
39297 @findex pre-commands annotation
39298 @findex commands annotation
39299 @findex post-commands annotation
39301 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
39302 command. The annotations are repeated for each command which is input.
39304 @findex pre-overload-choice annotation
39305 @findex overload-choice annotation
39306 @findex post-overload-choice annotation
39307 @item overload-choice
39308 When @value{GDBN} wants the user to select between various overloaded functions.
39310 @findex pre-query annotation
39311 @findex query annotation
39312 @findex post-query annotation
39314 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
39316 @findex pre-prompt-for-continue annotation
39317 @findex prompt-for-continue annotation
39318 @findex post-prompt-for-continue annotation
39319 @item prompt-for-continue
39320 When @value{GDBN} is asking the user to press return to continue. Note: Don't
39321 expect this to work well; instead use @code{set height 0} to disable
39322 prompting. This is because the counting of lines is buggy in the
39323 presence of annotations.
39328 @cindex annotations for errors, warnings and interrupts
39330 @findex quit annotation
39335 This annotation occurs right before @value{GDBN} responds to an interrupt.
39337 @findex error annotation
39342 This annotation occurs right before @value{GDBN} responds to an error.
39344 Quit and error annotations indicate that any annotations which @value{GDBN} was
39345 in the middle of may end abruptly. For example, if a
39346 @code{value-history-begin} annotation is followed by a @code{error}, one
39347 cannot expect to receive the matching @code{value-history-end}. One
39348 cannot expect not to receive it either, however; an error annotation
39349 does not necessarily mean that @value{GDBN} is immediately returning all the way
39352 @findex error-begin annotation
39353 A quit or error annotation may be preceded by
39359 Any output between that and the quit or error annotation is the error
39362 Warning messages are not yet annotated.
39363 @c If we want to change that, need to fix warning(), type_error(),
39364 @c range_error(), and possibly other places.
39367 @section Invalidation Notices
39369 @cindex annotations for invalidation messages
39370 The following annotations say that certain pieces of state may have
39374 @findex frames-invalid annotation
39375 @item ^Z^Zframes-invalid
39377 The frames (for example, output from the @code{backtrace} command) may
39380 @findex breakpoints-invalid annotation
39381 @item ^Z^Zbreakpoints-invalid
39383 The breakpoints may have changed. For example, the user just added or
39384 deleted a breakpoint.
39387 @node Annotations for Running
39388 @section Running the Program
39389 @cindex annotations for running programs
39391 @findex starting annotation
39392 @findex stopping annotation
39393 When the program starts executing due to a @value{GDBN} command such as
39394 @code{step} or @code{continue},
39400 is output. When the program stops,
39406 is output. Before the @code{stopped} annotation, a variety of
39407 annotations describe how the program stopped.
39410 @findex exited annotation
39411 @item ^Z^Zexited @var{exit-status}
39412 The program exited, and @var{exit-status} is the exit status (zero for
39413 successful exit, otherwise nonzero).
39415 @findex signalled annotation
39416 @findex signal-name annotation
39417 @findex signal-name-end annotation
39418 @findex signal-string annotation
39419 @findex signal-string-end annotation
39420 @item ^Z^Zsignalled
39421 The program exited with a signal. After the @code{^Z^Zsignalled}, the
39422 annotation continues:
39428 ^Z^Zsignal-name-end
39432 ^Z^Zsignal-string-end
39437 where @var{name} is the name of the signal, such as @code{SIGILL} or
39438 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
39439 as @code{Illegal Instruction} or @code{Segmentation fault}. The arguments
39440 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
39441 user's benefit and have no particular format.
39443 @findex signal annotation
39445 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
39446 just saying that the program received the signal, not that it was
39447 terminated with it.
39449 @findex breakpoint annotation
39450 @item ^Z^Zbreakpoint @var{number}
39451 The program hit breakpoint number @var{number}.
39453 @findex watchpoint annotation
39454 @item ^Z^Zwatchpoint @var{number}
39455 The program hit watchpoint number @var{number}.
39458 @node Source Annotations
39459 @section Displaying Source
39460 @cindex annotations for source display
39462 @findex source annotation
39463 The following annotation is used instead of displaying source code:
39466 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
39469 where @var{filename} is an absolute file name indicating which source
39470 file, @var{line} is the line number within that file (where 1 is the
39471 first line in the file), @var{character} is the character position
39472 within the file (where 0 is the first character in the file) (for most
39473 debug formats this will necessarily point to the beginning of a line),
39474 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
39475 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
39476 @var{addr} is the address in the target program associated with the
39477 source which is being displayed. The @var{addr} is in the form @samp{0x}
39478 followed by one or more lowercase hex digits (note that this does not
39479 depend on the language).
39481 @node Debugger Adapter Protocol
39482 @chapter Debugger Adapter Protocol
39484 Generally, @value{GDBN} implements the Debugger Adapter Protocol as
39485 written. However, in some cases, extensions are either needed or even
39488 @value{GDBN} defines some parameters that can be passed to the
39489 @code{launch} request:
39493 If provided, this should be an array of strings. These strings are
39494 provided as command-line arguments to the inferior, as if by
39495 @code{set args}. @xref{Arguments}.
39498 If provided, this should be a string. @value{GDBN} will change its
39499 working directory to this directory, as if by the @code{cd} command
39500 (@pxref{Working Directory}). The launched program will inherit this
39501 as its working directory. Note that change of directory happens
39502 before the @code{program} parameter is processed. This will affect
39503 the result if @code{program} is a relative filename.
39506 If provided, this should be an object. Each key of the object will be
39507 used as the name of an environment variable; each value must be a
39508 string and will be the value of that variable. The environment of the
39509 inferior will be set to exactly as passed in. @xref{Environment}.
39512 If provided, this is a string that specifies the program to use. This
39513 corresponds to the @code{file} command. @xref{Files}.
39515 @item stopAtBeginningOfMainSubprogram
39516 If provided, this must be a boolean. When @samp{True}, @value{GDBN}
39517 will set a temporary breakpoint at the program's main procedure, using
39518 the same approach as the @code{start} command. @xref{Starting}.
39521 @value{GDBN} defines some parameters that can be passed to the
39522 @code{attach} request. One of these must be specified.
39526 The process ID to which @value{GDBN} should attach. @xref{Attach}.
39529 The target to which @value{GDBN} should connect. This is a string and
39530 is passed to the @code{target remote} command. @xref{Connecting}.
39533 In response to the @code{disassemble} request, DAP allows the client
39534 to return the bytes of each instruction in an implementation-defined
39535 format. @value{GDBN} implements this by sending a string with the
39536 bytes encoded in hex, like @code{"55a2b900"}.
39538 @node JIT Interface
39539 @chapter JIT Compilation Interface
39540 @cindex just-in-time compilation
39541 @cindex JIT compilation interface
39543 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
39544 interface. A JIT compiler is a program or library that generates native
39545 executable code at runtime and executes it, usually in order to achieve good
39546 performance while maintaining platform independence.
39548 Programs that use JIT compilation are normally difficult to debug because
39549 portions of their code are generated at runtime, instead of being loaded from
39550 object files, which is where @value{GDBN} normally finds the program's symbols
39551 and debug information. In order to debug programs that use JIT compilation,
39552 @value{GDBN} has an interface that allows the program to register in-memory
39553 symbol files with @value{GDBN} at runtime.
39555 If you are using @value{GDBN} to debug a program that uses this interface, then
39556 it should work transparently so long as you have not stripped the binary. If
39557 you are developing a JIT compiler, then the interface is documented in the rest
39558 of this chapter. At this time, the only known client of this interface is the
39561 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
39562 JIT compiler communicates with @value{GDBN} by writing data into a global
39563 variable and calling a function at a well-known symbol. When @value{GDBN}
39564 attaches, it reads a linked list of symbol files from the global variable to
39565 find existing code, and puts a breakpoint in the function so that it can find
39566 out about additional code.
39569 * Declarations:: Relevant C struct declarations
39570 * Registering Code:: Steps to register code
39571 * Unregistering Code:: Steps to unregister code
39572 * Custom Debug Info:: Emit debug information in a custom format
39576 @section JIT Declarations
39578 These are the relevant struct declarations that a C program should include to
39579 implement the interface:
39589 struct jit_code_entry
39591 struct jit_code_entry *next_entry;
39592 struct jit_code_entry *prev_entry;
39593 const char *symfile_addr;
39594 uint64_t symfile_size;
39597 struct jit_descriptor
39600 /* This type should be jit_actions_t, but we use uint32_t
39601 to be explicit about the bitwidth. */
39602 uint32_t action_flag;
39603 struct jit_code_entry *relevant_entry;
39604 struct jit_code_entry *first_entry;
39607 /* GDB puts a breakpoint in this function. */
39608 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
39610 /* Make sure to specify the version statically, because the
39611 debugger may check the version before we can set it. */
39612 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
39615 If the JIT is multi-threaded, then it is important that the JIT synchronize any
39616 modifications to this global data properly, which can easily be done by putting
39617 a global mutex around modifications to these structures.
39619 @node Registering Code
39620 @section Registering Code
39622 To register code with @value{GDBN}, the JIT should follow this protocol:
39626 Generate an object file in memory with symbols and other desired debug
39627 information. The file must include the virtual addresses of the sections.
39630 Create a code entry for the file, which gives the start and size of the symbol
39634 Add it to the linked list in the JIT descriptor.
39637 Point the relevant_entry field of the descriptor at the entry.
39640 Set @code{action_flag} to @code{JIT_REGISTER} and call
39641 @code{__jit_debug_register_code}.
39644 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
39645 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
39646 new code. However, the linked list must still be maintained in order to allow
39647 @value{GDBN} to attach to a running process and still find the symbol files.
39649 @node Unregistering Code
39650 @section Unregistering Code
39652 If code is freed, then the JIT should use the following protocol:
39656 Remove the code entry corresponding to the code from the linked list.
39659 Point the @code{relevant_entry} field of the descriptor at the code entry.
39662 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
39663 @code{__jit_debug_register_code}.
39666 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
39667 and the JIT will leak the memory used for the associated symbol files.
39669 @node Custom Debug Info
39670 @section Custom Debug Info
39671 @cindex custom JIT debug info
39672 @cindex JIT debug info reader
39674 Generating debug information in platform-native file formats (like ELF
39675 or COFF) may be an overkill for JIT compilers; especially if all the
39676 debug info is used for is displaying a meaningful backtrace. The
39677 issue can be resolved by having the JIT writers decide on a debug info
39678 format and also provide a reader that parses the debug info generated
39679 by the JIT compiler. This section gives a brief overview on writing
39680 such a parser. More specific details can be found in the source file
39681 @file{gdb/jit-reader.in}, which is also installed as a header at
39682 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
39684 The reader is implemented as a shared object (so this functionality is
39685 not available on platforms which don't allow loading shared objects at
39686 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
39687 @code{jit-reader-unload} are provided, to be used to load and unload
39688 the readers from a preconfigured directory. Once loaded, the shared
39689 object is used the parse the debug information emitted by the JIT
39693 * Using JIT Debug Info Readers:: How to use supplied readers correctly
39694 * Writing JIT Debug Info Readers:: Creating a debug-info reader
39697 @node Using JIT Debug Info Readers
39698 @subsection Using JIT Debug Info Readers
39699 @kindex jit-reader-load
39700 @kindex jit-reader-unload
39702 Readers can be loaded and unloaded using the @code{jit-reader-load}
39703 and @code{jit-reader-unload} commands.
39706 @item jit-reader-load @var{reader}
39707 Load the JIT reader named @var{reader}, which is a shared
39708 object specified as either an absolute or a relative file name. In
39709 the latter case, @value{GDBN} will try to load the reader from a
39710 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
39711 system (here @var{libdir} is the system library directory, often
39712 @file{/usr/local/lib}).
39714 Only one reader can be active at a time; trying to load a second
39715 reader when one is already loaded will result in @value{GDBN}
39716 reporting an error. A new JIT reader can be loaded by first unloading
39717 the current one using @code{jit-reader-unload} and then invoking
39718 @code{jit-reader-load}.
39720 @item jit-reader-unload
39721 Unload the currently loaded JIT reader.
39725 @node Writing JIT Debug Info Readers
39726 @subsection Writing JIT Debug Info Readers
39727 @cindex writing JIT debug info readers
39729 As mentioned, a reader is essentially a shared object conforming to a
39730 certain ABI. This ABI is described in @file{jit-reader.h}.
39732 @file{jit-reader.h} defines the structures, macros and functions
39733 required to write a reader. It is installed (along with
39734 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
39735 the system include directory.
39737 Readers need to be released under a GPL compatible license. A reader
39738 can be declared as released under such a license by placing the macro
39739 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
39741 The entry point for readers is the symbol @code{gdb_init_reader},
39742 which is expected to be a function with the prototype
39744 @findex gdb_init_reader
39746 extern struct gdb_reader_funcs *gdb_init_reader (void);
39749 @cindex @code{struct gdb_reader_funcs}
39751 @code{struct gdb_reader_funcs} contains a set of pointers to callback
39752 functions. These functions are executed to read the debug info
39753 generated by the JIT compiler (@code{read}), to unwind stack frames
39754 (@code{unwind}) and to create canonical frame IDs
39755 (@code{get_frame_id}). It also has a callback that is called when the
39756 reader is being unloaded (@code{destroy}). The struct looks like this
39759 struct gdb_reader_funcs
39761 /* Must be set to GDB_READER_INTERFACE_VERSION. */
39762 int reader_version;
39764 /* For use by the reader. */
39767 gdb_read_debug_info *read;
39768 gdb_unwind_frame *unwind;
39769 gdb_get_frame_id *get_frame_id;
39770 gdb_destroy_reader *destroy;
39774 @cindex @code{struct gdb_symbol_callbacks}
39775 @cindex @code{struct gdb_unwind_callbacks}
39777 The callbacks are provided with another set of callbacks by
39778 @value{GDBN} to do their job. For @code{read}, these callbacks are
39779 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
39780 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
39781 @code{struct gdb_symbol_callbacks} has callbacks to create new object
39782 files and new symbol tables inside those object files. @code{struct
39783 gdb_unwind_callbacks} has callbacks to read registers off the current
39784 frame and to write out the values of the registers in the previous
39785 frame. Both have a callback (@code{target_read}) to read bytes off the
39786 target's address space.
39788 @node In-Process Agent
39789 @chapter In-Process Agent
39790 @cindex debugging agent
39791 The traditional debugging model is conceptually low-speed, but works fine,
39792 because most bugs can be reproduced in debugging-mode execution. However,
39793 as multi-core or many-core processors are becoming mainstream, and
39794 multi-threaded programs become more and more popular, there should be more
39795 and more bugs that only manifest themselves at normal-mode execution, for
39796 example, thread races, because debugger's interference with the program's
39797 timing may conceal the bugs. On the other hand, in some applications,
39798 it is not feasible for the debugger to interrupt the program's execution
39799 long enough for the developer to learn anything helpful about its behavior.
39800 If the program's correctness depends on its real-time behavior, delays
39801 introduced by a debugger might cause the program to fail, even when the
39802 code itself is correct. It is useful to be able to observe the program's
39803 behavior without interrupting it.
39805 Therefore, traditional debugging model is too intrusive to reproduce
39806 some bugs. In order to reduce the interference with the program, we can
39807 reduce the number of operations performed by debugger. The
39808 @dfn{In-Process Agent}, a shared library, is running within the same
39809 process with inferior, and is able to perform some debugging operations
39810 itself. As a result, debugger is only involved when necessary, and
39811 performance of debugging can be improved accordingly. Note that
39812 interference with program can be reduced but can't be removed completely,
39813 because the in-process agent will still stop or slow down the program.
39815 The in-process agent can interpret and execute Agent Expressions
39816 (@pxref{Agent Expressions}) during performing debugging operations. The
39817 agent expressions can be used for different purposes, such as collecting
39818 data in tracepoints, and condition evaluation in breakpoints.
39820 @anchor{Control Agent}
39821 You can control whether the in-process agent is used as an aid for
39822 debugging with the following commands:
39825 @kindex set agent on
39827 Causes the in-process agent to perform some operations on behalf of the
39828 debugger. Just which operations requested by the user will be done
39829 by the in-process agent depends on the its capabilities. For example,
39830 if you request to evaluate breakpoint conditions in the in-process agent,
39831 and the in-process agent has such capability as well, then breakpoint
39832 conditions will be evaluated in the in-process agent.
39834 @kindex set agent off
39835 @item set agent off
39836 Disables execution of debugging operations by the in-process agent. All
39837 of the operations will be performed by @value{GDBN}.
39841 Display the current setting of execution of debugging operations by
39842 the in-process agent.
39846 * In-Process Agent Protocol::
39849 @node In-Process Agent Protocol
39850 @section In-Process Agent Protocol
39851 @cindex in-process agent protocol
39853 The in-process agent is able to communicate with both @value{GDBN} and
39854 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
39855 used for communications between @value{GDBN} or GDBserver and the IPA.
39856 In general, @value{GDBN} or GDBserver sends commands
39857 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
39858 in-process agent replies back with the return result of the command, or
39859 some other information. The data sent to in-process agent is composed
39860 of primitive data types, such as 4-byte or 8-byte type, and composite
39861 types, which are called objects (@pxref{IPA Protocol Objects}).
39864 * IPA Protocol Objects::
39865 * IPA Protocol Commands::
39868 @node IPA Protocol Objects
39869 @subsection IPA Protocol Objects
39870 @cindex ipa protocol objects
39872 The commands sent to and results received from agent may contain some
39873 complex data types called @dfn{objects}.
39875 The in-process agent is running on the same machine with @value{GDBN}
39876 or GDBserver, so it doesn't have to handle as much differences between
39877 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
39878 However, there are still some differences of two ends in two processes:
39882 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
39883 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
39885 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
39886 GDBserver is compiled with one, and in-process agent is compiled with
39890 Here are the IPA Protocol Objects:
39894 agent expression object. It represents an agent expression
39895 (@pxref{Agent Expressions}).
39896 @anchor{agent expression object}
39898 tracepoint action object. It represents a tracepoint action
39899 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
39900 memory, static trace data and to evaluate expression.
39901 @anchor{tracepoint action object}
39903 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
39904 @anchor{tracepoint object}
39908 The following table describes important attributes of each IPA protocol
39911 @multitable @columnfractions .30 .20 .50
39912 @headitem Name @tab Size @tab Description
39913 @item @emph{agent expression object} @tab @tab
39914 @item length @tab 4 @tab length of bytes code
39915 @item byte code @tab @var{length} @tab contents of byte code
39916 @item @emph{tracepoint action for collecting memory} @tab @tab
39917 @item 'M' @tab 1 @tab type of tracepoint action
39918 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
39919 address of the lowest byte to collect, otherwise @var{addr} is the offset
39920 of @var{basereg} for memory collecting.
39921 @item len @tab 8 @tab length of memory for collecting
39922 @item basereg @tab 4 @tab the register number containing the starting
39923 memory address for collecting.
39924 @item @emph{tracepoint action for collecting registers} @tab @tab
39925 @item 'R' @tab 1 @tab type of tracepoint action
39926 @item @emph{tracepoint action for collecting static trace data} @tab @tab
39927 @item 'L' @tab 1 @tab type of tracepoint action
39928 @item @emph{tracepoint action for expression evaluation} @tab @tab
39929 @item 'X' @tab 1 @tab type of tracepoint action
39930 @item agent expression @tab length of @tab @ref{agent expression object}
39931 @item @emph{tracepoint object} @tab @tab
39932 @item number @tab 4 @tab number of tracepoint
39933 @item address @tab 8 @tab address of tracepoint inserted on
39934 @item type @tab 4 @tab type of tracepoint
39935 @item enabled @tab 1 @tab enable or disable of tracepoint
39936 @item step_count @tab 8 @tab step
39937 @item pass_count @tab 8 @tab pass
39938 @item numactions @tab 4 @tab number of tracepoint actions
39939 @item hit count @tab 8 @tab hit count
39940 @item trace frame usage @tab 8 @tab trace frame usage
39941 @item compiled_cond @tab 8 @tab compiled condition
39942 @item orig_size @tab 8 @tab orig size
39943 @item condition @tab 4 if condition is NULL otherwise length of
39944 @ref{agent expression object}
39945 @tab zero if condition is NULL, otherwise is
39946 @ref{agent expression object}
39947 @item actions @tab variable
39948 @tab numactions number of @ref{tracepoint action object}
39951 @node IPA Protocol Commands
39952 @subsection IPA Protocol Commands
39953 @cindex ipa protocol commands
39955 The spaces in each command are delimiters to ease reading this commands
39956 specification. They don't exist in real commands.
39960 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
39961 Installs a new fast tracepoint described by @var{tracepoint_object}
39962 (@pxref{tracepoint object}). The @var{gdb_jump_pad_head}, 8-byte long, is the
39963 head of @dfn{jumppad}, which is used to jump to data collection routine
39968 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
39969 @var{target_address} is address of tracepoint in the inferior.
39970 The @var{gdb_jump_pad_head} is updated head of jumppad. Both of
39971 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
39972 The @var{fjump} contains a sequence of instructions jump to jumppad entry.
39973 The @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
39980 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
39981 is about to kill inferiors.
39989 @item probe_marker_at:@var{address}
39990 Asks in-process agent to probe the marker at @var{address}.
39997 @item unprobe_marker_at:@var{address}
39998 Asks in-process agent to unprobe the marker at @var{address}.
40002 @chapter Reporting Bugs in @value{GDBN}
40003 @cindex bugs in @value{GDBN}
40004 @cindex reporting bugs in @value{GDBN}
40006 Your bug reports play an essential role in making @value{GDBN} reliable.
40008 Reporting a bug may help you by bringing a solution to your problem, or it
40009 may not. But in any case the principal function of a bug report is to help
40010 the entire community by making the next version of @value{GDBN} work better. Bug
40011 reports are your contribution to the maintenance of @value{GDBN}.
40013 In order for a bug report to serve its purpose, you must include the
40014 information that enables us to fix the bug.
40017 * Bug Criteria:: Have you found a bug?
40018 * Bug Reporting:: How to report bugs
40022 @section Have You Found a Bug?
40023 @cindex bug criteria
40025 If you are not sure whether you have found a bug, here are some guidelines:
40028 @cindex fatal signal
40029 @cindex debugger crash
40030 @cindex crash of debugger
40032 If the debugger gets a fatal signal, for any input whatever, that is a
40033 @value{GDBN} bug. Reliable debuggers never crash.
40035 @cindex error on valid input
40037 If @value{GDBN} produces an error message for valid input, that is a
40038 bug. (Note that if you're cross debugging, the problem may also be
40039 somewhere in the connection to the target.)
40041 @cindex invalid input
40043 If @value{GDBN} does not produce an error message for invalid input,
40044 that is a bug. However, you should note that your idea of
40045 ``invalid input'' might be our idea of ``an extension'' or ``support
40046 for traditional practice''.
40049 If you are an experienced user of debugging tools, your suggestions
40050 for improvement of @value{GDBN} are welcome in any case.
40053 @node Bug Reporting
40054 @section How to Report Bugs
40055 @cindex bug reports
40056 @cindex @value{GDBN} bugs, reporting
40058 A number of companies and individuals offer support for @sc{gnu} products.
40059 If you obtained @value{GDBN} from a support organization, we recommend you
40060 contact that organization first.
40062 You can find contact information for many support companies and
40063 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
40065 @c should add a web page ref...
40068 @ifset BUGURL_DEFAULT
40069 In any event, we also recommend that you submit bug reports for
40070 @value{GDBN}. The preferred method is to submit them directly using
40071 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
40072 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
40075 @strong{Do not send bug reports to @samp{info-gdb}, or to
40076 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
40077 not want to receive bug reports. Those that do have arranged to receive
40080 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
40081 serves as a repeater. The mailing list and the newsgroup carry exactly
40082 the same messages. Often people think of posting bug reports to the
40083 newsgroup instead of mailing them. This appears to work, but it has one
40084 problem which can be crucial: a newsgroup posting often lacks a mail
40085 path back to the sender. Thus, if we need to ask for more information,
40086 we may be unable to reach you. For this reason, it is better to send
40087 bug reports to the mailing list.
40089 @ifclear BUGURL_DEFAULT
40090 In any event, we also recommend that you submit bug reports for
40091 @value{GDBN} to @value{BUGURL}.
40095 The fundamental principle of reporting bugs usefully is this:
40096 @strong{report all the facts}. If you are not sure whether to state a
40097 fact or leave it out, state it!
40099 Often people omit facts because they think they know what causes the
40100 problem and assume that some details do not matter. Thus, you might
40101 assume that the name of the variable you use in an example does not matter.
40102 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
40103 stray memory reference which happens to fetch from the location where that
40104 name is stored in memory; perhaps, if the name were different, the contents
40105 of that location would fool the debugger into doing the right thing despite
40106 the bug. Play it safe and give a specific, complete example. That is the
40107 easiest thing for you to do, and the most helpful.
40109 Keep in mind that the purpose of a bug report is to enable us to fix the
40110 bug. It may be that the bug has been reported previously, but neither
40111 you nor we can know that unless your bug report is complete and
40114 Sometimes people give a few sketchy facts and ask, ``Does this ring a
40115 bell?'' Those bug reports are useless, and we urge everyone to
40116 @emph{refuse to respond to them} except to chide the sender to report
40119 To enable us to fix the bug, you should include all these things:
40123 The version of @value{GDBN}. @value{GDBN} announces it if you start
40124 with no arguments; you can also print it at any time using @code{show
40127 Without this, we will not know whether there is any point in looking for
40128 the bug in the current version of @value{GDBN}.
40131 The type of machine you are using, and the operating system name and
40135 The details of the @value{GDBN} build-time configuration.
40136 @value{GDBN} shows these details if you invoke it with the
40137 @option{--configuration} command-line option, or if you type
40138 @code{show configuration} at @value{GDBN}'s prompt.
40141 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
40142 ``@value{GCC}--2.8.1''.
40145 What compiler (and its version) was used to compile the program you are
40146 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
40147 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
40148 to get this information; for other compilers, see the documentation for
40152 The command arguments you gave the compiler to compile your example and
40153 observe the bug. For example, did you use @samp{-O}? To guarantee
40154 you will not omit something important, list them all. A copy of the
40155 Makefile (or the output from make) is sufficient.
40157 If we were to try to guess the arguments, we would probably guess wrong
40158 and then we might not encounter the bug.
40161 A complete input script, and all necessary source files, that will
40165 A description of what behavior you observe that you believe is
40166 incorrect. For example, ``It gets a fatal signal.''
40168 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
40169 will certainly notice it. But if the bug is incorrect output, we might
40170 not notice unless it is glaringly wrong. You might as well not give us
40171 a chance to make a mistake.
40173 Even if the problem you experience is a fatal signal, you should still
40174 say so explicitly. Suppose something strange is going on, such as, your
40175 copy of @value{GDBN} is out of synch, or you have encountered a bug in
40176 the C library on your system. (This has happened!) Your copy might
40177 crash and ours would not. If you told us to expect a crash, then when
40178 ours fails to crash, we would know that the bug was not happening for
40179 us. If you had not told us to expect a crash, then we would not be able
40180 to draw any conclusion from our observations.
40183 @cindex recording a session script
40184 To collect all this information, you can use a session recording program
40185 such as @command{script}, which is available on many Unix systems.
40186 Just run your @value{GDBN} session inside @command{script} and then
40187 include the @file{typescript} file with your bug report.
40189 Another way to record a @value{GDBN} session is to run @value{GDBN}
40190 inside Emacs and then save the entire buffer to a file.
40193 If you wish to suggest changes to the @value{GDBN} source, send us context
40194 diffs. If you even discuss something in the @value{GDBN} source, refer to
40195 it by context, not by line number.
40197 The line numbers in our development sources will not match those in your
40198 sources. Your line numbers would convey no useful information to us.
40202 Here are some things that are not necessary:
40206 A description of the envelope of the bug.
40208 Often people who encounter a bug spend a lot of time investigating
40209 which changes to the input file will make the bug go away and which
40210 changes will not affect it.
40212 This is often time consuming and not very useful, because the way we
40213 will find the bug is by running a single example under the debugger
40214 with breakpoints, not by pure deduction from a series of examples.
40215 We recommend that you save your time for something else.
40217 Of course, if you can find a simpler example to report @emph{instead}
40218 of the original one, that is a convenience for us. Errors in the
40219 output will be easier to spot, running under the debugger will take
40220 less time, and so on.
40222 However, simplification is not vital; if you do not want to do this,
40223 report the bug anyway and send us the entire test case you used.
40226 A patch for the bug.
40228 A patch for the bug does help us if it is a good one. But do not omit
40229 the necessary information, such as the test case, on the assumption that
40230 a patch is all we need. We might see problems with your patch and decide
40231 to fix the problem another way, or we might not understand it at all.
40233 Sometimes with a program as complicated as @value{GDBN} it is very hard to
40234 construct an example that will make the program follow a certain path
40235 through the code. If you do not send us the example, we will not be able
40236 to construct one, so we will not be able to verify that the bug is fixed.
40238 And if we cannot understand what bug you are trying to fix, or why your
40239 patch should be an improvement, we will not install it. A test case will
40240 help us to understand.
40243 A guess about what the bug is or what it depends on.
40245 Such guesses are usually wrong. Even we cannot guess right about such
40246 things without first using the debugger to find the facts.
40249 @c The readline documentation is distributed with the readline code
40250 @c and consists of the two following files:
40253 @c Use -I with makeinfo to point to the appropriate directory,
40254 @c environment var TEXINPUTS with TeX.
40255 @ifclear SYSTEM_READLINE
40256 @include rluser.texi
40257 @include hsuser.texi
40261 @appendix In Memoriam
40263 The @value{GDBN} project mourns the loss of the following long-time
40268 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
40269 to Free Software in general. Outside of @value{GDBN}, he was known in
40270 the Amiga world for his series of Fish Disks, and the GeekGadget project.
40272 @item Michael Snyder
40273 Michael was one of the Global Maintainers of the @value{GDBN} project,
40274 with contributions recorded as early as 1996, until 2011. In addition
40275 to his day to day participation, he was a large driving force behind
40276 adding Reverse Debugging to @value{GDBN}.
40279 Beyond their technical contributions to the project, they were also
40280 enjoyable members of the Free Software Community. We will miss them.
40282 @node Formatting Documentation
40283 @appendix Formatting Documentation
40285 @cindex @value{GDBN} reference card
40286 @cindex reference card
40287 The @value{GDBN} 4 release includes an already-formatted reference card, ready
40288 for printing with PostScript or Ghostscript, in the @file{gdb}
40289 subdirectory of the main source directory@footnote{In
40290 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
40291 release.}. If you can use PostScript or Ghostscript with your printer,
40292 you can print the reference card immediately with @file{refcard.ps}.
40294 The release also includes the source for the reference card. You
40295 can format it, using @TeX{}, by typing:
40301 The @value{GDBN} reference card is designed to print in @dfn{landscape}
40302 mode on US ``letter'' size paper;
40303 that is, on a sheet 11 inches wide by 8.5 inches
40304 high. You will need to specify this form of printing as an option to
40305 your @sc{dvi} output program.
40307 @cindex documentation
40309 All the documentation for @value{GDBN} comes as part of the machine-readable
40310 distribution. The documentation is written in Texinfo format, which is
40311 a documentation system that uses a single source file to produce both
40312 on-line information and a printed manual. You can use one of the Info
40313 formatting commands to create the on-line version of the documentation
40314 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
40316 @value{GDBN} includes an already formatted copy of the on-line Info
40317 version of this manual in the @file{gdb} subdirectory. The main Info
40318 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
40319 subordinate files matching @samp{gdb.info*} in the same directory. If
40320 necessary, you can print out these files, or read them with any editor;
40321 but they are easier to read using the @code{info} subsystem in @sc{gnu}
40322 Emacs or the standalone @code{info} program, available as part of the
40323 @sc{gnu} Texinfo distribution.
40325 If you want to format these Info files yourself, you need one of the
40326 Info formatting programs, such as @code{texinfo-format-buffer} or
40329 If you have @code{makeinfo} installed, and are in the top level
40330 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
40331 version @value{GDBVN}), you can make the Info file by typing:
40338 If you want to typeset and print copies of this manual, you need @TeX{},
40339 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
40340 Texinfo definitions file.
40342 @TeX{} is a typesetting program; it does not print files directly, but
40343 produces output files called @sc{dvi} files. To print a typeset
40344 document, you need a program to print @sc{dvi} files. If your system
40345 has @TeX{} installed, chances are it has such a program. The precise
40346 command to use depends on your system; @kbd{lpr -d} is common; another
40347 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
40348 require a file name without any extension or a @samp{.dvi} extension.
40350 @TeX{} also requires a macro definitions file called
40351 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
40352 written in Texinfo format. On its own, @TeX{} cannot either read or
40353 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
40354 and is located in the @file{gdb-@var{version-number}/texinfo}
40357 If you have @TeX{} and a @sc{dvi} printer program installed, you can
40358 typeset and print this manual. First switch to the @file{gdb}
40359 subdirectory of the main source directory (for example, to
40360 @file{gdb-@value{GDBVN}/gdb}) and type:
40366 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
40368 @node Installing GDB
40369 @appendix Installing @value{GDBN}
40370 @cindex installation
40373 * Requirements:: Requirements for building @value{GDBN}
40374 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
40375 * Separate Objdir:: Compiling @value{GDBN} in another directory
40376 * Config Names:: Specifying names for hosts and targets
40377 * Configure Options:: Summary of options for configure
40378 * System-wide configuration:: Having a system-wide init file
40382 @section Requirements for Building @value{GDBN}
40383 @cindex building @value{GDBN}, requirements for
40385 Building @value{GDBN} requires various tools and packages to be available.
40386 Other packages will be used only if they are found.
40388 @heading Tools/Packages Necessary for Building @value{GDBN}
40390 @item C@t{++}11 compiler
40391 @value{GDBN} is written in C@t{++}11. It should be buildable with any
40392 recent C@t{++}11 compiler, e.g.@: GCC.
40395 @value{GDBN}'s build system relies on features only found in the GNU
40396 make program. Other variants of @code{make} will not work.
40399 The following libraries are mandatory for building @value{GDBN}. The
40400 @file{configure} script searches for each of these libraries in
40401 several standard locations; if some library is installed in an unusual
40402 place, you can use either the @option{--with-@var{lib}}
40403 @file{configure} option to specify its installation directory, or
40404 the two separate options @option{---with-@var{library}-include} (to
40405 specify the location of its header files) and
40406 @option{--with-@var{library}-lib} (to specify the location of its
40407 libraries). For example, for the GMP library, the 3 options are
40408 @option{--with-gmp}, @option{--with-gmp-include}, and
40409 @option{--with-gmp-lib}. @xref{Configure Options}. We mention below
40410 the home site of each library, so that you could download and install
40411 them if your system doesn't already include them.
40414 @item GMP (The GNU Multiple Precision arithmetic library)
40415 @value{GDBN} uses GMP to perform some of its extended-precision
40416 arithmetics. The latest version of GMP is available from
40417 @url{https://gmplib.org/}.
40420 @item MPFR (The GNU Multiple-precision floating-point library)
40421 @value{GDBN} uses MPFR to emulate the target floating-point
40422 arithmetics during expression evaluation, if the target uses different
40423 floating-point formats than the host. The latest version of MPFR is
40424 available from @url{http://www.mpfr.org}.
40429 @heading Tools/Packages Optional for Building @value{GDBN}
40430 The tools/packages and libraries listed below are optional;
40431 @value{GDBN} can be build without them, at the expense of some run-time
40432 functionality that will be missing. As above, we list the home sites
40433 for each package/library, and the command-line options supported by
40434 the @file{configure} script to specify their installation directories
40435 if they are non-standard. In addition, for each package you can use
40436 the option @option{--with-@var{package}} to force @value{GDBN} to be
40437 compiled with the named @var{package}, and
40438 @option{--without-@var{package}} to disable building with it even if
40439 it is available. @xref{Configure Options}, for detailed description
40440 of the options to @file{configure}.
40444 @value{GDBN} can be scripted using Python language. @xref{Python}.
40445 The latest version is available from
40446 @url{https://www.python.org/downloads/}. Use the
40447 @option{--with-python=@var{dir}} to specify the non-standard directory
40448 where Python is installed.
40451 @value{GDBN} can also be scripted using GNU Guile. @xref{Guile}. The
40452 latest version can be found on
40453 @url{https://www.gnu.org/software/guile/download/}. If you have more
40454 than one version of Guile installed, use the
40455 @option{--with-guile=@var{guile-version}} to specify the Guile version
40456 to include in the build.
40460 If available, @value{GDBN} uses the Expat library for parsing XML
40461 files. @value{GDBN} uses XML files for the following functionalities:
40465 Remote protocol memory maps (@pxref{Memory Map Format})
40467 Target descriptions (@pxref{Target Descriptions})
40469 Remote shared library lists (@xref{Library List Format},
40470 or alternatively @pxref{Library List Format for SVR4 Targets})
40472 MS-Windows shared libraries (@pxref{Shared Libraries})
40474 Traceframe info (@pxref{Traceframe Info Format})
40476 Branch trace (@pxref{Branch Trace Format},
40477 @pxref{Branch Trace Configuration Format})
40480 The latest version of Expat is available from
40481 @url{http://expat.sourceforge.net}. Use the
40482 @option{--with-libexpat-prefix} to specify non-standard installation
40486 @value{GDBN}'s features related to character sets (@pxref{Character
40487 Sets}) require a functioning @code{iconv} implementation. If you are
40488 on a GNU system, then this is provided by the GNU C Library. Some
40489 other systems also provide a working @code{iconv}. Use the option
40490 @option{--with-iconv-bin} to specify where to find the @command{iconv}
40493 On systems without @code{iconv}, you can install the GNU Libiconv
40494 library; its latest version can be found on
40495 @url{https://ftp.gnu.org/pub/gnu/libiconv/} if your system doesn't
40496 provide it. Use the @option{--with-libiconv-prefix} option to
40497 @file{configure} to specify non-standard installation place for it.
40499 Alternatively, @value{GDBN}'s top-level @file{configure} and
40500 @file{Makefile} will arrange to build Libiconv if a directory named
40501 @file{libiconv} appears in the top-most source directory. If Libiconv
40502 is built this way, and if the operating system does not provide a
40503 suitable @code{iconv} implementation, then the just-built library will
40504 automatically be used by @value{GDBN}. One easy way to set this up is
40505 to download GNU Libiconv, unpack it inside the top-level directory of
40506 the @value{GDBN} source tree, and then rename the directory holding
40507 the Libiconv source code to @samp{libiconv}.
40509 @cindex compressed debug sections
40511 @value{GDBN} can support debugging sections that are compressed with
40512 the LZMA library. @xref{MiniDebugInfo}. If this library is not
40513 included with your operating system, you can find it in the xz package
40514 at @url{http://tukaani.org/xz/}. Use the
40515 @option{--with-liblzma-prefix} option to specify its non-standard
40519 @value{GDBN} will use the @samp{zlib} library, if available, to read
40520 compressed debug sections. Some linkers, such as GNU @command{gold},
40521 are capable of producing binaries with compressed debug sections. If
40522 @value{GDBN} is compiled with @samp{zlib}, it will be able to read the
40523 debug information in such binaries.
40525 The @samp{zlib} library is likely included with your operating system
40526 distribution; if it is not, you can get the latest version from
40527 @url{http://zlib.net}.
40529 @c FIXME: what about other optional libraries: debuginfod, zstd,
40530 @c libipt, babeltrace, xxhash, source-highlight?
40533 @node Running Configure
40534 @section Invoking the @value{GDBN} @file{configure} Script
40535 @cindex configuring @value{GDBN}
40536 @value{GDBN} comes with a @file{configure} script that automates the process
40537 of preparing @value{GDBN} for installation; you can then use @code{make} to
40538 build the @code{gdb} program.
40540 @c irrelevant in info file; it's as current as the code it lives with.
40541 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
40542 look at the @file{README} file in the sources; we may have improved the
40543 installation procedures since publishing this manual.}
40546 The @value{GDBN} distribution includes all the source code you need for
40547 @value{GDBN} in a single directory, whose name is usually composed by
40548 appending the version number to @samp{gdb}.
40550 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
40551 @file{gdb-@value{GDBVN}} directory. That directory contains:
40554 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
40555 script for configuring @value{GDBN} and all its supporting libraries
40557 @item gdb-@value{GDBVN}/gdb
40558 the source specific to @value{GDBN} itself
40560 @item gdb-@value{GDBVN}/bfd
40561 source for the Binary File Descriptor library
40563 @item gdb-@value{GDBVN}/include
40564 @sc{gnu} include files
40566 @item gdb-@value{GDBVN}/libiberty
40567 source for the @samp{-liberty} free software library
40569 @item gdb-@value{GDBVN}/opcodes
40570 source for the library of opcode tables and disassemblers
40572 @item gdb-@value{GDBVN}/readline
40573 source for the @sc{gnu} command-line interface
40576 There may be other subdirectories as well.
40578 The simplest way to configure and build @value{GDBN} is to run @file{configure}
40579 from the @file{gdb-@var{version-number}} source directory, which in
40580 this example is the @file{gdb-@value{GDBVN}} directory.
40582 First switch to the @file{gdb-@var{version-number}} source directory
40583 if you are not already in it; then run @file{configure}. Pass the
40584 identifier for the platform on which @value{GDBN} will run as an
40590 cd gdb-@value{GDBVN}
40595 Running @samp{configure} and then running @code{make} builds the
40596 included supporting libraries, then @code{gdb} itself. The configured
40597 source files, and the binaries, are left in the corresponding source
40601 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
40602 system does not recognize this automatically when you run a different
40603 shell, you may need to run @code{sh} on it explicitly:
40609 You should run the @file{configure} script from the top directory in the
40610 source tree, the @file{gdb-@var{version-number}} directory. If you run
40611 @file{configure} from one of the subdirectories, you will configure only
40612 that subdirectory. That is usually not what you want. In particular,
40613 if you run the first @file{configure} from the @file{gdb} subdirectory
40614 of the @file{gdb-@var{version-number}} directory, you will omit the
40615 configuration of @file{bfd}, @file{readline}, and other sibling
40616 directories of the @file{gdb} subdirectory. This leads to build errors
40617 about missing include files such as @file{bfd/bfd.h}.
40619 You can install @code{@value{GDBN}} anywhere. The best way to do this
40620 is to pass the @code{--prefix} option to @code{configure}, and then
40621 install it with @code{make install}.
40623 @node Separate Objdir
40624 @section Compiling @value{GDBN} in Another Directory
40626 If you want to run @value{GDBN} versions for several host or target machines,
40627 you need a different @code{gdb} compiled for each combination of
40628 host and target. @file{configure} is designed to make this easy by
40629 allowing you to generate each configuration in a separate subdirectory,
40630 rather than in the source directory. If your @code{make} program
40631 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
40632 @code{make} in each of these directories builds the @code{gdb}
40633 program specified there.
40635 To build @code{gdb} in a separate directory, run @file{configure}
40636 with the @samp{--srcdir} option to specify where to find the source.
40637 (You also need to specify a path to find @file{configure}
40638 itself from your working directory. If the path to @file{configure}
40639 would be the same as the argument to @samp{--srcdir}, you can leave out
40640 the @samp{--srcdir} option; it is assumed.)
40642 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
40643 separate directory for a Sun 4 like this:
40647 cd gdb-@value{GDBVN}
40650 ../gdb-@value{GDBVN}/configure
40655 When @file{configure} builds a configuration using a remote source
40656 directory, it creates a tree for the binaries with the same structure
40657 (and using the same names) as the tree under the source directory. In
40658 the example, you'd find the Sun 4 library @file{libiberty.a} in the
40659 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
40660 @file{gdb-sun4/gdb}.
40662 Make sure that your path to the @file{configure} script has just one
40663 instance of @file{gdb} in it. If your path to @file{configure} looks
40664 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
40665 one subdirectory of @value{GDBN}, not the whole package. This leads to
40666 build errors about missing include files such as @file{bfd/bfd.h}.
40668 One popular reason to build several @value{GDBN} configurations in separate
40669 directories is to configure @value{GDBN} for cross-compiling (where
40670 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
40671 programs that run on another machine---the @dfn{target}).
40672 You specify a cross-debugging target by
40673 giving the @samp{--target=@var{target}} option to @file{configure}.
40675 When you run @code{make} to build a program or library, you must run
40676 it in a configured directory---whatever directory you were in when you
40677 called @file{configure} (or one of its subdirectories).
40679 The @code{Makefile} that @file{configure} generates in each source
40680 directory also runs recursively. If you type @code{make} in a source
40681 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
40682 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
40683 will build all the required libraries, and then build GDB.
40685 When you have multiple hosts or targets configured in separate
40686 directories, you can run @code{make} on them in parallel (for example,
40687 if they are NFS-mounted on each of the hosts); they will not interfere
40691 @section Specifying Names for Hosts and Targets
40693 The specifications used for hosts and targets in the @file{configure}
40694 script are based on a three-part naming scheme, but some short predefined
40695 aliases are also supported. The full naming scheme encodes three pieces
40696 of information in the following pattern:
40699 @var{architecture}-@var{vendor}-@var{os}
40702 For example, you can use the alias @code{sun4} as a @var{host} argument,
40703 or as the value for @var{target} in a @code{--target=@var{target}}
40704 option. The equivalent full name is @samp{sparc-sun-sunos4}.
40706 The @file{configure} script accompanying @value{GDBN} does not provide
40707 any query facility to list all supported host and target names or
40708 aliases. @file{configure} calls the Bourne shell script
40709 @code{config.sub} to map abbreviations to full names; you can read the
40710 script, if you wish, or you can use it to test your guesses on
40711 abbreviations---for example:
40714 % sh config.sub i386-linux
40716 % sh config.sub alpha-linux
40717 alpha-unknown-linux-gnu
40718 % sh config.sub hp9k700
40720 % sh config.sub sun4
40721 sparc-sun-sunos4.1.1
40722 % sh config.sub sun3
40723 m68k-sun-sunos4.1.1
40724 % sh config.sub i986v
40725 Invalid configuration `i986v': machine `i986v' not recognized
40729 @code{config.sub} is also distributed in the @value{GDBN} source
40730 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
40732 @node Configure Options
40733 @section @file{configure} Options
40735 @c FIXME: This largely repeats what was already described in
40736 @c ``Requirements'', and OTOH doesn't describe the more fgine-granular
40737 @c options like --with-libexpat-prefix and --with-python-libdir.
40739 Here is a summary of the @file{configure} options and arguments that
40740 are most often useful for building @value{GDBN}. @file{configure}
40741 also has several other options not listed here. @xref{Running
40742 configure Scripts,,,autoconf}, for a full
40743 explanation of @file{configure}.
40746 configure @r{[}--help@r{]}
40747 @r{[}--prefix=@var{dir}@r{]}
40748 @r{[}--exec-prefix=@var{dir}@r{]}
40749 @r{[}--srcdir=@var{dirname}@r{]}
40750 @r{[}--target=@var{target}@r{]}
40754 You may introduce options with a single @samp{-} rather than
40755 @samp{--} if you prefer; but you may abbreviate option names if you use
40760 Display a quick summary of how to invoke @file{configure}.
40762 @item --prefix=@var{dir}
40763 Configure the source to install programs and files under directory
40766 @item --exec-prefix=@var{dir}
40767 Configure the source to install programs under directory
40770 @c avoid splitting the warning from the explanation:
40772 @item --srcdir=@var{dirname}
40773 Use this option to make configurations in directories separate from the
40774 @value{GDBN} source directories. Among other things, you can use this to
40775 build (or maintain) several configurations simultaneously, in separate
40776 directories. @file{configure} writes configuration-specific files in
40777 the current directory, but arranges for them to use the source in the
40778 directory @var{dirname}. @file{configure} creates directories under
40779 the working directory in parallel to the source directories below
40782 @item --target=@var{target}
40783 Configure @value{GDBN} for cross-debugging programs running on the specified
40784 @var{target}. Without this option, @value{GDBN} is configured to debug
40785 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
40787 There is no convenient way to generate a list of all available
40788 targets. Also see the @code{--enable-targets} option, below.
40791 There are many other options that are specific to @value{GDBN}. This
40792 lists just the most common ones; there are some very specialized
40793 options not described here.
40796 @item --enable-targets=@r{[}@var{target}@r{]}@dots{}
40797 @itemx --enable-targets=all
40798 Configure @value{GDBN} for cross-debugging programs running on the
40799 specified list of targets. The special value @samp{all} configures
40800 @value{GDBN} for debugging programs running on any target it supports.
40802 @item --with-gdb-datadir=@var{path}
40803 Set the @value{GDBN}-specific data directory. @value{GDBN} will look
40804 here for certain supporting files or scripts. This defaults to the
40805 @file{gdb} subdirectory of @samp{datadir} (which can be set using
40808 @item --with-relocated-sources=@var{dir}
40809 Sets up the default source path substitution rule so that directory
40810 names recorded in debug information will be automatically adjusted for
40811 any directory under @var{dir}. @var{dir} should be a subdirectory of
40812 @value{GDBN}'s configured prefix, the one mentioned in the
40813 @code{--prefix} or @code{--exec-prefix} options to configure. This
40814 option is useful if GDB is supposed to be moved to a different place
40817 @item --enable-64-bit-bfd
40818 Enable 64-bit support in BFD on 32-bit hosts.
40820 @item --disable-gdbmi
40821 Build @value{GDBN} without the GDB/MI machine interface
40825 Build @value{GDBN} with the text-mode full-screen user interface
40826 (TUI). Requires a curses library (ncurses and cursesX are also
40829 @item --with-curses
40830 Use the curses library instead of the termcap library, for text-mode
40831 terminal operations.
40833 @item --with-debuginfod
40834 Build @value{GDBN} with @file{libdebuginfod}, the @code{debuginfod} client
40835 library. Used to automatically fetch ELF, DWARF and source files from
40836 @code{debuginfod} servers using build IDs associated with any missing
40837 files. Enabled by default if @file{libdebuginfod} is installed and found
40838 at configure time. For more information regarding @code{debuginfod} see
40841 @item --with-libunwind-ia64
40842 Use the libunwind library for unwinding function call stack on ia64
40843 target platforms. See @url{http://www.nongnu.org/libunwind/index.html} for
40846 @item --with-system-readline
40847 Use the readline library installed on the host, rather than the
40848 library supplied as part of @value{GDBN}. Readline 7 or newer is
40849 required; this is enforced by the build system.
40851 @item --with-system-zlib
40852 Use the zlib library installed on the host, rather than the library
40853 supplied as part of @value{GDBN}.
40856 Build @value{GDBN} with Expat, a library for XML parsing. (Done by
40857 default if libexpat is installed and found at configure time.) This
40858 library is used to read XML files supplied with @value{GDBN}. If it
40859 is unavailable, some features, such as remote protocol memory maps,
40860 target descriptions, and shared library lists, that are based on XML
40861 files, will not be available in @value{GDBN}. If your host does not
40862 have libexpat installed, you can get the latest version from
40863 @url{http://expat.sourceforge.net}.
40865 @item --with-libiconv-prefix@r{[}=@var{dir}@r{]}
40866 Build @value{GDBN} with GNU libiconv, a character set encoding
40867 conversion library. This is not done by default, as on GNU systems
40868 the @code{iconv} that is built in to the C library is sufficient. If
40869 your host does not have a working @code{iconv}, you can get the latest
40870 version of GNU iconv from @url{https://www.gnu.org/software/libiconv/}.
40872 @value{GDBN}'s build system also supports building GNU libiconv as
40873 part of the overall build. @xref{Requirements}.
40876 Build @value{GDBN} with LZMA, a compression library. (Done by default
40877 if liblzma is installed and found at configure time.) LZMA is used by
40878 @value{GDBN}'s "mini debuginfo" feature, which is only useful on
40879 platforms using the ELF object file format. If your host does not
40880 have liblzma installed, you can get the latest version from
40881 @url{https://tukaani.org/xz/}.
40883 @item --with-python@r{[}=@var{python}@r{]}
40884 Build @value{GDBN} with Python scripting support. (Done by default if
40885 libpython is present and found at configure time.) Python makes
40886 @value{GDBN} scripting much more powerful than the restricted CLI
40887 scripting language. If your host does not have Python installed, you
40888 can find it on @url{http://www.python.org/download/}. The oldest version
40889 of Python supported by GDB is 3.0.1. The optional argument @var{python}
40890 is used to find the Python headers and libraries. It can be either
40891 the name of a Python executable, or the name of the directory in which
40892 Python is installed.
40894 @item --with-guile[=@var{guile}]
40895 Build @value{GDBN} with GNU Guile scripting support. (Done by default
40896 if libguile is present and found at configure time.) If your host
40897 does not have Guile installed, you can find it at
40898 @url{https://www.gnu.org/software/guile/}. The optional argument @var{guile}
40899 can be a version number, which will cause @code{configure} to try to
40900 use that version of Guile; or the file name of a @code{pkg-config}
40901 executable, which will be queried to find the information needed to
40902 compile and link against Guile.
40904 @item --without-included-regex
40905 Don't use the regex library included with @value{GDBN} (as part of the
40906 libiberty library). This is the default on hosts with version 2 of
40909 @item --with-sysroot=@var{dir}
40910 Use @var{dir} as the default system root directory for libraries whose
40911 file names begin with @file{/lib}' or @file{/usr/lib'}. (The value of
40912 @var{dir} can be modified at run time by using the @command{set
40913 sysroot} command.) If @var{dir} is under the @value{GDBN} configured
40914 prefix (set with @code{--prefix} or @code{--exec-prefix options}, the
40915 default system root will be automatically adjusted if and when
40916 @value{GDBN} is moved to a different location.
40918 @item --with-system-gdbinit=@var{file}
40919 Configure @value{GDBN} to automatically load a system-wide init file.
40920 @var{file} should be an absolute file name. If @var{file} is in a
40921 directory under the configured prefix, and @value{GDBN} is moved to
40922 another location after being built, the location of the system-wide
40923 init file will be adjusted accordingly.
40925 @item --with-system-gdbinit-dir=@var{directory}
40926 Configure @value{GDBN} to automatically load init files from a
40927 system-wide directory. @var{directory} should be an absolute directory
40928 name. If @var{directory} is in a directory under the configured
40929 prefix, and @value{GDBN} is moved to another location after being
40930 built, the location of the system-wide init directory will be
40931 adjusted accordingly.
40933 @item --enable-build-warnings
40934 When building the @value{GDBN} sources, ask the compiler to warn about
40935 any code which looks even vaguely suspicious. It passes many
40936 different warning flags, depending on the exact version of the
40937 compiler you are using.
40939 @item --enable-werror
40940 Treat compiler warnings as errors. It adds the @code{-Werror} flag
40941 to the compiler, which will fail the compilation if the compiler
40942 outputs any warning messages.
40944 @item --enable-ubsan
40945 Enable the GCC undefined behavior sanitizer. This is disabled by
40946 default, but passing @code{--enable-ubsan=yes} or
40947 @code{--enable-ubsan=auto} to @code{configure} will enable it. The
40948 undefined behavior sanitizer checks for C@t{++} undefined behavior.
40949 It has a performance cost, so if you are looking at @value{GDBN}'s
40950 performance, you should disable it. The undefined behavior sanitizer
40951 was first introduced in GCC 4.9.
40954 @node System-wide configuration
40955 @section System-wide configuration and settings
40956 @cindex system-wide init file
40958 @value{GDBN} can be configured to have a system-wide init file and a
40959 system-wide init file directory; this file and files in that directory
40960 (if they have a recognized file extension) will be read and executed at
40961 startup (@pxref{Startup, , What @value{GDBN} does during startup}).
40963 Here are the corresponding configure options:
40966 @item --with-system-gdbinit=@var{file}
40967 Specify that the default location of the system-wide init file is
40969 @item --with-system-gdbinit-dir=@var{directory}
40970 Specify that the default location of the system-wide init file directory
40971 is @var{directory}.
40974 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
40975 they may be subject to relocation. Two possible cases:
40979 If the default location of this init file/directory contains @file{$prefix},
40980 it will be subject to relocation. Suppose that the configure options
40981 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
40982 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
40983 init file is looked for as @file{$install/etc/gdbinit} instead of
40984 @file{$prefix/etc/gdbinit}.
40987 By contrast, if the default location does not contain the prefix,
40988 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
40989 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
40990 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
40991 wherever @value{GDBN} is installed.
40994 If the configured location of the system-wide init file (as given by the
40995 @option{--with-system-gdbinit} option at configure time) is in the
40996 data-directory (as specified by @option{--with-gdb-datadir} at configure
40997 time) or in one of its subdirectories, then @value{GDBN} will look for the
40998 system-wide init file in the directory specified by the
40999 @option{--data-directory} command-line option.
41000 Note that the system-wide init file is only read once, during @value{GDBN}
41001 initialization. If the data-directory is changed after @value{GDBN} has
41002 started with the @code{set data-directory} command, the file will not be
41005 This applies similarly to the system-wide directory specified in
41006 @option{--with-system-gdbinit-dir}.
41008 Any supported scripting language can be used for these init files, as long
41009 as the file extension matches the scripting language. To be interpreted
41010 as regular @value{GDBN} commands, the files needs to have a @file{.gdb}
41014 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
41017 @node System-wide Configuration Scripts
41018 @subsection Installed System-wide Configuration Scripts
41019 @cindex system-wide configuration scripts
41021 The @file{system-gdbinit} directory, located inside the data-directory
41022 (as specified by @option{--with-gdb-datadir} at configure time) contains
41023 a number of scripts which can be used as system-wide init files. To
41024 automatically source those scripts at startup, @value{GDBN} should be
41025 configured with @option{--with-system-gdbinit}. Otherwise, any user
41026 should be able to source them by hand as needed.
41028 The following scripts are currently available:
41031 @item @file{elinos.py}
41033 @cindex ELinOS system-wide configuration script
41034 This script is useful when debugging a program on an ELinOS target.
41035 It takes advantage of the environment variables defined in a standard
41036 ELinOS environment in order to determine the location of the system
41037 shared libraries, and then sets the @samp{solib-absolute-prefix}
41038 and @samp{solib-search-path} variables appropriately.
41040 @item @file{wrs-linux.py}
41041 @pindex wrs-linux.py
41042 @cindex Wind River Linux system-wide configuration script
41043 This script is useful when debugging a program on a target running
41044 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
41045 the host-side sysroot used by the target system.
41049 @node Maintenance Commands
41050 @appendix Maintenance Commands
41051 @cindex maintenance commands
41052 @cindex internal commands
41054 In addition to commands intended for @value{GDBN} users, @value{GDBN}
41055 includes a number of commands intended for @value{GDBN} developers,
41056 that are not documented elsewhere in this manual. These commands are
41057 provided here for reference. (For commands that turn on debugging
41058 messages, see @ref{Debugging Output}.)
41061 @kindex maint agent
41062 @kindex maint agent-eval
41063 @item maint agent @r{[}-at @var{linespec}@r{,}@r{]} @var{expression}
41064 @itemx maint agent-eval @r{[}-at @var{linespec}@r{,}@r{]} @var{expression}
41065 Translate the given @var{expression} into remote agent bytecodes.
41066 This command is useful for debugging the Agent Expression mechanism
41067 (@pxref{Agent Expressions}). The @samp{agent} version produces an
41068 expression useful for data collection, such as by tracepoints, while
41069 @samp{maint agent-eval} produces an expression that evaluates directly
41070 to a result. For instance, a collection expression for @code{globa +
41071 globb} will include bytecodes to record four bytes of memory at each
41072 of the addresses of @code{globa} and @code{globb}, while discarding
41073 the result of the addition, while an evaluation expression will do the
41074 addition and return the sum.
41075 If @code{-at} is given, generate remote agent bytecode for all the
41076 addresses to which @var{linespec} resolves (@pxref{Linespec
41078 If not, generate remote agent bytecode for current frame PC address.
41080 @kindex maint agent-printf
41081 @item maint agent-printf @var{format},@var{expr},...
41082 Translate the given format string and list of argument expressions
41083 into remote agent bytecodes and display them as a disassembled list.
41084 This command is useful for debugging the agent version of dynamic
41085 printf (@pxref{Dynamic Printf}).
41087 @kindex maint info breakpoints
41088 @item @anchor{maint info breakpoints}maint info breakpoints
41089 Using the same format as @samp{info breakpoints}, display both the
41090 breakpoints you've set explicitly, and those @value{GDBN} is using for
41091 internal purposes. Internal breakpoints are shown with negative
41092 breakpoint numbers. The type column identifies what kind of breakpoint
41097 Normal, explicitly set breakpoint.
41100 Normal, explicitly set watchpoint.
41103 Internal breakpoint, used to handle correctly stepping through
41104 @code{longjmp} calls.
41106 @item longjmp resume
41107 Internal breakpoint at the target of a @code{longjmp}.
41110 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
41113 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
41116 Shared library events.
41120 @kindex maint info btrace
41121 @item maint info btrace
41122 Pint information about raw branch tracing data.
41124 @kindex maint btrace packet-history
41125 @item maint btrace packet-history
41126 Print the raw branch trace packets that are used to compute the
41127 execution history for the @samp{record btrace} command. Both the
41128 information and the format in which it is printed depend on the btrace
41133 For the BTS recording format, print a list of blocks of sequential
41134 code. For each block, the following information is printed:
41138 Newer blocks have higher numbers. The oldest block has number zero.
41139 @item Lowest @samp{PC}
41140 @item Highest @samp{PC}
41144 For the Intel Processor Trace recording format, print a list of
41145 Intel Processor Trace packets. For each packet, the following
41146 information is printed:
41149 @item Packet number
41150 Newer packets have higher numbers. The oldest packet has number zero.
41152 The packet's offset in the trace stream.
41153 @item Packet opcode and payload
41157 @kindex maint btrace clear-packet-history
41158 @item maint btrace clear-packet-history
41159 Discards the cached packet history printed by the @samp{maint btrace
41160 packet-history} command. The history will be computed again when
41163 @kindex maint btrace clear
41164 @item maint btrace clear
41165 Discard the branch trace data. The data will be fetched anew and the
41166 branch trace will be recomputed when needed.
41168 This implicitly truncates the branch trace to a single branch trace
41169 buffer. When updating branch trace incrementally, the branch trace
41170 available to @value{GDBN} may be bigger than a single branch trace
41173 @kindex maint set btrace pt skip-pad
41174 @item maint set btrace pt skip-pad
41175 @kindex maint show btrace pt skip-pad
41176 @item maint show btrace pt skip-pad
41177 Control whether @value{GDBN} will skip PAD packets when computing the
41180 @kindex maint info jit
41181 @item maint info jit
41182 Print information about JIT code objects loaded in the current inferior.
41184 @anchor{maint info python-disassemblers}
41185 @kindex maint info python-disassemblers
41186 @item maint info python-disassemblers
41187 This command is defined within the @code{gdb.disassembler} Python
41188 module (@pxref{Disassembly In Python}), and will only be present after
41189 that module has been imported. To force the module to be imported do
41193 (@value{GDBP}) python import gdb.disassembler
41196 This command lists all the architectures for which a disassembler is
41197 currently registered, and the name of the disassembler. If a
41198 disassembler is registered for all architectures, then this is listed
41199 last against the @samp{GLOBAL} architecture.
41201 If one of the disassemblers would be selected for the architecture of
41202 the current inferior, then this disassembler will be marked.
41204 The following example shows a situation in which two disassemblers are
41205 registered, initially the @samp{i386} disassembler matches the current
41206 architecture, then the architecture is changed, now the @samp{GLOBAL}
41207 disassembler matches.
41211 (@value{GDBP}) show architecture
41212 The target architecture is set to "auto" (currently "i386").
41213 (@value{GDBP}) maint info python-disassemblers
41214 Architecture Disassember Name
41215 i386 Disassembler_1 (Matches current architecture)
41216 GLOBAL Disassembler_2
41219 (@value{GDBP}) set architecture arm
41220 The target architecture is set to "arm".
41221 (@value{GDBP}) maint info python-disassemblers
41223 Architecture Disassember Name
41224 i386 Disassembler_1
41225 GLOBAL Disassembler_2 (Matches current architecture)
41229 @kindex set displaced-stepping
41230 @kindex show displaced-stepping
41231 @cindex displaced stepping support
41232 @cindex out-of-line single-stepping
41233 @item set displaced-stepping
41234 @itemx show displaced-stepping
41235 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
41236 if the target supports it. Displaced stepping is a way to single-step
41237 over breakpoints without removing them from the inferior, by executing
41238 an out-of-line copy of the instruction that was originally at the
41239 breakpoint location. It is also known as out-of-line single-stepping.
41242 @item set displaced-stepping on
41243 If the target architecture supports it, @value{GDBN} will use
41244 displaced stepping to step over breakpoints.
41246 @item set displaced-stepping off
41247 @value{GDBN} will not use displaced stepping to step over breakpoints,
41248 even if such is supported by the target architecture.
41250 @cindex non-stop mode, and @samp{set displaced-stepping}
41251 @item set displaced-stepping auto
41252 This is the default mode. @value{GDBN} will use displaced stepping
41253 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
41254 architecture supports displaced stepping.
41257 @kindex maint check-psymtabs
41258 @item maint check-psymtabs
41259 Check the consistency of currently expanded psymtabs versus symtabs.
41260 Use this to check, for example, whether a symbol is in one but not the other.
41262 @kindex maint check-symtabs
41263 @item maint check-symtabs
41264 Check the consistency of currently expanded symtabs.
41266 @kindex maint expand-symtabs
41267 @item maint expand-symtabs [@var{regexp}]
41268 Expand symbol tables.
41269 If @var{regexp} is specified, only expand symbol tables for file
41270 names matching @var{regexp}.
41272 @kindex maint set catch-demangler-crashes
41273 @kindex maint show catch-demangler-crashes
41274 @cindex demangler crashes
41275 @item maint set catch-demangler-crashes [on|off]
41276 @itemx maint show catch-demangler-crashes
41277 Control whether @value{GDBN} should attempt to catch crashes in the
41278 symbol name demangler. The default is to attempt to catch crashes.
41279 If enabled, the first time a crash is caught, a core file is created,
41280 the offending symbol is displayed and the user is presented with the
41281 option to terminate the current session.
41283 @kindex maint cplus first_component
41284 @item maint cplus first_component @var{name}
41285 Print the first C@t{++} class/namespace component of @var{name}.
41287 @kindex maint cplus namespace
41288 @item maint cplus namespace
41289 Print the list of possible C@t{++} namespaces.
41291 @kindex maint deprecate
41292 @kindex maint undeprecate
41293 @cindex deprecated commands
41294 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
41295 @itemx maint undeprecate @var{command}
41296 Deprecate or undeprecate the named @var{command}. Deprecated commands
41297 cause @value{GDBN} to issue a warning when you use them. The optional
41298 argument @var{replacement} says which newer command should be used in
41299 favor of the deprecated one; if it is given, @value{GDBN} will mention
41300 the replacement as part of the warning.
41302 @kindex maint dump-me
41303 @item maint dump-me
41304 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
41305 Cause a fatal signal in the debugger and force it to dump its core.
41306 This is supported only on systems which support aborting a program
41307 with the @code{SIGQUIT} signal.
41309 @kindex maint internal-error
41310 @kindex maint internal-warning
41311 @kindex maint demangler-warning
41312 @cindex demangler crashes
41313 @item maint internal-error @r{[}@var{message-text}@r{]}
41314 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
41315 @itemx maint demangler-warning @r{[}@var{message-text}@r{]}
41317 Cause @value{GDBN} to call the internal function @code{internal_error},
41318 @code{internal_warning} or @code{demangler_warning} and hence behave
41319 as though an internal problem has been detected. In addition to
41320 reporting the internal problem, these functions give the user the
41321 opportunity to either quit @value{GDBN} or (for @code{internal_error}
41322 and @code{internal_warning}) create a core file of the current
41323 @value{GDBN} session.
41325 These commands take an optional parameter @var{message-text} that is
41326 used as the text of the error or warning message.
41328 Here's an example of using @code{internal-error}:
41331 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
41332 @dots{}/maint.c:121: internal-error: testing, 1, 2
41333 A problem internal to GDB has been detected. Further
41334 debugging may prove unreliable.
41335 Quit this debugging session? (y or n) @kbd{n}
41336 Create a core file? (y or n) @kbd{n}
41340 @kindex maint set debuginfod download-sections
41341 @kindex maint show debuginfod download-sections
41342 @cindex debuginfod, maintenance commands
41343 @item maint set debuginfod download-sections
41344 @itemx maint set debuginfod download-sections @r{[}on|off@r{]}
41345 @itemx maint show debuginfod download-sections
41346 Controls whether @value{GDBN} will attempt to download individual
41347 ELF/DWARF sections from @code{debuginfod}. If disabled, only
41348 whole debug info files will be downloaded; this could result
41349 in @value{GDBN} downloading larger amounts of data.
41351 @cindex @value{GDBN} internal error
41352 @cindex internal errors, control of @value{GDBN} behavior
41353 @cindex demangler crashes
41355 @kindex maint set internal-error
41356 @kindex maint show internal-error
41357 @kindex maint set internal-warning
41358 @kindex maint show internal-warning
41359 @kindex maint set demangler-warning
41360 @kindex maint show demangler-warning
41361 @item maint set internal-error @var{action} [ask|yes|no]
41362 @itemx maint show internal-error @var{action}
41363 @itemx maint set internal-warning @var{action} [ask|yes|no]
41364 @itemx maint show internal-warning @var{action}
41365 @itemx maint set demangler-warning @var{action} [ask|yes|no]
41366 @itemx maint show demangler-warning @var{action}
41367 When @value{GDBN} reports an internal problem (error or warning) it
41368 gives the user the opportunity to both quit @value{GDBN} and create a
41369 core file of the current @value{GDBN} session. These commands let you
41370 override the default behaviour for each particular @var{action},
41371 described in the table below.
41375 You can specify that @value{GDBN} should always (yes) or never (no)
41376 quit. The default is to ask the user what to do.
41379 You can specify that @value{GDBN} should always (yes) or never (no)
41380 create a core file. The default is to ask the user what to do. Note
41381 that there is no @code{corefile} option for @code{demangler-warning}:
41382 demangler warnings always create a core file and this cannot be
41386 @kindex maint set internal-error
41387 @kindex maint show internal-error
41388 @kindex maint set internal-warning
41389 @kindex maint show internal-warning
41390 @item maint set internal-error backtrace @r{[}on|off@r{]}
41391 @itemx maint show internal-error backtrace
41392 @itemx maint set internal-warning backtrace @r{[}on|off@r{]}
41393 @itemx maint show internal-warning backtrace
41394 When @value{GDBN} reports an internal problem (error or warning) it is
41395 possible to have a backtrace of @value{GDBN} printed to the standard
41396 error stream. This is @samp{on} by default for @code{internal-error}
41397 and @samp{off} by default for @code{internal-warning}.
41399 @anchor{maint packet}
41400 @kindex maint packet
41401 @item maint packet @var{text}
41402 If @value{GDBN} is talking to an inferior via the serial protocol,
41403 then this command sends the string @var{text} to the inferior, and
41404 displays the response packet. @value{GDBN} supplies the initial
41405 @samp{$} character, the terminating @samp{#} character, and the
41408 Any non-printable characters in the reply are printed as escaped hex,
41409 e.g. @samp{\x00}, @samp{\x01}, etc.
41411 @kindex maint print architecture
41412 @item maint print architecture @r{[}@var{file}@r{]}
41413 Print the entire architecture configuration. The optional argument
41414 @var{file} names the file where the output goes.
41416 @kindex maint print c-tdesc
41417 @item maint print c-tdesc @r{[}-single-feature@r{]} @r{[}@var{file}@r{]}
41418 Print the target description (@pxref{Target Descriptions}) as
41419 a C source file. By default, the target description is for the current
41420 target, but if the optional argument @var{file} is provided, that file
41421 is used to produce the description. The @var{file} should be an XML
41422 document, of the form described in @ref{Target Description Format}.
41423 The created source file is built into @value{GDBN} when @value{GDBN} is
41424 built again. This command is used by developers after they add or
41425 modify XML target descriptions.
41427 When the optional flag @samp{-single-feature} is provided then the
41428 target description being processed (either the default, or from
41429 @var{file}) must only contain a single feature. The source file
41430 produced is different in this case.
41432 @kindex maint print xml-tdesc
41433 @item maint print xml-tdesc @r{[}@var{file}@r{]}
41434 Print the target description (@pxref{Target Descriptions}) as an XML
41435 file. By default print the target description for the current target,
41436 but if the optional argument @var{file} is provided, then that file is
41437 read in by GDB and then used to produce the description. The
41438 @var{file} should be an XML document, of the form described in
41439 @ref{Target Description Format}.
41441 @kindex maint check xml-descriptions
41442 @item maint check xml-descriptions @var{dir}
41443 Check that the target descriptions dynamically created by @value{GDBN}
41444 equal the descriptions created from XML files found in @var{dir}.
41446 @anchor{maint check libthread-db}
41447 @kindex maint check libthread-db
41448 @item maint check libthread-db
41449 Run integrity checks on the current inferior's thread debugging
41450 library. This exercises all @code{libthread_db} functionality used by
41451 @value{GDBN} on GNU/Linux systems, and by extension also exercises the
41452 @code{proc_service} functions provided by @value{GDBN} that
41453 @code{libthread_db} uses. Note that parts of the test may be skipped
41454 on some platforms when debugging core files.
41456 @kindex maint print core-file-backed-mappings
41457 @cindex memory address space mappings
41458 @item maint print core-file-backed-mappings
41459 Print the file-backed mappings which were loaded from a core file note.
41460 This output represents state internal to @value{GDBN} and should be
41461 similar to the mappings displayed by the @code{info proc mappings}
41464 @kindex maint print dummy-frames
41465 @item maint print dummy-frames
41466 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
41469 (@value{GDBP}) @kbd{b add}
41471 (@value{GDBP}) @kbd{print add(2,3)}
41472 Breakpoint 2, add (a=2, b=3) at @dots{}
41474 The program being debugged stopped while in a function called from GDB.
41476 (@value{GDBP}) @kbd{maint print dummy-frames}
41477 0xa8206d8: id=@{stack=0xbfffe734,code=0xbfffe73f,!special@}, ptid=process 9353
41481 Takes an optional file parameter.
41483 @kindex maint print frame-id
41484 @item maint print frame-id
41485 @itemx maint print frame-id @var{level}
41486 Print @value{GDBN}'s internal frame-id for the frame at relative
41487 @var{level}, or for the currently selected frame when @var{level} is
41490 If used, @var{level} should be an integer, as displayed in the
41491 @command{backtrace} output.
41494 (@value{GDBP}) maint print frame-id
41495 frame-id for frame #0: @{stack=0x7fffffffac70,code=0x0000000000401106,!special@}
41496 (@value{GDBP}) maint print frame-id 2
41497 frame-id for frame #2: @{stack=0x7fffffffac90,code=0x000000000040111c,!special@}
41500 @kindex maint print registers
41501 @kindex maint print raw-registers
41502 @kindex maint print cooked-registers
41503 @kindex maint print register-groups
41504 @kindex maint print remote-registers
41505 @item maint print registers @r{[}@var{file}@r{]}
41506 @itemx maint print raw-registers @r{[}@var{file}@r{]}
41507 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
41508 @itemx maint print register-groups @r{[}@var{file}@r{]}
41509 @itemx maint print remote-registers @r{[}@var{file}@r{]}
41510 Print @value{GDBN}'s internal register data structures.
41512 The command @code{maint print raw-registers} includes the contents of
41513 the raw register cache; the command @code{maint print
41514 cooked-registers} includes the (cooked) value of all registers,
41515 including registers which aren't available on the target nor visible
41516 to user; the command @code{maint print register-groups} includes the
41517 groups that each register is a member of; and the command @code{maint
41518 print remote-registers} includes the remote target's register numbers
41519 and offsets in the `G' packets.
41521 These commands take an optional parameter, a file name to which to
41522 write the information.
41524 @kindex maint print reggroups
41525 @item maint print reggroups @r{[}@var{file}@r{]}
41526 Print @value{GDBN}'s internal register group data structures. The
41527 optional argument @var{file} tells to what file to write the
41530 The register groups info looks like this:
41533 (@value{GDBP}) @kbd{maint print reggroups}
41544 @kindex maint flush register-cache
41546 @cindex register cache, flushing
41547 @item maint flush register-cache
41549 Flush the contents of the register cache and as a consequence the
41550 frame cache. This command is useful when debugging issues related to
41551 register fetching, or frame unwinding. The command @code{flushregs}
41552 is deprecated in favor of @code{maint flush register-cache}.
41554 @kindex maint flush source-cache
41555 @cindex source code, caching
41556 @item maint flush source-cache
41557 Flush @value{GDBN}'s cache of source code file contents. After
41558 @value{GDBN} reads a source file, and optionally applies styling
41559 (@pxref{Output Styling}), the file contents are cached. This command
41560 clears that cache. The next time @value{GDBN} wants to show lines
41561 from a source file, the content will be re-read.
41563 This command is useful when debugging issues related to source code
41564 styling. After flushing the cache any source code displayed by
41565 @value{GDBN} will be re-read and re-styled.
41567 @kindex maint print objfiles
41568 @cindex info for known object files
41569 @item maint print objfiles @r{[}@var{regexp}@r{]}
41570 Print a dump of all known object files.
41571 If @var{regexp} is specified, only print object files whose names
41572 match @var{regexp}. For each object file, this command prints its name,
41573 address in memory, and all of its psymtabs and symtabs.
41575 @kindex maint print user-registers
41576 @cindex user registers
41577 @item maint print user-registers
41578 List all currently available @dfn{user registers}. User registers
41579 typically provide alternate names for actual hardware registers. They
41580 include the four ``standard'' registers @code{$fp}, @code{$pc},
41581 @code{$sp}, and @code{$ps}. @xref{standard registers}. User
41582 registers can be used in expressions in the same way as the canonical
41583 register names, but only the latter are listed by the @code{info
41584 registers} and @code{maint print registers} commands.
41586 @kindex maint print section-scripts
41587 @cindex info for known .debug_gdb_scripts-loaded scripts
41588 @item maint print section-scripts [@var{regexp}]
41589 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
41590 If @var{regexp} is specified, only print scripts loaded by object files
41591 matching @var{regexp}.
41592 For each script, this command prints its name as specified in the objfile,
41593 and the full path if known.
41594 @xref{dotdebug_gdb_scripts section}.
41596 @kindex maint print statistics
41597 @cindex bcache statistics
41598 @item maint print statistics
41599 This command prints, for each object file in the program, various data
41600 about that object file followed by the byte cache (@dfn{bcache})
41601 statistics for the object file. The objfile data includes the number
41602 of minimal, partial, full, and stabs symbols, the number of types
41603 defined by the objfile, the number of as yet unexpanded psym tables,
41604 the number of line tables and string tables, and the amount of memory
41605 used by the various tables. The bcache statistics include the counts,
41606 sizes, and counts of duplicates of all and unique objects, max,
41607 average, and median entry size, total memory used and its overhead and
41608 savings, and various measures of the hash table size and chain
41611 @kindex maint print target-stack
41612 @cindex target stack description
41613 @item maint print target-stack
41614 A @dfn{target} is an interface between the debugger and a particular
41615 kind of file or process. Targets can be stacked in @dfn{strata},
41616 so that more than one target can potentially respond to a request.
41617 In particular, memory accesses will walk down the stack of targets
41618 until they find a target that is interested in handling that particular
41621 This command prints a short description of each layer that was pushed on
41622 the @dfn{target stack}, starting from the top layer down to the bottom one.
41624 @kindex maint print type
41625 @cindex type chain of a data type
41626 @item maint print type @var{expr}
41627 Print the type chain for a type specified by @var{expr}. The argument
41628 can be either a type name or a symbol. If it is a symbol, the type of
41629 that symbol is described. The type chain produced by this command is
41630 a recursive definition of the data type as stored in @value{GDBN}'s
41631 data structures, including its flags and contained types.
41633 @kindex maint print record-instruction
41634 @item maint print record-instruction
41635 @itemx maint print record-instruction @var{N}
41636 print how GDB recorded a given instruction. If @var{n} is not positive
41637 number, it prints the values stored by the inferior before the @var{n}-th previous
41638 instruction was exectued. If @var{n} is positive, print the values after the @var{n}-th
41639 following instruction is executed. If @var{n} is not given, 0 is assumed.
41641 @kindex maint selftest
41643 @item maint selftest @r{[}-verbose@r{]} @r{[}@var{filter}@r{]}
41644 Run any self tests that were compiled in to @value{GDBN}. This will
41645 print a message showing how many tests were run, and how many failed.
41646 If a @var{filter} is passed, only the tests with @var{filter} in their
41647 name will be ran. If @code{-verbose} is passed, the self tests can be
41650 @kindex maint set selftest verbose
41651 @kindex maint show selftest verbose
41653 @item maint set selftest verbose
41654 @item maint show selftest verbose
41655 Control whether self tests are run verbosely or not.
41657 @kindex maint info selftests
41659 @item maint info selftests
41660 List the selftests compiled in to @value{GDBN}.
41662 @kindex maint set dwarf always-disassemble
41663 @kindex maint show dwarf always-disassemble
41664 @item maint set dwarf always-disassemble
41665 @item maint show dwarf always-disassemble
41666 Control the behavior of @code{info address} when using DWARF debugging
41669 The default is @code{off}, which means that @value{GDBN} should try to
41670 describe a variable's location in an easily readable format. When
41671 @code{on}, @value{GDBN} will instead display the DWARF location
41672 expression in an assembly-like format. Note that some locations are
41673 too complex for @value{GDBN} to describe simply; in this case you will
41674 always see the disassembly form.
41676 Here is an example of the resulting disassembly:
41679 (@value{GDBP}) info addr argc
41680 Symbol "argc" is a complex DWARF expression:
41684 For more information on these expressions, see
41685 @uref{http://www.dwarfstd.org/, the DWARF standard}.
41687 @kindex maint set dwarf max-cache-age
41688 @kindex maint show dwarf max-cache-age
41689 @item maint set dwarf max-cache-age
41690 @itemx maint show dwarf max-cache-age
41691 Control the DWARF compilation unit cache.
41693 @cindex DWARF compilation units cache
41694 In object files with inter-compilation-unit references, such as those
41695 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF
41696 reader needs to frequently refer to previously read compilation units.
41697 This setting controls how long a compilation unit will remain in the
41698 cache if it is not referenced. A higher limit means that cached
41699 compilation units will be stored in memory longer, and more total
41700 memory will be used. Setting it to zero disables caching, which will
41701 slow down @value{GDBN} startup, but reduce memory consumption.
41703 @kindex maint set dwarf unwinders
41704 @kindex maint show dwarf unwinders
41705 @item maint set dwarf unwinders
41706 @itemx maint show dwarf unwinders
41707 Control use of the DWARF frame unwinders.
41709 @cindex DWARF frame unwinders
41710 Many targets that support DWARF debugging use @value{GDBN}'s DWARF
41711 frame unwinders to build the backtrace. Many of these targets will
41712 also have a second mechanism for building the backtrace for use in
41713 cases where DWARF information is not available, this second mechanism
41714 is often an analysis of a function's prologue.
41716 In order to extend testing coverage of the second level stack
41717 unwinding mechanisms it is helpful to be able to disable the DWARF
41718 stack unwinders, this can be done with this switch.
41720 In normal use of @value{GDBN} disabling the DWARF unwinders is not
41721 advisable, there are cases that are better handled through DWARF than
41722 prologue analysis, and the debug experience is likely to be better
41723 with the DWARF frame unwinders enabled.
41725 If DWARF frame unwinders are not supported for a particular target
41726 architecture, then enabling this flag does not cause them to be used.
41728 @kindex maint info frame-unwinders
41729 @item maint info frame-unwinders
41730 List the frame unwinders currently in effect, starting with the highest priority.
41732 @kindex maint set worker-threads
41733 @kindex maint show worker-threads
41734 @item maint set worker-threads
41735 @item maint show worker-threads
41736 Control the number of worker threads that may be used by @value{GDBN}.
41737 On capable hosts, @value{GDBN} may use multiple threads to speed up
41738 certain CPU-intensive operations, such as demangling symbol names.
41739 While the number of threads used by @value{GDBN} may vary, this
41740 command can be used to set an upper bound on this number. The default
41741 is @code{unlimited}, which lets @value{GDBN} choose a reasonable
41742 number. Note that this only controls worker threads started by
41743 @value{GDBN} itself; libraries used by @value{GDBN} may start threads
41746 @kindex maint set profile
41747 @kindex maint show profile
41748 @cindex profiling GDB
41749 @item maint set profile
41750 @itemx maint show profile
41751 Control profiling of @value{GDBN}.
41753 Profiling will be disabled until you use the @samp{maint set profile}
41754 command to enable it. When you enable profiling, the system will begin
41755 collecting timing and execution count data; when you disable profiling or
41756 exit @value{GDBN}, the results will be written to a log file. Remember that
41757 if you use profiling, @value{GDBN} will overwrite the profiling log file
41758 (often called @file{gmon.out}). If you have a record of important profiling
41759 data in a @file{gmon.out} file, be sure to move it to a safe location.
41761 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
41762 compiled with the @samp{-pg} compiler option.
41764 @kindex maint set show-debug-regs
41765 @kindex maint show show-debug-regs
41766 @cindex hardware debug registers
41767 @item maint set show-debug-regs
41768 @itemx maint show show-debug-regs
41769 Control whether to show variables that mirror the hardware debug
41770 registers. Use @code{on} to enable, @code{off} to disable. If
41771 enabled, the debug registers values are shown when @value{GDBN} inserts or
41772 removes a hardware breakpoint or watchpoint, and when the inferior
41773 triggers a hardware-assisted breakpoint or watchpoint.
41775 @kindex maint set show-all-tib
41776 @kindex maint show show-all-tib
41777 @item maint set show-all-tib
41778 @itemx maint show show-all-tib
41779 Control whether to show all non zero areas within a 1k block starting
41780 at thread local base, when using the @samp{info w32 thread-information-block}
41783 @kindex maint set target-async
41784 @kindex maint show target-async
41785 @item maint set target-async
41786 @itemx maint show target-async
41787 This controls whether @value{GDBN} targets operate in synchronous or
41788 asynchronous mode (@pxref{Background Execution}). Normally the
41789 default is asynchronous, if it is available; but this can be changed
41790 to more easily debug problems occurring only in synchronous mode.
41792 @kindex maint set target-non-stop @var{mode} [on|off|auto]
41793 @kindex maint show target-non-stop
41794 @item maint set target-non-stop
41795 @itemx maint show target-non-stop
41797 This controls whether @value{GDBN} targets always operate in non-stop
41798 mode even if @code{set non-stop} is @code{off} (@pxref{Non-Stop
41799 Mode}). The default is @code{auto}, meaning non-stop mode is enabled
41800 if supported by the target.
41803 @item maint set target-non-stop auto
41804 This is the default mode. @value{GDBN} controls the target in
41805 non-stop mode if the target supports it.
41807 @item maint set target-non-stop on
41808 @value{GDBN} controls the target in non-stop mode even if the target
41809 does not indicate support.
41811 @item maint set target-non-stop off
41812 @value{GDBN} does not control the target in non-stop mode even if the
41813 target supports it.
41816 @kindex maint set tui-resize-message
41817 @kindex maint show tui-resize-message
41818 @item maint set tui-resize-message
41819 @item maint show tui-resize-message
41820 Control whether @value{GDBN} displays a message each time the terminal
41821 is resized when in TUI mode. The default is @code{off}, which means
41822 that @value{GDBN} is silent during resizes. When @code{on},
41823 @value{GDBN} will display a message after a resize is completed; the
41824 message will include a number indicating how many times the terminal
41825 has been resized. This setting is intended for use by the test suite,
41826 where it would otherwise be difficult to determine when a resize and
41827 refresh has been completed.
41829 @kindex maint set tui-left-margin-verbose
41830 @kindex maint show tui-left-margin-verbose
41831 @item maint set tui-left-margin-verbose
41832 @item maint show tui-left-margin-verbose
41833 Control whether the left margin of the TUI source and disassembly windows
41834 uses @samp{_} and @samp{0} at locations where otherwise there would be a
41835 space. The default is @code{off}, which means spaces are used. The
41836 setting is intended to make it clear where the left margin begins and
41837 ends, to avoid incorrectly interpreting a space as being part of the
41840 @kindex maint set per-command
41841 @kindex maint show per-command
41842 @item maint set per-command
41843 @itemx maint show per-command
41844 @cindex resources used by commands
41846 @value{GDBN} can display the resources used by each command.
41847 This is useful in debugging performance problems.
41850 @item maint set per-command space [on|off]
41851 @itemx maint show per-command space
41852 Enable or disable the printing of the memory used by GDB for each command.
41853 If enabled, @value{GDBN} will display how much memory each command
41854 took, following the command's own output.
41855 This can also be requested by invoking @value{GDBN} with the
41856 @option{--statistics} command-line switch (@pxref{Mode Options}).
41858 @item maint set per-command time [on|off]
41859 @itemx maint show per-command time
41860 Enable or disable the printing of the execution time of @value{GDBN}
41862 If enabled, @value{GDBN} will display how much time it
41863 took to execute each command, following the command's own output.
41864 Both CPU time and wallclock time are printed.
41865 Printing both is useful when trying to determine whether the cost is
41866 CPU or, e.g., disk/network latency.
41867 Note that the CPU time printed is for @value{GDBN} only, it does not include
41868 the execution time of the inferior because there's no mechanism currently
41869 to compute how much time was spent by @value{GDBN} and how much time was
41870 spent by the program been debugged.
41871 This can also be requested by invoking @value{GDBN} with the
41872 @option{--statistics} command-line switch (@pxref{Mode Options}).
41874 @item maint set per-command symtab [on|off]
41875 @itemx maint show per-command symtab
41876 Enable or disable the printing of basic symbol table statistics
41878 If enabled, @value{GDBN} will display the following information:
41882 number of symbol tables
41884 number of primary symbol tables
41886 number of blocks in the blockvector
41890 @kindex maint set check-libthread-db
41891 @kindex maint show check-libthread-db
41892 @item maint set check-libthread-db [on|off]
41893 @itemx maint show check-libthread-db
41894 Control whether @value{GDBN} should run integrity checks on inferior
41895 specific thread debugging libraries as they are loaded. The default
41896 is not to perform such checks. If any check fails @value{GDBN} will
41897 unload the library and continue searching for a suitable candidate as
41898 described in @ref{set libthread-db-search-path}. For more information
41899 about the tests, see @ref{maint check libthread-db}.
41901 @kindex maint set gnu-source-highlight enabled
41902 @kindex maint show gnu-source-highlight enabled
41903 @item maint set gnu-source-highlight enabled @r{[}on|off@r{]}
41904 @itemx maint show gnu-source-highlight enabled
41905 Control whether @value{GDBN} should use the GNU Source Highlight
41906 library for applying styling to source code (@pxref{Output Styling}).
41907 This will be @samp{on} by default if the GNU Source Highlight library
41908 is available. If the GNU Source Highlight library is not available,
41909 then this will be @samp{off} by default, and attempting to change this
41910 value to @samp{on} will give an error.
41912 If the GNU Source Highlight library is not being used, then
41913 @value{GDBN} will use the Python Pygments package for source code
41914 styling, if it is available.
41916 This option is useful for debugging @value{GDBN}'s use of the Pygments
41917 library when @value{GDBN} is linked against the GNU Source Highlight
41920 @anchor{maint_libopcodes_styling}
41921 @kindex maint set libopcodes-styling enabled
41922 @kindex maint show libopcodes-styling enabled
41923 @item maint set libopcodes-styling enabled @r{[}on|off@r{]}
41924 @itemx maint show libopcodes-styling enabled
41925 Control whether @value{GDBN} should use its builtin disassembler
41926 (@file{libopcodes}) to style disassembler output (@pxref{Output
41927 Styling}). The builtin disassembler does not support styling for all
41930 When this option is @samp{off} the builtin disassembler will not be
41931 used for styling, @value{GDBN} will fall back to using the Python
41932 Pygments package if possible.
41934 Trying to set this option @samp{on} for an architecture that the
41935 builtin disassembler is unable to style will give an error, otherwise,
41936 the builtin disassembler will be used to style disassembler output.
41938 This option is @samp{on} by default for supported architectures.
41940 This option is useful for debugging @value{GDBN}'s use of the Pygments
41941 library when @value{GDBN} is built for an architecture that supports
41942 styling with the builtin disassembler
41944 @kindex maint info screen
41945 @cindex show screen characteristics
41946 @item maint info screen
41947 Print various characteristics of the screen, such as various notions
41948 of width and height.
41950 @kindex maint space
41951 @cindex memory used by commands
41952 @item maint space @var{value}
41953 An alias for @code{maint set per-command space}.
41954 A non-zero value enables it, zero disables it.
41957 @cindex time of command execution
41958 @item maint time @var{value}
41959 An alias for @code{maint set per-command time}.
41960 A non-zero value enables it, zero disables it.
41962 @kindex maint translate-address
41963 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
41964 Find the symbol stored at the location specified by the address
41965 @var{addr} and an optional section name @var{section}. If found,
41966 @value{GDBN} prints the name of the closest symbol and an offset from
41967 the symbol's location to the specified address. This is similar to
41968 the @code{info address} command (@pxref{Symbols}), except that this
41969 command also allows to find symbols in other sections.
41971 If section was not specified, the section in which the symbol was found
41972 is also printed. For dynamically linked executables, the name of
41973 executable or shared library containing the symbol is printed as well.
41975 @kindex maint test-options
41976 @item maint test-options require-delimiter
41977 @itemx maint test-options unknown-is-error
41978 @itemx maint test-options unknown-is-operand
41979 These commands are used by the testsuite to validate the command
41980 options framework. The @code{require-delimiter} variant requires a
41981 double-dash delimiter to indicate end of options. The
41982 @code{unknown-is-error} and @code{unknown-is-operand} do not. The
41983 @code{unknown-is-error} variant throws an error on unknown option,
41984 while @code{unknown-is-operand} treats unknown options as the start of
41985 the command's operands. When run, the commands output the result of
41986 the processed options. When completed, the commands store the
41987 internal result of completion in a variable exposed by the @code{maint
41988 show test-options-completion-result} command.
41990 @kindex maint show test-options-completion-result
41991 @item maint show test-options-completion-result
41992 Shows the result of completing the @code{maint test-options}
41993 subcommands. This is used by the testsuite to validate completion
41994 support in the command options framework.
41996 @kindex maint set test-settings
41997 @kindex maint show test-settings
41998 @item maint set test-settings @var{kind}
41999 @itemx maint show test-settings @var{kind}
42000 These are representative commands for each @var{kind} of setting type
42001 @value{GDBN} supports. They are used by the testsuite for exercising
42002 the settings infrastructure.
42004 @kindex maint set backtrace-on-fatal-signal
42005 @kindex maint show backtrace-on-fatal-signal
42006 @item maint set backtrace-on-fatal-signal [on|off]
42007 @itemx maint show backtrace-on-fatal-signal
42008 When this setting is @code{on}, if @value{GDBN} itself terminates with
42009 a fatal signal (e.g.@: SIGSEGV), then a limited backtrace will be
42010 printed to the standard error stream. This backtrace can be used to
42011 help diagnose crashes within @value{GDBN} in situations where a user
42012 is unable to share a corefile with the @value{GDBN} developers.
42014 If the functionality to provide this backtrace is not available for
42015 the platform on which GDB is running then this feature will be
42016 @code{off} by default, and attempting to turn this feature on will
42019 For platforms that do support creating the backtrace this feature is
42020 @code{on} by default.
42022 @kindex maint wait-for-index-cache
42023 @item maint wait-for-index-cache
42024 Wait until all pending writes to the index cache have completed. This
42025 is used by the test suite to avoid races when the index cache is being
42026 updated by a worker thread.
42029 @item maint with @var{setting} [@var{value}] [-- @var{command}]
42030 Like the @code{with} command, but works with @code{maintenance set}
42031 variables. This is used by the testsuite to exercise the @code{with}
42032 command's infrastructure.
42034 @kindex maint ignore-probes
42035 @item maint ignore-probes [@var{-v}|@var{-verbose}] [@var{provider} [@var{name} [@var{objfile}]]]
42036 @itemx maint ignore-probes @var{-reset}
42037 Set or reset the ignore-probes filter. The @var{provider}, @var{name}
42038 and @var{objfile} arguments are as in @code{enable probes} and
42039 @code{disable probes} (@pxref{enable probes}). Only supported for
42042 Here's an example of using @code{maint ignore-probes}:
42044 (gdb) maint ignore-probes -verbose libc ^longjmp$
42045 ignore-probes filter has been set to:
42047 PROBE_NAME: '^longjmp$'
42050 <... more output ...>
42051 Ignoring SystemTap probe libc longjmp in /lib64/libc.so.6.^M
42052 Ignoring SystemTap probe libc longjmp in /lib64/libc.so.6.^M
42053 Ignoring SystemTap probe libc longjmp in /lib64/libc.so.6.^M
42057 The following command is useful for non-interactive invocations of
42058 @value{GDBN}, such as in the test suite.
42061 @item set watchdog @var{nsec}
42062 @kindex set watchdog
42063 @cindex watchdog timer
42064 @cindex timeout for commands
42065 Set the maximum number of seconds @value{GDBN} will wait for the
42066 target operation to finish. If this time expires, @value{GDBN}
42067 reports and error and the command is aborted.
42069 @item show watchdog
42070 Show the current setting of the target wait timeout.
42073 @node Remote Protocol
42074 @appendix @value{GDBN} Remote Serial Protocol
42079 * Stop Reply Packets::
42080 * General Query Packets::
42081 * Architecture-Specific Protocol Details::
42082 * Tracepoint Packets::
42083 * Host I/O Packets::
42085 * Notification Packets::
42086 * Remote Non-Stop::
42087 * Packet Acknowledgment::
42089 * File-I/O Remote Protocol Extension::
42090 * Library List Format::
42091 * Library List Format for SVR4 Targets::
42092 * Memory Map Format::
42093 * Thread List Format::
42094 * Traceframe Info Format::
42095 * Branch Trace Format::
42096 * Branch Trace Configuration Format::
42102 There may be occasions when you need to know something about the
42103 protocol---for example, if there is only one serial port to your target
42104 machine, you might want your program to do something special if it
42105 recognizes a packet meant for @value{GDBN}.
42107 In the examples below, @samp{->} and @samp{<-} are used to indicate
42108 transmitted and received data, respectively.
42110 @cindex protocol, @value{GDBN} remote serial
42111 @cindex serial protocol, @value{GDBN} remote
42112 @cindex remote serial protocol
42113 All @value{GDBN} commands and responses (other than acknowledgments
42114 and notifications, see @ref{Notification Packets}) are sent as a
42115 @var{packet}. A @var{packet} is introduced with the character
42116 @samp{$}, the actual @var{packet-data}, and the terminating character
42117 @samp{#} followed by a two-digit @var{checksum}:
42120 @code{$}@var{packet-data}@code{#}@var{checksum}
42124 @cindex checksum, for @value{GDBN} remote
42126 The two-digit @var{checksum} is computed as the modulo 256 sum of all
42127 characters between the leading @samp{$} and the trailing @samp{#} (an
42128 eight bit unsigned checksum).
42130 Implementors should note that prior to @value{GDBN} 5.0 the protocol
42131 specification also included an optional two-digit @var{sequence-id}:
42134 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
42137 @cindex sequence-id, for @value{GDBN} remote
42139 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
42140 has never output @var{sequence-id}s. Stubs that handle packets added
42141 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
42143 When either the host or the target machine receives a packet, the first
42144 response expected is an acknowledgment: either @samp{+} (to indicate
42145 the package was received correctly) or @samp{-} (to request
42149 -> @code{$}@var{packet-data}@code{#}@var{checksum}
42154 The @samp{+}/@samp{-} acknowledgments can be disabled
42155 once a connection is established.
42156 @xref{Packet Acknowledgment}, for details.
42158 The host (@value{GDBN}) sends @var{command}s, and the target (the
42159 debugging stub incorporated in your program) sends a @var{response}. In
42160 the case of step and continue @var{command}s, the response is only sent
42161 when the operation has completed, and the target has again stopped all
42162 threads in all attached processes. This is the default all-stop mode
42163 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
42164 execution mode; see @ref{Remote Non-Stop}, for details.
42166 @var{packet-data} consists of a sequence of characters with the
42167 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
42170 @cindex remote protocol, field separator
42171 Fields within the packet should be separated using @samp{,} @samp{;} or
42172 @samp{:}. Except where otherwise noted all numbers are represented in
42173 @sc{hex} with leading zeros suppressed.
42175 Implementors should note that prior to @value{GDBN} 5.0, the character
42176 @samp{:} could not appear as the third character in a packet (as it
42177 would potentially conflict with the @var{sequence-id}).
42179 @cindex remote protocol, binary data
42180 @anchor{Binary Data}
42181 Binary data in most packets is encoded either as two hexadecimal
42182 digits per byte of binary data. This allowed the traditional remote
42183 protocol to work over connections which were only seven-bit clean.
42184 Some packets designed more recently assume an eight-bit clean
42185 connection, and use a more efficient encoding to send and receive
42188 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
42189 as an escape character. Any escaped byte is transmitted as the escape
42190 character followed by the original character XORed with @code{0x20}.
42191 For example, the byte @code{0x7d} would be transmitted as the two
42192 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
42193 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
42194 @samp{@}}) must always be escaped. Responses sent by the stub
42195 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
42196 is not interpreted as the start of a run-length encoded sequence
42199 Response @var{data} can be run-length encoded to save space.
42200 Run-length encoding replaces runs of identical characters with one
42201 instance of the repeated character, followed by a @samp{*} and a
42202 repeat count. The repeat count is itself sent encoded, to avoid
42203 binary characters in @var{data}: a value of @var{n} is sent as
42204 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
42205 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
42206 code 32) for a repeat count of 3. (This is because run-length
42207 encoding starts to win for counts 3 or more.) Thus, for example,
42208 @samp{0* } is a run-length encoding of ``0000'': the space character
42209 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
42212 The printable characters @samp{#} and @samp{$} or with a numeric value
42213 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
42214 seven repeats (@samp{$}) can be expanded using a repeat count of only
42215 five (@samp{"}). For example, @samp{00000000} can be encoded as
42218 The error response returned for some packets includes a two character
42219 error number. That number is not well defined.
42221 @cindex empty response, for unsupported packets
42222 For any @var{command} not supported by the stub, an empty response
42223 (@samp{$#00}) should be returned. That way it is possible to extend the
42224 protocol. A newer @value{GDBN} can tell if a packet is supported based
42227 At a minimum, a stub is required to support the @samp{?} command to
42228 tell @value{GDBN} the reason for halting, @samp{g} and @samp{G}
42229 commands for register access, and the @samp{m} and @samp{M} commands
42230 for memory access. Stubs that only control single-threaded targets
42231 can implement run control with the @samp{c} (continue) command, and if
42232 the target architecture supports hardware-assisted single-stepping,
42233 the @samp{s} (step) command. Stubs that support multi-threading
42234 targets should support the @samp{vCont} command. All other commands
42240 The following table provides a complete list of all currently defined
42241 @var{command}s and their corresponding response @var{data}.
42242 @xref{File-I/O Remote Protocol Extension}, for details about the File
42243 I/O extension of the remote protocol.
42245 Each packet's description has a template showing the packet's overall
42246 syntax, followed by an explanation of the packet's meaning. We
42247 include spaces in some of the templates for clarity; these are not
42248 part of the packet's syntax. No @value{GDBN} packet uses spaces to
42249 separate its components. For example, a template like @samp{foo
42250 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
42251 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
42252 @var{baz}. @value{GDBN} does not transmit a space character between the
42253 @samp{foo} and the @var{bar}, or between the @var{bar} and the
42256 @cindex @var{thread-id}, in remote protocol
42257 @anchor{thread-id syntax}
42258 Several packets and replies include a @var{thread-id} field to identify
42259 a thread. Normally these are positive numbers with a target-specific
42260 interpretation, formatted as big-endian hex strings. A @var{thread-id}
42261 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
42264 In addition, the remote protocol supports a multiprocess feature in
42265 which the @var{thread-id} syntax is extended to optionally include both
42266 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
42267 The @var{pid} (process) and @var{tid} (thread) components each have the
42268 format described above: a positive number with target-specific
42269 interpretation formatted as a big-endian hex string, literal @samp{-1}
42270 to indicate all processes or threads (respectively), or @samp{0} to
42271 indicate an arbitrary process or thread. Specifying just a process, as
42272 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
42273 error to specify all processes but a specific thread, such as
42274 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
42275 for those packets and replies explicitly documented to include a process
42276 ID, rather than a @var{thread-id}.
42278 The multiprocess @var{thread-id} syntax extensions are only used if both
42279 @value{GDBN} and the stub report support for the @samp{multiprocess}
42280 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
42283 Note that all packet forms beginning with an upper- or lower-case
42284 letter, other than those described here, are reserved for future use.
42286 Here are the packet descriptions.
42291 @cindex @samp{!} packet
42292 @anchor{extended mode}
42293 Enable extended mode. In extended mode, the remote server is made
42294 persistent. The @samp{R} packet is used to restart the program being
42300 The remote target both supports and has enabled extended mode.
42304 @cindex @samp{?} packet
42306 This is sent when connection is first established to query the reason
42307 the target halted. The reply is the same as for step and continue.
42308 This packet has a special interpretation when the target is in
42309 non-stop mode; see @ref{Remote Non-Stop}.
42312 @xref{Stop Reply Packets}, for the reply specifications.
42314 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
42315 @cindex @samp{A} packet
42316 Initialized @code{argv[]} array passed into program. @var{arglen}
42317 specifies the number of bytes in the hex encoded byte stream
42318 @var{arg}. See @code{gdbserver} for more details.
42323 The arguments were set.
42329 @cindex @samp{b} packet
42330 (Don't use this packet; its behavior is not well-defined.)
42331 Change the serial line speed to @var{baud}.
42333 JTC: @emph{When does the transport layer state change? When it's
42334 received, or after the ACK is transmitted. In either case, there are
42335 problems if the command or the acknowledgment packet is dropped.}
42337 Stan: @emph{If people really wanted to add something like this, and get
42338 it working for the first time, they ought to modify ser-unix.c to send
42339 some kind of out-of-band message to a specially-setup stub and have the
42340 switch happen "in between" packets, so that from remote protocol's point
42341 of view, nothing actually happened.}
42343 @item B @var{addr},@var{mode}
42344 @cindex @samp{B} packet
42345 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
42346 breakpoint at @var{addr}.
42348 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
42349 (@pxref{insert breakpoint or watchpoint packet}).
42351 @cindex @samp{bc} packet
42354 Backward continue. Execute the target system in reverse. No parameter.
42355 @xref{Reverse Execution}, for more information.
42358 @xref{Stop Reply Packets}, for the reply specifications.
42360 @cindex @samp{bs} packet
42363 Backward single step. Execute one instruction in reverse. No parameter.
42364 @xref{Reverse Execution}, for more information.
42367 @xref{Stop Reply Packets}, for the reply specifications.
42369 @item c @r{[}@var{addr}@r{]}
42370 @cindex @samp{c} packet
42371 Continue at @var{addr}, which is the address to resume. If @var{addr}
42372 is omitted, resume at current address.
42374 This packet is deprecated for multi-threading support. @xref{vCont
42378 @xref{Stop Reply Packets}, for the reply specifications.
42380 @item C @var{sig}@r{[};@var{addr}@r{]}
42381 @cindex @samp{C} packet
42382 Continue with signal @var{sig} (hex signal number). If
42383 @samp{;@var{addr}} is omitted, resume at same address.
42385 This packet is deprecated for multi-threading support. @xref{vCont
42389 @xref{Stop Reply Packets}, for the reply specifications.
42392 @cindex @samp{d} packet
42395 Don't use this packet; instead, define a general set packet
42396 (@pxref{General Query Packets}).
42400 @cindex @samp{D} packet
42401 The first form of the packet is used to detach @value{GDBN} from the
42402 remote system. It is sent to the remote target
42403 before @value{GDBN} disconnects via the @code{detach} command.
42405 The second form, including a process ID, is used when multiprocess
42406 protocol extensions are enabled (@pxref{multiprocess extensions}), to
42407 detach only a specific process. The @var{pid} is specified as a
42408 big-endian hex string.
42418 @item F @var{RC},@var{EE},@var{CF};@var{XX}
42419 @cindex @samp{F} packet
42420 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
42421 This is part of the File-I/O protocol extension. @xref{File-I/O
42422 Remote Protocol Extension}, for the specification.
42425 @anchor{read registers packet}
42426 @cindex @samp{g} packet
42427 Read general registers.
42431 @item @var{XX@dots{}}
42432 Each byte of register data is described by two hex digits. The bytes
42433 with the register are transmitted in target byte order. The size of
42434 each register and their position within the @samp{g} packet are
42435 determined by the target description (@pxref{Target Descriptions}); in
42436 the absence of a target description, this is done using code internal
42437 to @value{GDBN}; typically this is some customary register layout for
42438 the architecture in question.
42440 When reading registers, the stub may also return a string of literal
42441 @samp{x}'s in place of the register data digits, to indicate that the
42442 corresponding register's value is unavailable. For example, when
42443 reading registers from a trace frame (@pxref{Analyze Collected
42444 Data,,Using the Collected Data}), this means that the register has not
42445 been collected in the trace frame. When reading registers from a live
42446 program, this indicates that the stub has no means to access the
42447 register contents, even though the corresponding register is known to
42448 exist. Note that if a register truly does not exist on the target,
42449 then it is better to not include it in the target description in the
42452 For example, for an architecture with 4 registers of
42453 4 bytes each, the following reply indicates to @value{GDBN} that
42454 registers 0 and 2 are unavailable, while registers 1 and 3
42455 are available, and both have zero value:
42459 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
42466 @item G @var{XX@dots{}}
42467 @cindex @samp{G} packet
42468 Write general registers. @xref{read registers packet}, for a
42469 description of the @var{XX@dots{}} data.
42479 @item H @var{op} @var{thread-id}
42480 @cindex @samp{H} packet
42481 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
42482 @samp{G}, et.al.). Depending on the operation to be performed, @var{op}
42483 should be @samp{c} for step and continue operations (note that this
42484 is deprecated, supporting the @samp{vCont} command is a better
42485 option), and @samp{g} for other operations. The thread designator
42486 @var{thread-id} has the format and interpretation described in
42487 @ref{thread-id syntax}.
42498 @c 'H': How restrictive (or permissive) is the thread model. If a
42499 @c thread is selected and stopped, are other threads allowed
42500 @c to continue to execute? As I mentioned above, I think the
42501 @c semantics of each command when a thread is selected must be
42502 @c described. For example:
42504 @c 'g': If the stub supports threads and a specific thread is
42505 @c selected, returns the register block from that thread;
42506 @c otherwise returns current registers.
42508 @c 'G' If the stub supports threads and a specific thread is
42509 @c selected, sets the registers of the register block of
42510 @c that thread; otherwise sets current registers.
42512 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
42513 @anchor{cycle step packet}
42514 @cindex @samp{i} packet
42515 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
42516 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
42517 step starting at that address.
42520 @cindex @samp{I} packet
42521 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
42525 @cindex @samp{k} packet
42528 The exact effect of this packet is not specified.
42530 For a bare-metal target, it may power cycle or reset the target
42531 system. For that reason, the @samp{k} packet has no reply.
42533 For a single-process target, it may kill that process if possible.
42535 A multiple-process target may choose to kill just one process, or all
42536 that are under @value{GDBN}'s control. For more precise control, use
42537 the vKill packet (@pxref{vKill packet}).
42539 If the target system immediately closes the connection in response to
42540 @samp{k}, @value{GDBN} does not consider the lack of packet
42541 acknowledgment to be an error, and assumes the kill was successful.
42543 If connected using @kbd{target extended-remote}, and the target does
42544 not close the connection in response to a kill request, @value{GDBN}
42545 probes the target state as if a new connection was opened
42546 (@pxref{? packet}).
42548 @item m @var{addr},@var{length}
42549 @cindex @samp{m} packet
42550 Read @var{length} addressable memory units starting at address @var{addr}
42551 (@pxref{addressable memory unit}). Note that @var{addr} may not be aligned to
42552 any particular boundary.
42554 The stub need not use any particular size or alignment when gathering
42555 data from memory for the response; even if @var{addr} is word-aligned
42556 and @var{length} is a multiple of the word size, the stub is free to
42557 use byte accesses, or not. For this reason, this packet may not be
42558 suitable for accessing memory-mapped I/O devices.
42559 @cindex alignment of remote memory accesses
42560 @cindex size of remote memory accesses
42561 @cindex memory, alignment and size of remote accesses
42565 @item @var{XX@dots{}}
42566 Memory contents; each byte is transmitted as a two-digit hexadecimal number.
42567 The reply may contain fewer addressable memory units than requested if the
42568 server was able to read only part of the region of memory.
42573 @item M @var{addr},@var{length}:@var{XX@dots{}}
42574 @cindex @samp{M} packet
42575 Write @var{length} addressable memory units starting at address @var{addr}
42576 (@pxref{addressable memory unit}). The data is given by @var{XX@dots{}}; each
42577 byte is transmitted as a two-digit hexadecimal number.
42584 for an error (this includes the case where only part of the data was
42589 @cindex @samp{p} packet
42590 Read the value of register @var{n}; @var{n} is in hex.
42591 @xref{read registers packet}, for a description of how the returned
42592 register value is encoded.
42596 @item @var{XX@dots{}}
42597 the register's value
42601 Indicating an unrecognized @var{query}.
42604 @item P @var{n@dots{}}=@var{r@dots{}}
42605 @anchor{write register packet}
42606 @cindex @samp{P} packet
42607 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
42608 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
42609 digits for each byte in the register (target byte order).
42619 @item q @var{name} @var{params}@dots{}
42620 @itemx Q @var{name} @var{params}@dots{}
42621 @cindex @samp{q} packet
42622 @cindex @samp{Q} packet
42623 General query (@samp{q}) and set (@samp{Q}). These packets are
42624 described fully in @ref{General Query Packets}.
42627 @cindex @samp{r} packet
42628 Reset the entire system.
42630 Don't use this packet; use the @samp{R} packet instead.
42633 @cindex @samp{R} packet
42634 Restart the program being debugged. The @var{XX}, while needed, is ignored.
42635 This packet is only available in extended mode (@pxref{extended mode}).
42637 The @samp{R} packet has no reply.
42639 @item s @r{[}@var{addr}@r{]}
42640 @cindex @samp{s} packet
42641 Single step, resuming at @var{addr}. If
42642 @var{addr} is omitted, resume at same address.
42644 This packet is deprecated for multi-threading support. @xref{vCont
42648 @xref{Stop Reply Packets}, for the reply specifications.
42650 @item S @var{sig}@r{[};@var{addr}@r{]}
42651 @anchor{step with signal packet}
42652 @cindex @samp{S} packet
42653 Step with signal. This is analogous to the @samp{C} packet, but
42654 requests a single-step, rather than a normal resumption of execution.
42656 This packet is deprecated for multi-threading support. @xref{vCont
42660 @xref{Stop Reply Packets}, for the reply specifications.
42662 @item t @var{addr}:@var{PP},@var{MM}
42663 @cindex @samp{t} packet
42664 Search backwards starting at address @var{addr} for a match with pattern
42665 @var{PP} and mask @var{MM}, both of which are are 4 byte long.
42666 There must be at least 3 digits in @var{addr}.
42668 @item T @var{thread-id}
42669 @cindex @samp{T} packet
42670 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
42675 thread is still alive
42681 Packets starting with @samp{v} are identified by a multi-letter name,
42682 up to the first @samp{;} or @samp{?} (or the end of the packet).
42684 @item vAttach;@var{pid}
42685 @cindex @samp{vAttach} packet
42686 Attach to a new process with the specified process ID @var{pid}.
42687 The process ID is a
42688 hexadecimal integer identifying the process. In all-stop mode, all
42689 threads in the attached process are stopped; in non-stop mode, it may be
42690 attached without being stopped if that is supported by the target.
42692 @c In non-stop mode, on a successful vAttach, the stub should set the
42693 @c current thread to a thread of the newly-attached process. After
42694 @c attaching, GDB queries for the attached process's thread ID with qC.
42695 @c Also note that, from a user perspective, whether or not the
42696 @c target is stopped on attach in non-stop mode depends on whether you
42697 @c use the foreground or background version of the attach command, not
42698 @c on what vAttach does; GDB does the right thing with respect to either
42699 @c stopping or restarting threads.
42701 This packet is only available in extended mode (@pxref{extended mode}).
42707 @item @r{Any stop packet}
42708 for success in all-stop mode (@pxref{Stop Reply Packets})
42710 for success in non-stop mode (@pxref{Remote Non-Stop})
42713 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
42714 @cindex @samp{vCont} packet
42715 @anchor{vCont packet}
42716 Resume the inferior, specifying different actions for each thread.
42718 For each inferior thread, the leftmost action with a matching
42719 @var{thread-id} is applied. Threads that don't match any action
42720 remain in their current state. Thread IDs are specified using the
42721 syntax described in @ref{thread-id syntax}. If multiprocess
42722 extensions (@pxref{multiprocess extensions}) are supported, actions
42723 can be specified to match all threads in a process by using the
42724 @samp{p@var{pid}.-1} form of the @var{thread-id}. An action with no
42725 @var{thread-id} matches all threads. Specifying no actions is an
42728 Currently supported actions are:
42734 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
42738 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
42741 @item r @var{start},@var{end}
42742 Step once, and then keep stepping as long as the thread stops at
42743 addresses between @var{start} (inclusive) and @var{end} (exclusive).
42744 The remote stub reports a stop reply when either the thread goes out
42745 of the range or is stopped due to an unrelated reason, such as hitting
42746 a breakpoint. @xref{range stepping}.
42748 If the range is empty (@var{start} == @var{end}), then the action
42749 becomes equivalent to the @samp{s} action. In other words,
42750 single-step once, and report the stop (even if the stepped instruction
42751 jumps to @var{start}).
42753 (A stop reply may be sent at any point even if the PC is still within
42754 the stepping range; for example, it is valid to implement this packet
42755 in a degenerate way as a single instruction step operation.)
42759 The optional argument @var{addr} normally associated with the
42760 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
42761 not supported in @samp{vCont}.
42763 The @samp{t} action is only relevant in non-stop mode
42764 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
42765 A stop reply should be generated for any affected thread not already stopped.
42766 When a thread is stopped by means of a @samp{t} action,
42767 the corresponding stop reply should indicate that the thread has stopped with
42768 signal @samp{0}, regardless of whether the target uses some other signal
42769 as an implementation detail.
42771 The server must ignore @samp{c}, @samp{C}, @samp{s}, @samp{S}, and
42772 @samp{r} actions for threads that are already running. Conversely,
42773 the server must ignore @samp{t} actions for threads that are already
42776 @emph{Note:} In non-stop mode, a thread is considered running until
42777 @value{GDBN} acknowledges an asynchronous stop notification for it with
42778 the @samp{vStopped} packet (@pxref{Remote Non-Stop}).
42780 The stub must support @samp{vCont} if it reports support for
42781 multiprocess extensions (@pxref{multiprocess extensions}).
42784 @xref{Stop Reply Packets}, for the reply specifications.
42787 @cindex @samp{vCont?} packet
42788 Request a list of actions supported by the @samp{vCont} packet.
42792 @item vCont@r{[};@var{action}@dots{}@r{]}
42793 The @samp{vCont} packet is supported. Each @var{action} is a supported
42794 command in the @samp{vCont} packet.
42796 The @samp{vCont} packet is not supported.
42799 @anchor{vCtrlC packet}
42801 @cindex @samp{vCtrlC} packet
42802 Interrupt remote target as if a control-C was pressed on the remote
42803 terminal. This is the equivalent to reacting to the @code{^C}
42804 (@samp{\003}, the control-C character) character in all-stop mode
42805 while the target is running, except this works in non-stop mode.
42806 @xref{interrupting remote targets}, for more info on the all-stop
42817 @item vFile:@var{operation}:@var{parameter}@dots{}
42818 @cindex @samp{vFile} packet
42819 Perform a file operation on the target system. For details,
42820 see @ref{Host I/O Packets}.
42822 @item vFlashErase:@var{addr},@var{length}
42823 @cindex @samp{vFlashErase} packet
42824 Direct the stub to erase @var{length} bytes of flash starting at
42825 @var{addr}. The region may enclose any number of flash blocks, but
42826 its start and end must fall on block boundaries, as indicated by the
42827 flash block size appearing in the memory map (@pxref{Memory Map
42828 Format}). @value{GDBN} groups flash memory programming operations
42829 together, and sends a @samp{vFlashDone} request after each group; the
42830 stub is allowed to delay erase operation until the @samp{vFlashDone}
42831 packet is received.
42841 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
42842 @cindex @samp{vFlashWrite} packet
42843 Direct the stub to write data to flash address @var{addr}. The data
42844 is passed in binary form using the same encoding as for the @samp{X}
42845 packet (@pxref{Binary Data}). The memory ranges specified by
42846 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
42847 not overlap, and must appear in order of increasing addresses
42848 (although @samp{vFlashErase} packets for higher addresses may already
42849 have been received; the ordering is guaranteed only between
42850 @samp{vFlashWrite} packets). If a packet writes to an address that was
42851 neither erased by a preceding @samp{vFlashErase} packet nor by some other
42852 target-specific method, the results are unpredictable.
42860 for vFlashWrite addressing non-flash memory
42866 @cindex @samp{vFlashDone} packet
42867 Indicate to the stub that flash programming operation is finished.
42868 The stub is permitted to delay or batch the effects of a group of
42869 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
42870 @samp{vFlashDone} packet is received. The contents of the affected
42871 regions of flash memory are unpredictable until the @samp{vFlashDone}
42872 request is completed.
42874 @item vKill;@var{pid}
42875 @cindex @samp{vKill} packet
42876 @anchor{vKill packet}
42877 Kill the process with the specified process ID @var{pid}, which is a
42878 hexadecimal integer identifying the process. This packet is used in
42879 preference to @samp{k} when multiprocess protocol extensions are
42880 supported; see @ref{multiprocess extensions}.
42890 @item vMustReplyEmpty
42891 @cindex @samp{vMustReplyEmpty} packet
42892 The correct reply to an unknown @samp{v} packet is to return the empty
42893 string, however, some older versions of @command{gdbserver} would
42894 incorrectly return @samp{OK} for unknown @samp{v} packets.
42896 The @samp{vMustReplyEmpty} is used as a feature test to check how
42897 @command{gdbserver} handles unknown packets, it is important that this
42898 packet be handled in the same way as other unknown @samp{v} packets.
42899 If this packet is handled differently to other unknown @samp{v}
42900 packets then it is possible that @value{GDBN} may run into problems in
42901 other areas, specifically around use of @samp{vFile:setfs:}.
42903 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
42904 @cindex @samp{vRun} packet
42905 Run the program @var{filename}, passing it each @var{argument} on its
42906 command line. The file and arguments are hex-encoded strings. If
42907 @var{filename} is an empty string, the stub may use a default program
42908 (e.g.@: the last program run). The program is created in the stopped
42911 @c FIXME: What about non-stop mode?
42913 This packet is only available in extended mode (@pxref{extended mode}).
42919 @item @r{Any stop packet}
42920 for success (@pxref{Stop Reply Packets})
42924 @cindex @samp{vStopped} packet
42925 @xref{Notification Packets}.
42927 @item X @var{addr},@var{length}:@var{XX@dots{}}
42929 @cindex @samp{X} packet
42930 Write data to memory, where the data is transmitted in binary.
42931 Memory is specified by its address @var{addr} and number of addressable memory
42932 units @var{length} (@pxref{addressable memory unit});
42933 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
42943 @item z @var{type},@var{addr},@var{kind}
42944 @itemx Z @var{type},@var{addr},@var{kind}
42945 @anchor{insert breakpoint or watchpoint packet}
42946 @cindex @samp{z} packet
42947 @cindex @samp{Z} packets
42948 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
42949 watchpoint starting at address @var{address} of kind @var{kind}.
42951 Each breakpoint and watchpoint packet @var{type} is documented
42954 @emph{Implementation notes: A remote target shall return an empty string
42955 for an unrecognized breakpoint or watchpoint packet @var{type}. A
42956 remote target shall support either both or neither of a given
42957 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
42958 avoid potential problems with duplicate packets, the operations should
42959 be implemented in an idempotent way.}
42961 @item z0,@var{addr},@var{kind}
42962 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
42963 @cindex @samp{z0} packet
42964 @cindex @samp{Z0} packet
42965 Insert (@samp{Z0}) or remove (@samp{z0}) a software breakpoint at address
42966 @var{addr} of type @var{kind}.
42968 A software breakpoint is implemented by replacing the instruction at
42969 @var{addr} with a software breakpoint or trap instruction. The
42970 @var{kind} is target-specific and typically indicates the size of the
42971 breakpoint in bytes that should be inserted. E.g., the @sc{arm} and
42972 @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
42973 architectures have additional meanings for @var{kind}
42974 (@pxref{Architecture-Specific Protocol Details}); if no
42975 architecture-specific value is being used, it should be @samp{0}.
42976 @var{kind} is hex-encoded. @var{cond_list} is an optional list of
42977 conditional expressions in bytecode form that should be evaluated on
42978 the target's side. These are the conditions that should be taken into
42979 consideration when deciding if the breakpoint trigger should be
42980 reported back to @value{GDBN}.
42982 See also the @samp{swbreak} stop reason (@pxref{swbreak stop reason})
42983 for how to best report a software breakpoint event to @value{GDBN}.
42985 The @var{cond_list} parameter is comprised of a series of expressions,
42986 concatenated without separators. Each expression has the following form:
42990 @item X @var{len},@var{expr}
42991 @var{len} is the length of the bytecode expression and @var{expr} is the
42992 actual conditional expression in bytecode form.
42996 The optional @var{cmd_list} parameter introduces commands that may be
42997 run on the target, rather than being reported back to @value{GDBN}.
42998 The parameter starts with a numeric flag @var{persist}; if the flag is
42999 nonzero, then the breakpoint may remain active and the commands
43000 continue to be run even when @value{GDBN} disconnects from the target.
43001 Following this flag is a series of expressions concatenated with no
43002 separators. Each expression has the following form:
43006 @item X @var{len},@var{expr}
43007 @var{len} is the length of the bytecode expression and @var{expr} is the
43008 actual commands expression in bytecode form.
43012 @emph{Implementation note: It is possible for a target to copy or move
43013 code that contains software breakpoints (e.g., when implementing
43014 overlays). The behavior of this packet, in the presence of such a
43015 target, is not defined.}
43027 @item z1,@var{addr},@var{kind}
43028 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
43029 @cindex @samp{z1} packet
43030 @cindex @samp{Z1} packet
43031 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
43032 address @var{addr}.
43034 A hardware breakpoint is implemented using a mechanism that is not
43035 dependent on being able to modify the target's memory. The
43036 @var{kind}, @var{cond_list}, and @var{cmd_list} arguments have the
43037 same meaning as in @samp{Z0} packets.
43039 @emph{Implementation note: A hardware breakpoint is not affected by code
43052 @item z2,@var{addr},@var{kind}
43053 @itemx Z2,@var{addr},@var{kind}
43054 @cindex @samp{z2} packet
43055 @cindex @samp{Z2} packet
43056 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
43057 The number of bytes to watch is specified by @var{kind}.
43069 @item z3,@var{addr},@var{kind}
43070 @itemx Z3,@var{addr},@var{kind}
43071 @cindex @samp{z3} packet
43072 @cindex @samp{Z3} packet
43073 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
43074 The number of bytes to watch is specified by @var{kind}.
43086 @item z4,@var{addr},@var{kind}
43087 @itemx Z4,@var{addr},@var{kind}
43088 @cindex @samp{z4} packet
43089 @cindex @samp{Z4} packet
43090 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
43091 The number of bytes to watch is specified by @var{kind}.
43105 @node Stop Reply Packets
43106 @section Stop Reply Packets
43107 @cindex stop reply packets
43109 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
43110 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
43111 receive any of the below as a reply. Except for @samp{?}
43112 and @samp{vStopped}, that reply is only returned
43113 when the target halts. In the below the exact meaning of @dfn{signal
43114 number} is defined by the header @file{include/gdb/signals.h} in the
43115 @value{GDBN} source code.
43117 In non-stop mode, the server will simply reply @samp{OK} to commands
43118 such as @samp{vCont}; any stop will be the subject of a future
43119 notification. @xref{Remote Non-Stop}.
43121 As in the description of request packets, we include spaces in the
43122 reply templates for clarity; these are not part of the reply packet's
43123 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
43129 The program received signal number @var{AA} (a two-digit hexadecimal
43130 number). This is equivalent to a @samp{T} response with no
43131 @var{n}:@var{r} pairs.
43133 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
43134 @cindex @samp{T} packet reply
43135 The program received signal number @var{AA} (a two-digit hexadecimal
43136 number). This is equivalent to an @samp{S} response, except that the
43137 @samp{@var{n}:@var{r}} pairs can carry values of important registers
43138 and other information directly in the stop reply packet, reducing
43139 round-trip latency. Single-step and breakpoint traps are reported
43140 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
43144 If @var{n} is a hexadecimal number, it is a register number, and the
43145 corresponding @var{r} gives that register's value. The data @var{r} is a
43146 series of bytes in target byte order, with each byte given by a
43147 two-digit hex number.
43150 If @var{n} is @samp{thread}, then @var{r} is the thread ID of
43151 the stopped thread, as specified in @ref{thread-id syntax}.
43154 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
43155 the core on which the stop event was detected.
43158 If @var{n} is a recognized @dfn{stop reason}, it describes a more
43159 specific event that stopped the target. The currently defined stop
43160 reasons are listed below. The @var{aa} should be @samp{05}, the trap
43161 signal. At most one stop reason should be present.
43164 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
43165 and go on to the next; this allows us to extend the protocol in the
43169 The currently defined stop reasons are:
43175 The packet indicates a watchpoint hit, and @var{r} is the data address, in
43178 @item syscall_entry
43179 @itemx syscall_return
43180 The packet indicates a syscall entry or return, and @var{r} is the
43181 syscall number, in hex.
43183 @cindex shared library events, remote reply
43185 The packet indicates that the loaded libraries have changed.
43186 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
43187 list of loaded libraries. The @var{r} part is ignored.
43189 @cindex replay log events, remote reply
43191 The packet indicates that the target cannot continue replaying
43192 logged execution events, because it has reached the end (or the
43193 beginning when executing backward) of the log. The value of @var{r}
43194 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
43195 for more information.
43198 @anchor{swbreak stop reason}
43199 The packet indicates a software breakpoint instruction was executed,
43200 irrespective of whether it was @value{GDBN} that planted the
43201 breakpoint or the breakpoint is hardcoded in the program. The @var{r}
43202 part must be left empty.
43204 On some architectures, such as x86, at the architecture level, when a
43205 breakpoint instruction executes the program counter points at the
43206 breakpoint address plus an offset. On such targets, the stub is
43207 responsible for adjusting the PC to point back at the breakpoint
43210 This packet should not be sent by default; older @value{GDBN} versions
43211 did not support it. @value{GDBN} requests it, by supplying an
43212 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
43213 remote stub must also supply the appropriate @samp{qSupported} feature
43214 indicating support.
43216 This packet is required for correct non-stop mode operation.
43219 The packet indicates the target stopped for a hardware breakpoint.
43220 The @var{r} part must be left empty.
43222 The same remarks about @samp{qSupported} and non-stop mode above
43225 @cindex fork events, remote reply
43227 The packet indicates that @code{fork} was called, and @var{r} is the
43228 thread ID of the new child process, as specified in @ref{thread-id
43229 syntax}. This packet is only applicable to targets that support fork
43232 This packet should not be sent by default; older @value{GDBN} versions
43233 did not support it. @value{GDBN} requests it, by supplying an
43234 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
43235 remote stub must also supply the appropriate @samp{qSupported} feature
43236 indicating support.
43238 @cindex vfork events, remote reply
43240 The packet indicates that @code{vfork} was called, and @var{r} is the
43241 thread ID of the new child process, as specified in @ref{thread-id
43242 syntax}. This packet is only applicable to targets that support vfork
43245 This packet should not be sent by default; older @value{GDBN} versions
43246 did not support it. @value{GDBN} requests it, by supplying an
43247 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
43248 remote stub must also supply the appropriate @samp{qSupported} feature
43249 indicating support.
43251 @cindex vforkdone events, remote reply
43253 The packet indicates that a child process created by a vfork
43254 has either called @code{exec} or terminated, so that the
43255 address spaces of the parent and child process are no longer
43256 shared. The @var{r} part is ignored. This packet is only
43257 applicable to targets that support vforkdone events.
43259 This packet should not be sent by default; older @value{GDBN} versions
43260 did not support it. @value{GDBN} requests it, by supplying an
43261 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
43262 remote stub must also supply the appropriate @samp{qSupported} feature
43263 indicating support.
43265 @cindex exec events, remote reply
43267 The packet indicates that @code{execve} was called, and @var{r}
43268 is the absolute pathname of the file that was executed, in hex.
43269 This packet is only applicable to targets that support exec events.
43271 This packet should not be sent by default; older @value{GDBN} versions
43272 did not support it. @value{GDBN} requests it, by supplying an
43273 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
43274 remote stub must also supply the appropriate @samp{qSupported} feature
43275 indicating support.
43277 @cindex thread create event, remote reply
43278 @anchor{thread create event}
43280 The packet indicates that the thread was just created. The new thread
43281 is stopped until @value{GDBN} sets it running with a resumption packet
43282 (@pxref{vCont packet}). This packet should not be sent by default;
43283 @value{GDBN} requests it with the @ref{QThreadEvents} packet. See
43284 also the @samp{w} (@pxref{thread exit event}) remote reply below. The
43285 @var{r} part is ignored.
43290 @itemx W @var{AA} ; process:@var{pid}
43291 The process exited, and @var{AA} is the exit status. This is only
43292 applicable to certain targets.
43294 The second form of the response, including the process ID of the
43295 exited process, can be used only when @value{GDBN} has reported
43296 support for multiprocess protocol extensions; see @ref{multiprocess
43297 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
43301 @itemx X @var{AA} ; process:@var{pid}
43302 The process terminated with signal @var{AA}.
43304 The second form of the response, including the process ID of the
43305 terminated process, can be used only when @value{GDBN} has reported
43306 support for multiprocess protocol extensions; see @ref{multiprocess
43307 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
43310 @anchor{thread exit event}
43311 @cindex thread exit event, remote reply
43312 @item w @var{AA} ; @var{tid}
43314 The thread exited, and @var{AA} is the exit status. This response
43315 should not be sent by default; @value{GDBN} requests it with the
43316 @ref{QThreadEvents} packet. See also @ref{thread create event} above.
43317 @var{AA} is formatted as a big-endian hex string.
43320 There are no resumed threads left in the target. In other words, even
43321 though the process is alive, the last resumed thread has exited. For
43322 example, say the target process has two threads: thread 1 and thread
43323 2. The client leaves thread 1 stopped, and resumes thread 2, which
43324 subsequently exits. At this point, even though the process is still
43325 alive, and thus no @samp{W} stop reply is sent, no thread is actually
43326 executing either. The @samp{N} stop reply thus informs the client
43327 that it can stop waiting for stop replies. This packet should not be
43328 sent by default; older @value{GDBN} versions did not support it.
43329 @value{GDBN} requests it, by supplying an appropriate
43330 @samp{qSupported} feature (@pxref{qSupported}). The remote stub must
43331 also supply the appropriate @samp{qSupported} feature indicating
43334 @item O @var{XX}@dots{}
43335 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
43336 written as the program's console output. This can happen at any time
43337 while the program is running and the debugger should continue to wait
43338 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
43340 @item F @var{call-id},@var{parameter}@dots{}
43341 @var{call-id} is the identifier which says which host system call should
43342 be called. This is just the name of the function. Translation into the
43343 correct system call is only applicable as it's defined in @value{GDBN}.
43344 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
43347 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
43348 this very system call.
43350 The target replies with this packet when it expects @value{GDBN} to
43351 call a host system call on behalf of the target. @value{GDBN} replies
43352 with an appropriate @samp{F} packet and keeps up waiting for the next
43353 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
43354 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
43355 Protocol Extension}, for more details.
43359 @node General Query Packets
43360 @section General Query Packets
43361 @cindex remote query requests
43363 Packets starting with @samp{q} are @dfn{general query packets};
43364 packets starting with @samp{Q} are @dfn{general set packets}. General
43365 query and set packets are a semi-unified form for retrieving and
43366 sending information to and from the stub.
43368 The initial letter of a query or set packet is followed by a name
43369 indicating what sort of thing the packet applies to. For example,
43370 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
43371 definitions with the stub. These packet names follow some
43376 The name must not contain commas, colons or semicolons.
43378 Most @value{GDBN} query and set packets have a leading upper case
43381 The names of custom vendor packets should use a company prefix, in
43382 lower case, followed by a period. For example, packets designed at
43383 the Acme Corporation might begin with @samp{qacme.foo} (for querying
43384 foos) or @samp{Qacme.bar} (for setting bars).
43387 The name of a query or set packet should be separated from any
43388 parameters by a @samp{:}; the parameters themselves should be
43389 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
43390 full packet name, and check for a separator or the end of the packet,
43391 in case two packet names share a common prefix. New packets should not begin
43392 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
43393 packets predate these conventions, and have arguments without any terminator
43394 for the packet name; we suspect they are in widespread use in places that
43395 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
43396 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
43399 Like the descriptions of the other packets, each description here
43400 has a template showing the packet's overall syntax, followed by an
43401 explanation of the packet's meaning. We include spaces in some of the
43402 templates for clarity; these are not part of the packet's syntax. No
43403 @value{GDBN} packet uses spaces to separate its components.
43405 Here are the currently defined query and set packets:
43411 Turn on or off the agent as a helper to perform some debugging operations
43412 delegated from @value{GDBN} (@pxref{Control Agent}).
43414 @item QAllow:@var{op}:@var{val}@dots{}
43415 @cindex @samp{QAllow} packet
43416 Specify which operations @value{GDBN} expects to request of the
43417 target, as a semicolon-separated list of operation name and value
43418 pairs. Possible values for @var{op} include @samp{WriteReg},
43419 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
43420 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
43421 indicating that @value{GDBN} will not request the operation, or 1,
43422 indicating that it may. (The target can then use this to set up its
43423 own internals optimally, for instance if the debugger never expects to
43424 insert breakpoints, it may not need to install its own trap handler.)
43427 @cindex current thread, remote request
43428 @cindex @samp{qC} packet
43429 Return the current thread ID.
43433 @item QC @var{thread-id}
43434 Where @var{thread-id} is a thread ID as documented in
43435 @ref{thread-id syntax}.
43436 @item @r{(anything else)}
43437 Any other reply implies the old thread ID.
43440 @item qCRC:@var{addr},@var{length}
43441 @cindex CRC of memory block, remote request
43442 @cindex @samp{qCRC} packet
43443 @anchor{qCRC packet}
43444 Compute the CRC checksum of a block of memory using CRC-32 defined in
43445 IEEE 802.3. The CRC is computed byte at a time, taking the most
43446 significant bit of each byte first. The initial pattern code
43447 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
43449 @emph{Note:} This is the same CRC used in validating separate debug
43450 files (@pxref{Separate Debug Files, , Debugging Information in Separate
43451 Files}). However the algorithm is slightly different. When validating
43452 separate debug files, the CRC is computed taking the @emph{least}
43453 significant bit of each byte first, and the final result is inverted to
43454 detect trailing zeros.
43459 An error (such as memory fault)
43460 @item C @var{crc32}
43461 The specified memory region's checksum is @var{crc32}.
43464 @item QDisableRandomization:@var{value}
43465 @cindex disable address space randomization, remote request
43466 @cindex @samp{QDisableRandomization} packet
43467 Some target operating systems will randomize the virtual address space
43468 of the inferior process as a security feature, but provide a feature
43469 to disable such randomization, e.g.@: to allow for a more deterministic
43470 debugging experience. On such systems, this packet with a @var{value}
43471 of 1 directs the target to disable address space randomization for
43472 processes subsequently started via @samp{vRun} packets, while a packet
43473 with a @var{value} of 0 tells the target to enable address space
43476 This packet is only available in extended mode (@pxref{extended mode}).
43481 The request succeeded.
43484 An error occurred. The error number @var{nn} is given as hex digits.
43487 An empty reply indicates that @samp{QDisableRandomization} is not supported
43491 This packet is not probed by default; the remote stub must request it,
43492 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43493 This should only be done on targets that actually support disabling
43494 address space randomization.
43496 @item QStartupWithShell:@var{value}
43497 @cindex startup with shell, remote request
43498 @cindex @samp{QStartupWithShell} packet
43499 On UNIX-like targets, it is possible to start the inferior using a
43500 shell program. This is the default behavior on both @value{GDBN} and
43501 @command{gdbserver} (@pxref{set startup-with-shell}). This packet is
43502 used to inform @command{gdbserver} whether it should start the
43503 inferior using a shell or not.
43505 If @var{value} is @samp{0}, @command{gdbserver} will not use a shell
43506 to start the inferior. If @var{value} is @samp{1},
43507 @command{gdbserver} will use a shell to start the inferior. All other
43508 values are considered an error.
43510 This packet is only available in extended mode (@pxref{extended
43516 The request succeeded.
43519 An error occurred. The error number @var{nn} is given as hex digits.
43522 This packet is not probed by default; the remote stub must request it,
43523 by supplying an appropriate @samp{qSupported} response
43524 (@pxref{qSupported}). This should only be done on targets that
43525 actually support starting the inferior using a shell.
43527 Use of this packet is controlled by the @code{set startup-with-shell}
43528 command; @pxref{set startup-with-shell}.
43530 @item QEnvironmentHexEncoded:@var{hex-value}
43531 @anchor{QEnvironmentHexEncoded}
43532 @cindex set environment variable, remote request
43533 @cindex @samp{QEnvironmentHexEncoded} packet
43534 On UNIX-like targets, it is possible to set environment variables that
43535 will be passed to the inferior during the startup process. This
43536 packet is used to inform @command{gdbserver} of an environment
43537 variable that has been defined by the user on @value{GDBN} (@pxref{set
43540 The packet is composed by @var{hex-value}, an hex encoded
43541 representation of the @var{name=value} format representing an
43542 environment variable. The name of the environment variable is
43543 represented by @var{name}, and the value to be assigned to the
43544 environment variable is represented by @var{value}. If the variable
43545 has no value (i.e., the value is @code{null}), then @var{value} will
43548 This packet is only available in extended mode (@pxref{extended
43554 The request succeeded.
43557 This packet is not probed by default; the remote stub must request it,
43558 by supplying an appropriate @samp{qSupported} response
43559 (@pxref{qSupported}). This should only be done on targets that
43560 actually support passing environment variables to the starting
43563 This packet is related to the @code{set environment} command;
43564 @pxref{set environment}.
43566 @item QEnvironmentUnset:@var{hex-value}
43567 @anchor{QEnvironmentUnset}
43568 @cindex unset environment variable, remote request
43569 @cindex @samp{QEnvironmentUnset} packet
43570 On UNIX-like targets, it is possible to unset environment variables
43571 before starting the inferior in the remote target. This packet is
43572 used to inform @command{gdbserver} of an environment variable that has
43573 been unset by the user on @value{GDBN} (@pxref{unset environment}).
43575 The packet is composed by @var{hex-value}, an hex encoded
43576 representation of the name of the environment variable to be unset.
43578 This packet is only available in extended mode (@pxref{extended
43584 The request succeeded.
43587 This packet is not probed by default; the remote stub must request it,
43588 by supplying an appropriate @samp{qSupported} response
43589 (@pxref{qSupported}). This should only be done on targets that
43590 actually support passing environment variables to the starting
43593 This packet is related to the @code{unset environment} command;
43594 @pxref{unset environment}.
43596 @item QEnvironmentReset
43597 @anchor{QEnvironmentReset}
43598 @cindex reset environment, remote request
43599 @cindex @samp{QEnvironmentReset} packet
43600 On UNIX-like targets, this packet is used to reset the state of
43601 environment variables in the remote target before starting the
43602 inferior. In this context, reset means unsetting all environment
43603 variables that were previously set by the user (i.e., were not
43604 initially present in the environment). It is sent to
43605 @command{gdbserver} before the @samp{QEnvironmentHexEncoded}
43606 (@pxref{QEnvironmentHexEncoded}) and the @samp{QEnvironmentUnset}
43607 (@pxref{QEnvironmentUnset}) packets.
43609 This packet is only available in extended mode (@pxref{extended
43615 The request succeeded.
43618 This packet is not probed by default; the remote stub must request it,
43619 by supplying an appropriate @samp{qSupported} response
43620 (@pxref{qSupported}). This should only be done on targets that
43621 actually support passing environment variables to the starting
43624 @item QSetWorkingDir:@r{[}@var{directory}@r{]}
43625 @anchor{QSetWorkingDir packet}
43626 @cindex set working directory, remote request
43627 @cindex @samp{QSetWorkingDir} packet
43628 This packet is used to inform the remote server of the intended
43629 current working directory for programs that are going to be executed.
43631 The packet is composed by @var{directory}, an hex encoded
43632 representation of the directory that the remote inferior will use as
43633 its current working directory. If @var{directory} is an empty string,
43634 the remote server should reset the inferior's current working
43635 directory to its original, empty value.
43637 This packet is only available in extended mode (@pxref{extended
43643 The request succeeded.
43647 @itemx qsThreadInfo
43648 @cindex list active threads, remote request
43649 @cindex @samp{qfThreadInfo} packet
43650 @cindex @samp{qsThreadInfo} packet
43651 Obtain a list of all active thread IDs from the target (OS). Since there
43652 may be too many active threads to fit into one reply packet, this query
43653 works iteratively: it may require more than one query/reply sequence to
43654 obtain the entire list of threads. The first query of the sequence will
43655 be the @samp{qfThreadInfo} query; subsequent queries in the
43656 sequence will be the @samp{qsThreadInfo} query.
43658 NOTE: This packet replaces the @samp{qL} query (see below).
43662 @item m @var{thread-id}
43664 @item m @var{thread-id},@var{thread-id}@dots{}
43665 a comma-separated list of thread IDs
43667 (lower case letter @samp{L}) denotes end of list.
43670 In response to each query, the target will reply with a list of one or
43671 more thread IDs, separated by commas.
43672 @value{GDBN} will respond to each reply with a request for more thread
43673 ids (using the @samp{qs} form of the query), until the target responds
43674 with @samp{l} (lower-case ell, for @dfn{last}).
43675 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
43678 @emph{Note: @value{GDBN} will send the @code{qfThreadInfo} query during the
43679 initial connection with the remote target, and the very first thread ID
43680 mentioned in the reply will be stopped by @value{GDBN} in a subsequent
43681 message. Therefore, the stub should ensure that the first thread ID in
43682 the @code{qfThreadInfo} reply is suitable for being stopped by @value{GDBN}.}
43684 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
43685 @cindex get thread-local storage address, remote request
43686 @cindex @samp{qGetTLSAddr} packet
43687 Fetch the address associated with thread local storage specified
43688 by @var{thread-id}, @var{offset}, and @var{lm}.
43690 @var{thread-id} is the thread ID associated with the
43691 thread for which to fetch the TLS address. @xref{thread-id syntax}.
43693 @var{offset} is the (big endian, hex encoded) offset associated with the
43694 thread local variable. (This offset is obtained from the debug
43695 information associated with the variable.)
43697 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
43698 load module associated with the thread local storage. For example,
43699 a @sc{gnu}/Linux system will pass the link map address of the shared
43700 object associated with the thread local storage under consideration.
43701 Other operating environments may choose to represent the load module
43702 differently, so the precise meaning of this parameter will vary.
43706 @item @var{XX}@dots{}
43707 Hex encoded (big endian) bytes representing the address of the thread
43708 local storage requested.
43711 An error occurred. The error number @var{nn} is given as hex digits.
43714 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
43717 @item qGetTIBAddr:@var{thread-id}
43718 @cindex get thread information block address
43719 @cindex @samp{qGetTIBAddr} packet
43720 Fetch address of the Windows OS specific Thread Information Block.
43722 @var{thread-id} is the thread ID associated with the thread.
43726 @item @var{XX}@dots{}
43727 Hex encoded (big endian) bytes representing the linear address of the
43728 thread information block.
43731 An error occurred. This means that either the thread was not found, or the
43732 address could not be retrieved.
43735 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
43738 @item qL @var{startflag} @var{threadcount} @var{nextthread}
43739 Obtain thread information from RTOS. Where: @var{startflag} (one hex
43740 digit) is one to indicate the first query and zero to indicate a
43741 subsequent query; @var{threadcount} (two hex digits) is the maximum
43742 number of threads the response packet can contain; and @var{nextthread}
43743 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
43744 returned in the response as @var{argthread}.
43746 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
43750 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
43751 Where: @var{count} (two hex digits) is the number of threads being
43752 returned; @var{done} (one hex digit) is zero to indicate more threads
43753 and one indicates no further threads; @var{argthreadid} (eight hex
43754 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
43755 is a sequence of thread IDs, @var{threadid} (eight hex
43756 digits), from the target. See @code{remote.c:parse_threadlist_response()}.
43759 @item qMemTags:@var{start address},@var{length}:@var{type}
43761 @cindex fetch memory tags
43762 @cindex @samp{qMemTags} packet
43763 Fetch memory tags of type @var{type} from the address range
43764 @w{@r{[}@var{start address}, @var{start address} + @var{length}@r{)}}. The
43765 target is responsible for calculating how many tags will be returned, as this
43766 is architecture-specific.
43768 @var{start address} is the starting address of the memory range.
43770 @var{length} is the length, in bytes, of the memory range.
43772 @var{type} is the type of tag the request wants to fetch. The type is a signed
43777 @item @var{mxx}@dots{}
43778 Hex encoded sequence of uninterpreted bytes, @var{xx}@dots{}, representing the
43779 tags found in the requested memory range.
43782 An error occurred. This means that fetching of memory tags failed for some
43786 An empty reply indicates that @samp{qMemTags} is not supported by the stub,
43787 although this should not happen given @value{GDBN} will only send this packet
43788 if the stub has advertised support for memory tagging via @samp{qSupported}.
43791 @item QMemTags:@var{start address},@var{length}:@var{type}:@var{tag bytes}
43793 @cindex store memory tags
43794 @cindex @samp{QMemTags} packet
43795 Store memory tags of type @var{type} to the address range
43796 @w{@r{[}@var{start address}, @var{start address} + @var{length}@r{)}}. The
43797 target is responsible for interpreting the type, the tag bytes and modifying
43798 the memory tag granules accordingly, given this is architecture-specific.
43800 The interpretation of how many tags (@var{nt}) should be written to how many
43801 memory tag granules (@var{ng}) is also architecture-specific. The behavior is
43802 implementation-specific, but the following is suggested.
43804 If the number of memory tags, @var{nt}, is greater than or equal to the
43805 number of memory tag granules, @var{ng}, only @var{ng} tags will be
43808 If @var{nt} is less than @var{ng}, the behavior is that of a fill operation,
43809 and the tag bytes will be used as a pattern that will get repeated until
43810 @var{ng} tags are stored.
43812 @var{start address} is the starting address of the memory range. The address
43813 does not have any restriction on alignment or size.
43815 @var{length} is the length, in bytes, of the memory range.
43817 @var{type} is the type of tag the request wants to fetch. The type is a signed
43820 @var{tag bytes} is a sequence of hex encoded uninterpreted bytes which will be
43821 interpreted by the target. Each pair of hex digits is interpreted as a
43827 The request was successful and the memory tag granules were modified
43831 An error occurred. This means that modifying the memory tag granules failed
43835 An empty reply indicates that @samp{QMemTags} is not supported by the stub,
43836 although this should not happen given @value{GDBN} will only send this packet
43837 if the stub has advertised support for memory tagging via @samp{qSupported}.
43841 @cindex section offsets, remote request
43842 @cindex @samp{qOffsets} packet
43843 Get section offsets that the target used when relocating the downloaded
43848 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
43849 Relocate the @code{Text} section by @var{xxx} from its original address.
43850 Relocate the @code{Data} section by @var{yyy} from its original address.
43851 If the object file format provides segment information (e.g.@: @sc{elf}
43852 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
43853 segments by the supplied offsets.
43855 @emph{Note: while a @code{Bss} offset may be included in the response,
43856 @value{GDBN} ignores this and instead applies the @code{Data} offset
43857 to the @code{Bss} section.}
43859 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
43860 Relocate the first segment of the object file, which conventionally
43861 contains program code, to a starting address of @var{xxx}. If
43862 @samp{DataSeg} is specified, relocate the second segment, which
43863 conventionally contains modifiable data, to a starting address of
43864 @var{yyy}. @value{GDBN} will report an error if the object file
43865 does not contain segment information, or does not contain at least
43866 as many segments as mentioned in the reply. Extra segments are
43867 kept at fixed offsets relative to the last relocated segment.
43870 @item qP @var{mode} @var{thread-id}
43871 @cindex thread information, remote request
43872 @cindex @samp{qP} packet
43873 Returns information on @var{thread-id}. Where: @var{mode} is a hex
43874 encoded 32 bit mode; @var{thread-id} is a thread ID
43875 (@pxref{thread-id syntax}).
43877 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
43880 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
43884 @cindex non-stop mode, remote request
43885 @cindex @samp{QNonStop} packet
43887 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
43888 @xref{Remote Non-Stop}, for more information.
43893 The request succeeded.
43896 An error occurred. The error number @var{nn} is given as hex digits.
43899 An empty reply indicates that @samp{QNonStop} is not supported by
43903 This packet is not probed by default; the remote stub must request it,
43904 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43905 Use of this packet is controlled by the @code{set non-stop} command;
43906 @pxref{Non-Stop Mode}.
43908 @item QCatchSyscalls:1 @r{[};@var{sysno}@r{]}@dots{}
43909 @itemx QCatchSyscalls:0
43910 @cindex catch syscalls from inferior, remote request
43911 @cindex @samp{QCatchSyscalls} packet
43912 @anchor{QCatchSyscalls}
43913 Enable (@samp{QCatchSyscalls:1}) or disable (@samp{QCatchSyscalls:0})
43914 catching syscalls from the inferior process.
43916 For @samp{QCatchSyscalls:1}, each listed syscall @var{sysno} (encoded
43917 in hex) should be reported to @value{GDBN}. If no syscall @var{sysno}
43918 is listed, every system call should be reported.
43920 Note that if a syscall not in the list is reported, @value{GDBN} will
43921 still filter the event according to its own list from all corresponding
43922 @code{catch syscall} commands. However, it is more efficient to only
43923 report the requested syscalls.
43925 Multiple @samp{QCatchSyscalls:1} packets do not combine; any earlier
43926 @samp{QCatchSyscalls:1} list is completely replaced by the new list.
43928 If the inferior process execs, the state of @samp{QCatchSyscalls} is
43929 kept for the new process too. On targets where exec may affect syscall
43930 numbers, for example with exec between 32 and 64-bit processes, the
43931 client should send a new packet with the new syscall list.
43936 The request succeeded.
43939 An error occurred. @var{nn} are hex digits.
43942 An empty reply indicates that @samp{QCatchSyscalls} is not supported by
43946 Use of this packet is controlled by the @code{set remote catch-syscalls}
43947 command (@pxref{Remote Configuration, set remote catch-syscalls}).
43948 This packet is not probed by default; the remote stub must request it,
43949 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43951 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
43952 @cindex pass signals to inferior, remote request
43953 @cindex @samp{QPassSignals} packet
43954 @anchor{QPassSignals}
43955 Each listed @var{signal} should be passed directly to the inferior process.
43956 Signals are numbered identically to continue packets and stop replies
43957 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
43958 strictly greater than the previous item. These signals do not need to stop
43959 the inferior, or be reported to @value{GDBN}. All other signals should be
43960 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
43961 combine; any earlier @samp{QPassSignals} list is completely replaced by the
43962 new list. This packet improves performance when using @samp{handle
43963 @var{signal} nostop noprint pass}.
43968 The request succeeded.
43971 An error occurred. The error number @var{nn} is given as hex digits.
43974 An empty reply indicates that @samp{QPassSignals} is not supported by
43978 Use of this packet is controlled by the @code{set remote pass-signals}
43979 command (@pxref{Remote Configuration, set remote pass-signals}).
43980 This packet is not probed by default; the remote stub must request it,
43981 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43983 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
43984 @cindex signals the inferior may see, remote request
43985 @cindex @samp{QProgramSignals} packet
43986 @anchor{QProgramSignals}
43987 Each listed @var{signal} may be delivered to the inferior process.
43988 Others should be silently discarded.
43990 In some cases, the remote stub may need to decide whether to deliver a
43991 signal to the program or not without @value{GDBN} involvement. One
43992 example of that is while detaching --- the program's threads may have
43993 stopped for signals that haven't yet had a chance of being reported to
43994 @value{GDBN}, and so the remote stub can use the signal list specified
43995 by this packet to know whether to deliver or ignore those pending
43998 This does not influence whether to deliver a signal as requested by a
43999 resumption packet (@pxref{vCont packet}).
44001 Signals are numbered identically to continue packets and stop replies
44002 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
44003 strictly greater than the previous item. Multiple
44004 @samp{QProgramSignals} packets do not combine; any earlier
44005 @samp{QProgramSignals} list is completely replaced by the new list.
44010 The request succeeded.
44013 An error occurred. The error number @var{nn} is given as hex digits.
44016 An empty reply indicates that @samp{QProgramSignals} is not supported
44020 Use of this packet is controlled by the @code{set remote program-signals}
44021 command (@pxref{Remote Configuration, set remote program-signals}).
44022 This packet is not probed by default; the remote stub must request it,
44023 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
44025 @anchor{QThreadEvents}
44026 @item QThreadEvents:1
44027 @itemx QThreadEvents:0
44028 @cindex thread create/exit events, remote request
44029 @cindex @samp{QThreadEvents} packet
44031 Enable (@samp{QThreadEvents:1}) or disable (@samp{QThreadEvents:0})
44032 reporting of thread create and exit events. @xref{thread create
44033 event}, for the reply specifications. For example, this is used in
44034 non-stop mode when @value{GDBN} stops a set of threads and
44035 synchronously waits for the their corresponding stop replies. Without
44036 exit events, if one of the threads exits, @value{GDBN} would hang
44037 forever not knowing that it should no longer expect a stop for that
44038 same thread. @value{GDBN} does not enable this feature unless the
44039 stub reports that it supports it by including @samp{QThreadEvents+} in
44040 its @samp{qSupported} reply.
44045 The request succeeded.
44048 An error occurred. The error number @var{nn} is given as hex digits.
44051 An empty reply indicates that @samp{QThreadEvents} is not supported by
44055 Use of this packet is controlled by the @code{set remote thread-events}
44056 command (@pxref{Remote Configuration, set remote thread-events}).
44058 @item qRcmd,@var{command}
44059 @cindex execute remote command, remote request
44060 @cindex @samp{qRcmd} packet
44061 @var{command} (hex encoded) is passed to the local interpreter for
44062 execution. Invalid commands should be reported using the output
44063 string. Before the final result packet, the target may also respond
44064 with a number of intermediate @samp{O@var{output}} console output
44065 packets. @emph{Implementors should note that providing access to a
44066 stubs's interpreter may have security implications}.
44071 A command response with no output.
44073 A command response with the hex encoded output string @var{OUTPUT}.
44075 Indicate a badly formed request. The error number @var{NN} is given as
44078 An empty reply indicates that @samp{qRcmd} is not recognized.
44081 (Note that the @code{qRcmd} packet's name is separated from the
44082 command by a @samp{,}, not a @samp{:}, contrary to the naming
44083 conventions above. Please don't use this packet as a model for new
44086 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
44087 @cindex searching memory, in remote debugging
44089 @cindex @samp{qSearch:memory} packet
44091 @cindex @samp{qSearch memory} packet
44092 @anchor{qSearch memory}
44093 Search @var{length} bytes at @var{address} for @var{search-pattern}.
44094 Both @var{address} and @var{length} are encoded in hex;
44095 @var{search-pattern} is a sequence of bytes, also hex encoded.
44100 The pattern was not found.
44102 The pattern was found at @var{address}.
44104 A badly formed request or an error was encountered while searching memory.
44106 An empty reply indicates that @samp{qSearch:memory} is not recognized.
44109 @item QStartNoAckMode
44110 @cindex @samp{QStartNoAckMode} packet
44111 @anchor{QStartNoAckMode}
44112 Request that the remote stub disable the normal @samp{+}/@samp{-}
44113 protocol acknowledgments (@pxref{Packet Acknowledgment}).
44118 The stub has switched to no-acknowledgment mode.
44119 @value{GDBN} acknowledges this response,
44120 but neither the stub nor @value{GDBN} shall send or expect further
44121 @samp{+}/@samp{-} acknowledgments in the current connection.
44123 An empty reply indicates that the stub does not support no-acknowledgment mode.
44126 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
44127 @cindex supported packets, remote query
44128 @cindex features of the remote protocol
44129 @cindex @samp{qSupported} packet
44130 @anchor{qSupported}
44131 Tell the remote stub about features supported by @value{GDBN}, and
44132 query the stub for features it supports. This packet allows
44133 @value{GDBN} and the remote stub to take advantage of each others'
44134 features. @samp{qSupported} also consolidates multiple feature probes
44135 at startup, to improve @value{GDBN} performance---a single larger
44136 packet performs better than multiple smaller probe packets on
44137 high-latency links. Some features may enable behavior which must not
44138 be on by default, e.g.@: because it would confuse older clients or
44139 stubs. Other features may describe packets which could be
44140 automatically probed for, but are not. These features must be
44141 reported before @value{GDBN} will use them. This ``default
44142 unsupported'' behavior is not appropriate for all packets, but it
44143 helps to keep the initial connection time under control with new
44144 versions of @value{GDBN} which support increasing numbers of packets.
44148 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
44149 The stub supports or does not support each returned @var{stubfeature},
44150 depending on the form of each @var{stubfeature} (see below for the
44153 An empty reply indicates that @samp{qSupported} is not recognized,
44154 or that no features needed to be reported to @value{GDBN}.
44157 The allowed forms for each feature (either a @var{gdbfeature} in the
44158 @samp{qSupported} packet, or a @var{stubfeature} in the response)
44162 @item @var{name}=@var{value}
44163 The remote protocol feature @var{name} is supported, and associated
44164 with the specified @var{value}. The format of @var{value} depends
44165 on the feature, but it must not include a semicolon.
44167 The remote protocol feature @var{name} is supported, and does not
44168 need an associated value.
44170 The remote protocol feature @var{name} is not supported.
44172 The remote protocol feature @var{name} may be supported, and
44173 @value{GDBN} should auto-detect support in some other way when it is
44174 needed. This form will not be used for @var{gdbfeature} notifications,
44175 but may be used for @var{stubfeature} responses.
44178 Whenever the stub receives a @samp{qSupported} request, the
44179 supplied set of @value{GDBN} features should override any previous
44180 request. This allows @value{GDBN} to put the stub in a known
44181 state, even if the stub had previously been communicating with
44182 a different version of @value{GDBN}.
44184 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
44189 This feature indicates whether @value{GDBN} supports multiprocess
44190 extensions to the remote protocol. @value{GDBN} does not use such
44191 extensions unless the stub also reports that it supports them by
44192 including @samp{multiprocess+} in its @samp{qSupported} reply.
44193 @xref{multiprocess extensions}, for details.
44196 This feature indicates that @value{GDBN} supports the XML target
44197 description. If the stub sees @samp{xmlRegisters=} with target
44198 specific strings separated by a comma, it will report register
44202 This feature indicates whether @value{GDBN} supports the
44203 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
44204 instruction reply packet}).
44207 This feature indicates whether @value{GDBN} supports the swbreak stop
44208 reason in stop replies. @xref{swbreak stop reason}, for details.
44211 This feature indicates whether @value{GDBN} supports the hwbreak stop
44212 reason in stop replies. @xref{swbreak stop reason}, for details.
44215 This feature indicates whether @value{GDBN} supports fork event
44216 extensions to the remote protocol. @value{GDBN} does not use such
44217 extensions unless the stub also reports that it supports them by
44218 including @samp{fork-events+} in its @samp{qSupported} reply.
44221 This feature indicates whether @value{GDBN} supports vfork event
44222 extensions to the remote protocol. @value{GDBN} does not use such
44223 extensions unless the stub also reports that it supports them by
44224 including @samp{vfork-events+} in its @samp{qSupported} reply.
44227 This feature indicates whether @value{GDBN} supports exec event
44228 extensions to the remote protocol. @value{GDBN} does not use such
44229 extensions unless the stub also reports that it supports them by
44230 including @samp{exec-events+} in its @samp{qSupported} reply.
44232 @item vContSupported
44233 This feature indicates whether @value{GDBN} wants to know the
44234 supported actions in the reply to @samp{vCont?} packet.
44237 Stubs should ignore any unknown values for
44238 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
44239 packet supports receiving packets of unlimited length (earlier
44240 versions of @value{GDBN} may reject overly long responses). Additional values
44241 for @var{gdbfeature} may be defined in the future to let the stub take
44242 advantage of new features in @value{GDBN}, e.g.@: incompatible
44243 improvements in the remote protocol---the @samp{multiprocess} feature is
44244 an example of such a feature. The stub's reply should be independent
44245 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
44246 describes all the features it supports, and then the stub replies with
44247 all the features it supports.
44249 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
44250 responses, as long as each response uses one of the standard forms.
44252 Some features are flags. A stub which supports a flag feature
44253 should respond with a @samp{+} form response. Other features
44254 require values, and the stub should respond with an @samp{=}
44257 Each feature has a default value, which @value{GDBN} will use if
44258 @samp{qSupported} is not available or if the feature is not mentioned
44259 in the @samp{qSupported} response. The default values are fixed; a
44260 stub is free to omit any feature responses that match the defaults.
44262 Not all features can be probed, but for those which can, the probing
44263 mechanism is useful: in some cases, a stub's internal
44264 architecture may not allow the protocol layer to know some information
44265 about the underlying target in advance. This is especially common in
44266 stubs which may be configured for multiple targets.
44268 These are the currently defined stub features and their properties:
44270 @multitable @columnfractions 0.35 0.2 0.12 0.2
44271 @c NOTE: The first row should be @headitem, but we do not yet require
44272 @c a new enough version of Texinfo (4.7) to use @headitem.
44274 @tab Value Required
44278 @item @samp{PacketSize}
44283 @item @samp{qXfer:auxv:read}
44288 @item @samp{qXfer:btrace:read}
44293 @item @samp{qXfer:btrace-conf:read}
44298 @item @samp{qXfer:exec-file:read}
44303 @item @samp{qXfer:features:read}
44308 @item @samp{qXfer:libraries:read}
44313 @item @samp{qXfer:libraries-svr4:read}
44318 @item @samp{augmented-libraries-svr4-read}
44323 @item @samp{qXfer:memory-map:read}
44328 @item @samp{qXfer:sdata:read}
44333 @item @samp{qXfer:siginfo:read}
44338 @item @samp{qXfer:siginfo:write}
44343 @item @samp{qXfer:threads:read}
44348 @item @samp{qXfer:traceframe-info:read}
44353 @item @samp{qXfer:uib:read}
44358 @item @samp{qXfer:fdpic:read}
44363 @item @samp{Qbtrace:off}
44368 @item @samp{Qbtrace:bts}
44373 @item @samp{Qbtrace:pt}
44378 @item @samp{Qbtrace-conf:bts:size}
44383 @item @samp{Qbtrace-conf:pt:size}
44388 @item @samp{QNonStop}
44393 @item @samp{QCatchSyscalls}
44398 @item @samp{QPassSignals}
44403 @item @samp{QStartNoAckMode}
44408 @item @samp{multiprocess}
44413 @item @samp{ConditionalBreakpoints}
44418 @item @samp{ConditionalTracepoints}
44423 @item @samp{ReverseContinue}
44428 @item @samp{ReverseStep}
44433 @item @samp{TracepointSource}
44438 @item @samp{QAgent}
44443 @item @samp{QAllow}
44448 @item @samp{QDisableRandomization}
44453 @item @samp{EnableDisableTracepoints}
44458 @item @samp{QTBuffer:size}
44463 @item @samp{tracenz}
44468 @item @samp{BreakpointCommands}
44473 @item @samp{swbreak}
44478 @item @samp{hwbreak}
44483 @item @samp{fork-events}
44488 @item @samp{vfork-events}
44493 @item @samp{exec-events}
44498 @item @samp{QThreadEvents}
44503 @item @samp{no-resumed}
44508 @item @samp{memory-tagging}
44515 These are the currently defined stub features, in more detail:
44518 @cindex packet size, remote protocol
44519 @item PacketSize=@var{bytes}
44520 The remote stub can accept packets up to at least @var{bytes} in
44521 length. @value{GDBN} will send packets up to this size for bulk
44522 transfers, and will never send larger packets. This is a limit on the
44523 data characters in the packet, including the frame and checksum.
44524 There is no trailing NUL byte in a remote protocol packet; if the stub
44525 stores packets in a NUL-terminated format, it should allow an extra
44526 byte in its buffer for the NUL. If this stub feature is not supported,
44527 @value{GDBN} guesses based on the size of the @samp{g} packet response.
44529 @item qXfer:auxv:read
44530 The remote stub understands the @samp{qXfer:auxv:read} packet
44531 (@pxref{qXfer auxiliary vector read}).
44533 @item qXfer:btrace:read
44534 The remote stub understands the @samp{qXfer:btrace:read}
44535 packet (@pxref{qXfer btrace read}).
44537 @item qXfer:btrace-conf:read
44538 The remote stub understands the @samp{qXfer:btrace-conf:read}
44539 packet (@pxref{qXfer btrace-conf read}).
44541 @item qXfer:exec-file:read
44542 The remote stub understands the @samp{qXfer:exec-file:read} packet
44543 (@pxref{qXfer executable filename read}).
44545 @item qXfer:features:read
44546 The remote stub understands the @samp{qXfer:features:read} packet
44547 (@pxref{qXfer target description read}).
44549 @item qXfer:libraries:read
44550 The remote stub understands the @samp{qXfer:libraries:read} packet
44551 (@pxref{qXfer library list read}).
44553 @item qXfer:libraries-svr4:read
44554 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
44555 (@pxref{qXfer svr4 library list read}).
44557 @item augmented-libraries-svr4-read
44558 The remote stub understands the augmented form of the
44559 @samp{qXfer:libraries-svr4:read} packet
44560 (@pxref{qXfer svr4 library list read}).
44562 @item qXfer:memory-map:read
44563 The remote stub understands the @samp{qXfer:memory-map:read} packet
44564 (@pxref{qXfer memory map read}).
44566 @item qXfer:sdata:read
44567 The remote stub understands the @samp{qXfer:sdata:read} packet
44568 (@pxref{qXfer sdata read}).
44570 @item qXfer:siginfo:read
44571 The remote stub understands the @samp{qXfer:siginfo:read} packet
44572 (@pxref{qXfer siginfo read}).
44574 @item qXfer:siginfo:write
44575 The remote stub understands the @samp{qXfer:siginfo:write} packet
44576 (@pxref{qXfer siginfo write}).
44578 @item qXfer:threads:read
44579 The remote stub understands the @samp{qXfer:threads:read} packet
44580 (@pxref{qXfer threads read}).
44582 @item qXfer:traceframe-info:read
44583 The remote stub understands the @samp{qXfer:traceframe-info:read}
44584 packet (@pxref{qXfer traceframe info read}).
44586 @item qXfer:uib:read
44587 The remote stub understands the @samp{qXfer:uib:read}
44588 packet (@pxref{qXfer unwind info block}).
44590 @item qXfer:fdpic:read
44591 The remote stub understands the @samp{qXfer:fdpic:read}
44592 packet (@pxref{qXfer fdpic loadmap read}).
44595 The remote stub understands the @samp{QNonStop} packet
44596 (@pxref{QNonStop}).
44598 @item QCatchSyscalls
44599 The remote stub understands the @samp{QCatchSyscalls} packet
44600 (@pxref{QCatchSyscalls}).
44603 The remote stub understands the @samp{QPassSignals} packet
44604 (@pxref{QPassSignals}).
44606 @item QStartNoAckMode
44607 The remote stub understands the @samp{QStartNoAckMode} packet and
44608 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
44611 @anchor{multiprocess extensions}
44612 @cindex multiprocess extensions, in remote protocol
44613 The remote stub understands the multiprocess extensions to the remote
44614 protocol syntax. The multiprocess extensions affect the syntax of
44615 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
44616 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
44617 replies. Note that reporting this feature indicates support for the
44618 syntactic extensions only, not that the stub necessarily supports
44619 debugging of more than one process at a time. The stub must not use
44620 multiprocess extensions in packet replies unless @value{GDBN} has also
44621 indicated it supports them in its @samp{qSupported} request.
44623 @item qXfer:osdata:read
44624 The remote stub understands the @samp{qXfer:osdata:read} packet
44625 ((@pxref{qXfer osdata read}).
44627 @item ConditionalBreakpoints
44628 The target accepts and implements evaluation of conditional expressions
44629 defined for breakpoints. The target will only report breakpoint triggers
44630 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
44632 @item ConditionalTracepoints
44633 The remote stub accepts and implements conditional expressions defined
44634 for tracepoints (@pxref{Tracepoint Conditions}).
44636 @item ReverseContinue
44637 The remote stub accepts and implements the reverse continue packet
44641 The remote stub accepts and implements the reverse step packet
44644 @item TracepointSource
44645 The remote stub understands the @samp{QTDPsrc} packet that supplies
44646 the source form of tracepoint definitions.
44649 The remote stub understands the @samp{QAgent} packet.
44652 The remote stub understands the @samp{QAllow} packet.
44654 @item QDisableRandomization
44655 The remote stub understands the @samp{QDisableRandomization} packet.
44657 @item StaticTracepoint
44658 @cindex static tracepoints, in remote protocol
44659 The remote stub supports static tracepoints.
44661 @item InstallInTrace
44662 @anchor{install tracepoint in tracing}
44663 The remote stub supports installing tracepoint in tracing.
44665 @item EnableDisableTracepoints
44666 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
44667 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
44668 to be enabled and disabled while a trace experiment is running.
44670 @item QTBuffer:size
44671 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
44672 packet that allows to change the size of the trace buffer.
44675 @cindex string tracing, in remote protocol
44676 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
44677 See @ref{Bytecode Descriptions} for details about the bytecode.
44679 @item BreakpointCommands
44680 @cindex breakpoint commands, in remote protocol
44681 The remote stub supports running a breakpoint's command list itself,
44682 rather than reporting the hit to @value{GDBN}.
44685 The remote stub understands the @samp{Qbtrace:off} packet.
44688 The remote stub understands the @samp{Qbtrace:bts} packet.
44691 The remote stub understands the @samp{Qbtrace:pt} packet.
44693 @item Qbtrace-conf:bts:size
44694 The remote stub understands the @samp{Qbtrace-conf:bts:size} packet.
44696 @item Qbtrace-conf:pt:size
44697 The remote stub understands the @samp{Qbtrace-conf:pt:size} packet.
44700 The remote stub reports the @samp{swbreak} stop reason for memory
44704 The remote stub reports the @samp{hwbreak} stop reason for hardware
44708 The remote stub reports the @samp{fork} stop reason for fork events.
44711 The remote stub reports the @samp{vfork} stop reason for vfork events
44712 and vforkdone events.
44715 The remote stub reports the @samp{exec} stop reason for exec events.
44717 @item vContSupported
44718 The remote stub reports the supported actions in the reply to
44719 @samp{vCont?} packet.
44721 @item QThreadEvents
44722 The remote stub understands the @samp{QThreadEvents} packet.
44725 The remote stub reports the @samp{N} stop reply.
44728 @item memory-tagging
44729 The remote stub supports and implements the required memory tagging
44730 functionality and understands the @samp{qMemTags} (@pxref{qMemTags}) and
44731 @samp{QMemTags} (@pxref{QMemTags}) packets.
44733 For AArch64 GNU/Linux systems, this feature also requires access to the
44734 @file{/proc/@var{pid}/smaps} file so memory mapping page flags can be inspected.
44735 This is done via the @samp{vFile} requests.
44740 @cindex symbol lookup, remote request
44741 @cindex @samp{qSymbol} packet
44742 Notify the target that @value{GDBN} is prepared to serve symbol lookup
44743 requests. Accept requests from the target for the values of symbols.
44748 The target does not need to look up any (more) symbols.
44749 @item qSymbol:@var{sym_name}
44750 The target requests the value of symbol @var{sym_name} (hex encoded).
44751 @value{GDBN} may provide the value by using the
44752 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
44756 @item qSymbol:@var{sym_value}:@var{sym_name}
44757 Set the value of @var{sym_name} to @var{sym_value}.
44759 @var{sym_name} (hex encoded) is the name of a symbol whose value the
44760 target has previously requested.
44762 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
44763 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
44769 The target does not need to look up any (more) symbols.
44770 @item qSymbol:@var{sym_name}
44771 The target requests the value of a new symbol @var{sym_name} (hex
44772 encoded). @value{GDBN} will continue to supply the values of symbols
44773 (if available), until the target ceases to request them.
44778 @itemx QTDisconnected
44785 @itemx qTMinFTPILen
44787 @xref{Tracepoint Packets}.
44789 @anchor{qThreadExtraInfo}
44790 @item qThreadExtraInfo,@var{thread-id}
44791 @cindex thread attributes info, remote request
44792 @cindex @samp{qThreadExtraInfo} packet
44793 Obtain from the target OS a printable string description of thread
44794 attributes for the thread @var{thread-id}; see @ref{thread-id syntax},
44795 for the forms of @var{thread-id}. This
44796 string may contain anything that the target OS thinks is interesting
44797 for @value{GDBN} to tell the user about the thread. The string is
44798 displayed in @value{GDBN}'s @code{info threads} display. Some
44799 examples of possible thread extra info strings are @samp{Runnable}, or
44800 @samp{Blocked on Mutex}.
44804 @item @var{XX}@dots{}
44805 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
44806 comprising the printable string containing the extra information about
44807 the thread's attributes.
44810 (Note that the @code{qThreadExtraInfo} packet's name is separated from
44811 the command by a @samp{,}, not a @samp{:}, contrary to the naming
44812 conventions above. Please don't use this packet as a model for new
44831 @xref{Tracepoint Packets}.
44833 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
44834 @cindex read special object, remote request
44835 @cindex @samp{qXfer} packet
44836 @anchor{qXfer read}
44837 Read uninterpreted bytes from the target's special data area
44838 identified by the keyword @var{object}. Request @var{length} bytes
44839 starting at @var{offset} bytes into the data. The content and
44840 encoding of @var{annex} is specific to @var{object}; it can supply
44841 additional details about what data to access.
44846 Data @var{data} (@pxref{Binary Data}) has been read from the
44847 target. There may be more data at a higher address (although
44848 it is permitted to return @samp{m} even for the last valid
44849 block of data, as long as at least one byte of data was read).
44850 It is possible for @var{data} to have fewer bytes than the @var{length} in the
44854 Data @var{data} (@pxref{Binary Data}) has been read from the target.
44855 There is no more data to be read. It is possible for @var{data} to
44856 have fewer bytes than the @var{length} in the request.
44859 The @var{offset} in the request is at the end of the data.
44860 There is no more data to be read.
44863 The request was malformed, or @var{annex} was invalid.
44866 The offset was invalid, or there was an error encountered reading the data.
44867 The @var{nn} part is a hex-encoded @code{errno} value.
44870 An empty reply indicates the @var{object} string was not recognized by
44871 the stub, or that the object does not support reading.
44874 Here are the specific requests of this form defined so far. All the
44875 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
44876 formats, listed above.
44879 @item qXfer:auxv:read::@var{offset},@var{length}
44880 @anchor{qXfer auxiliary vector read}
44881 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
44882 auxiliary vector}. Note @var{annex} must be empty.
44884 This packet is not probed by default; the remote stub must request it,
44885 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
44887 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
44888 @anchor{qXfer btrace read}
44890 Return a description of the current branch trace.
44891 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
44892 packet may have one of the following values:
44896 Returns all available branch trace.
44899 Returns all available branch trace if the branch trace changed since
44900 the last read request.
44903 Returns the new branch trace since the last read request. Adds a new
44904 block to the end of the trace that begins at zero and ends at the source
44905 location of the first branch in the trace buffer. This extra block is
44906 used to stitch traces together.
44908 If the trace buffer overflowed, returns an error indicating the overflow.
44911 This packet is not probed by default; the remote stub must request it
44912 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
44914 @item qXfer:btrace-conf:read::@var{offset},@var{length}
44915 @anchor{qXfer btrace-conf read}
44917 Return a description of the current branch trace configuration.
44918 @xref{Branch Trace Configuration Format}.
44920 This packet is not probed by default; the remote stub must request it
44921 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
44923 @item qXfer:exec-file:read:@var{annex}:@var{offset},@var{length}
44924 @anchor{qXfer executable filename read}
44925 Return the full absolute name of the file that was executed to create
44926 a process running on the remote system. The annex specifies the
44927 numeric process ID of the process to query, encoded as a hexadecimal
44928 number. If the annex part is empty the remote stub should return the
44929 filename corresponding to the currently executing process.
44931 This packet is not probed by default; the remote stub must request it,
44932 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
44934 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
44935 @anchor{qXfer target description read}
44936 Access the @dfn{target description}. @xref{Target Descriptions}. The
44937 annex specifies which XML document to access. The main description is
44938 always loaded from the @samp{target.xml} annex.
44940 This packet is not probed by default; the remote stub must request it,
44941 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
44943 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
44944 @anchor{qXfer library list read}
44945 Access the target's list of loaded libraries. @xref{Library List Format}.
44946 The annex part of the generic @samp{qXfer} packet must be empty
44947 (@pxref{qXfer read}).
44949 Targets which maintain a list of libraries in the program's memory do
44950 not need to implement this packet; it is designed for platforms where
44951 the operating system manages the list of loaded libraries.
44953 This packet is not probed by default; the remote stub must request it,
44954 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
44956 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
44957 @anchor{qXfer svr4 library list read}
44958 Access the target's list of loaded libraries when the target is an SVR4
44959 platform. @xref{Library List Format for SVR4 Targets}. The annex part
44960 of the generic @samp{qXfer} packet must be empty unless the remote
44961 stub indicated it supports the augmented form of this packet
44962 by supplying an appropriate @samp{qSupported} response
44963 (@pxref{qXfer read}, @ref{qSupported}).
44965 This packet is optional for better performance on SVR4 targets.
44966 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
44968 This packet is not probed by default; the remote stub must request it,
44969 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
44971 If the remote stub indicates it supports the augmented form of this
44972 packet then the annex part of the generic @samp{qXfer} packet may
44973 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
44974 arguments. The currently supported arguments are:
44977 @item start=@var{address}
44978 A hexadecimal number specifying the address of the @samp{struct
44979 link_map} to start reading the library list from. If unset or zero
44980 then the first @samp{struct link_map} in the library list will be
44981 chosen as the starting point.
44983 @item prev=@var{address}
44984 A hexadecimal number specifying the address of the @samp{struct
44985 link_map} immediately preceding the @samp{struct link_map}
44986 specified by the @samp{start} argument. If unset or zero then
44987 the remote stub will expect that no @samp{struct link_map}
44988 exists prior to the starting point.
44990 @item lmid=@var{lmid}
44991 A hexadecimal number specifying a namespace identifier. This is
44992 currently only used together with @samp{start} to provide the
44993 namespace identifier back to @value{GDBN} in the response.
44994 @value{GDBN} will only provide values that were previously reported to
44995 it. If unset, the response will include @samp{lmid="0x0"}.
44998 Arguments that are not understood by the remote stub will be silently
45001 @item qXfer:memory-map:read::@var{offset},@var{length}
45002 @anchor{qXfer memory map read}
45003 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
45004 annex part of the generic @samp{qXfer} packet must be empty
45005 (@pxref{qXfer read}).
45007 This packet is not probed by default; the remote stub must request it,
45008 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
45010 @item qXfer:sdata:read::@var{offset},@var{length}
45011 @anchor{qXfer sdata read}
45013 Read contents of the extra collected static tracepoint marker
45014 information. The annex part of the generic @samp{qXfer} packet must
45015 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
45018 This packet is not probed by default; the remote stub must request it,
45019 by supplying an appropriate @samp{qSupported} response
45020 (@pxref{qSupported}).
45022 @item qXfer:siginfo:read::@var{offset},@var{length}
45023 @anchor{qXfer siginfo read}
45024 Read contents of the extra signal information on the target
45025 system. The annex part of the generic @samp{qXfer} packet must be
45026 empty (@pxref{qXfer read}).
45028 This packet is not probed by default; the remote stub must request it,
45029 by supplying an appropriate @samp{qSupported} response
45030 (@pxref{qSupported}).
45032 @item qXfer:threads:read::@var{offset},@var{length}
45033 @anchor{qXfer threads read}
45034 Access the list of threads on target. @xref{Thread List Format}. The
45035 annex part of the generic @samp{qXfer} packet must be empty
45036 (@pxref{qXfer read}).
45038 This packet is not probed by default; the remote stub must request it,
45039 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
45041 @item qXfer:traceframe-info:read::@var{offset},@var{length}
45042 @anchor{qXfer traceframe info read}
45044 Return a description of the current traceframe's contents.
45045 @xref{Traceframe Info Format}. The annex part of the generic
45046 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
45048 This packet is not probed by default; the remote stub must request it,
45049 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
45051 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
45052 @anchor{qXfer unwind info block}
45054 Return the unwind information block for @var{pc}. This packet is used
45055 on OpenVMS/ia64 to ask the kernel unwind information.
45057 This packet is not probed by default.
45059 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
45060 @anchor{qXfer fdpic loadmap read}
45061 Read contents of @code{loadmap}s on the target system. The
45062 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
45063 executable @code{loadmap} or interpreter @code{loadmap} to read.
45065 This packet is not probed by default; the remote stub must request it,
45066 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
45068 @item qXfer:osdata:read::@var{offset},@var{length}
45069 @anchor{qXfer osdata read}
45070 Access the target's @dfn{operating system information}.
45071 @xref{Operating System Information}.
45075 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
45076 @cindex write data into object, remote request
45077 @anchor{qXfer write}
45078 Write uninterpreted bytes into the target's special data area
45079 identified by the keyword @var{object}, starting at @var{offset} bytes
45080 into the data. The binary-encoded data (@pxref{Binary Data}) to be
45081 written is given by @var{data}@dots{}. The content and encoding of @var{annex}
45082 is specific to @var{object}; it can supply additional details about what data
45088 @var{nn} (hex encoded) is the number of bytes written.
45089 This may be fewer bytes than supplied in the request.
45092 The request was malformed, or @var{annex} was invalid.
45095 The offset was invalid, or there was an error encountered writing the data.
45096 The @var{nn} part is a hex-encoded @code{errno} value.
45099 An empty reply indicates the @var{object} string was not
45100 recognized by the stub, or that the object does not support writing.
45103 Here are the specific requests of this form defined so far. All the
45104 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
45105 formats, listed above.
45108 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
45109 @anchor{qXfer siginfo write}
45110 Write @var{data} to the extra signal information on the target system.
45111 The annex part of the generic @samp{qXfer} packet must be
45112 empty (@pxref{qXfer write}).
45114 This packet is not probed by default; the remote stub must request it,
45115 by supplying an appropriate @samp{qSupported} response
45116 (@pxref{qSupported}).
45119 @item qXfer:@var{object}:@var{operation}:@dots{}
45120 Requests of this form may be added in the future. When a stub does
45121 not recognize the @var{object} keyword, or its support for
45122 @var{object} does not recognize the @var{operation} keyword, the stub
45123 must respond with an empty packet.
45125 @item qAttached:@var{pid}
45126 @cindex query attached, remote request
45127 @cindex @samp{qAttached} packet
45128 Return an indication of whether the remote server attached to an
45129 existing process or created a new process. When the multiprocess
45130 protocol extensions are supported (@pxref{multiprocess extensions}),
45131 @var{pid} is an integer in hexadecimal format identifying the target
45132 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
45133 the query packet will be simplified as @samp{qAttached}.
45135 This query is used, for example, to know whether the remote process
45136 should be detached or killed when a @value{GDBN} session is ended with
45137 the @code{quit} command.
45142 The remote server attached to an existing process.
45144 The remote server created a new process.
45146 A badly formed request or an error was encountered.
45150 Enable branch tracing for the current thread using Branch Trace Store.
45155 Branch tracing has been enabled.
45157 A badly formed request or an error was encountered.
45161 Enable branch tracing for the current thread using Intel Processor Trace.
45166 Branch tracing has been enabled.
45168 A badly formed request or an error was encountered.
45172 Disable branch tracing for the current thread.
45177 Branch tracing has been disabled.
45179 A badly formed request or an error was encountered.
45182 @item Qbtrace-conf:bts:size=@var{value}
45183 Set the requested ring buffer size for new threads that use the
45184 btrace recording method in bts format.
45189 The ring buffer size has been set.
45191 A badly formed request or an error was encountered.
45194 @item Qbtrace-conf:pt:size=@var{value}
45195 Set the requested ring buffer size for new threads that use the
45196 btrace recording method in pt format.
45201 The ring buffer size has been set.
45203 A badly formed request or an error was encountered.
45208 @node Architecture-Specific Protocol Details
45209 @section Architecture-Specific Protocol Details
45211 This section describes how the remote protocol is applied to specific
45212 target architectures. Also see @ref{Standard Target Features}, for
45213 details of XML target descriptions for each architecture.
45216 * ARM-Specific Protocol Details::
45217 * MIPS-Specific Protocol Details::
45220 @node ARM-Specific Protocol Details
45221 @subsection @acronym{ARM}-specific Protocol Details
45224 * ARM Breakpoint Kinds::
45225 * ARM Memory Tag Types::
45228 @node ARM Breakpoint Kinds
45229 @subsubsection @acronym{ARM} Breakpoint Kinds
45230 @cindex breakpoint kinds, @acronym{ARM}
45232 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
45237 16-bit Thumb mode breakpoint.
45240 32-bit Thumb mode (Thumb-2) breakpoint.
45243 32-bit @acronym{ARM} mode breakpoint.
45247 @node ARM Memory Tag Types
45248 @subsubsection @acronym{ARM} Memory Tag Types
45249 @cindex memory tag types, @acronym{ARM}
45251 These memory tag types are defined for the @samp{qMemTag} and @samp{QMemTag}
45264 @node MIPS-Specific Protocol Details
45265 @subsection @acronym{MIPS}-specific Protocol Details
45268 * MIPS Register packet Format::
45269 * MIPS Breakpoint Kinds::
45272 @node MIPS Register packet Format
45273 @subsubsection @acronym{MIPS} Register Packet Format
45274 @cindex register packet format, @acronym{MIPS}
45276 The following @code{g}/@code{G} packets have previously been defined.
45277 In the below, some thirty-two bit registers are transferred as
45278 sixty-four bits. Those registers should be zero/sign extended (which?)
45279 to fill the space allocated. Register bytes are transferred in target
45280 byte order. The two nibbles within a register byte are transferred
45281 most-significant -- least-significant.
45286 All registers are transferred as thirty-two bit quantities in the order:
45287 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
45288 registers; fsr; fir; fp.
45291 All registers are transferred as sixty-four bit quantities (including
45292 thirty-two bit registers such as @code{sr}). The ordering is the same
45297 @node MIPS Breakpoint Kinds
45298 @subsubsection @acronym{MIPS} Breakpoint Kinds
45299 @cindex breakpoint kinds, @acronym{MIPS}
45301 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
45306 16-bit @acronym{MIPS16} mode breakpoint.
45309 16-bit @acronym{microMIPS} mode breakpoint.
45312 32-bit standard @acronym{MIPS} mode breakpoint.
45315 32-bit @acronym{microMIPS} mode breakpoint.
45319 @node Tracepoint Packets
45320 @section Tracepoint Packets
45321 @cindex tracepoint packets
45322 @cindex packets, tracepoint
45324 Here we describe the packets @value{GDBN} uses to implement
45325 tracepoints (@pxref{Tracepoints}).
45329 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
45330 @cindex @samp{QTDP} packet
45331 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
45332 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
45333 the tracepoint is disabled. The @var{step} gives the tracepoint's step
45334 count, and @var{pass} gives its pass count. If an @samp{F} is present,
45335 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
45336 the number of bytes that the target should copy elsewhere to make room
45337 for the tracepoint. If an @samp{X} is present, it introduces a
45338 tracepoint condition, which consists of a hexadecimal length, followed
45339 by a comma and hex-encoded bytes, in a manner similar to action
45340 encodings as described below. If the trailing @samp{-} is present,
45341 further @samp{QTDP} packets will follow to specify this tracepoint's
45347 The packet was understood and carried out.
45349 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
45351 The packet was not recognized.
45354 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
45355 Define actions to be taken when a tracepoint is hit. The @var{n} and
45356 @var{addr} must be the same as in the initial @samp{QTDP} packet for
45357 this tracepoint. This packet may only be sent immediately after
45358 another @samp{QTDP} packet that ended with a @samp{-}. If the
45359 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
45360 specifying more actions for this tracepoint.
45362 In the series of action packets for a given tracepoint, at most one
45363 can have an @samp{S} before its first @var{action}. If such a packet
45364 is sent, it and the following packets define ``while-stepping''
45365 actions. Any prior packets define ordinary actions --- that is, those
45366 taken when the tracepoint is first hit. If no action packet has an
45367 @samp{S}, then all the packets in the series specify ordinary
45368 tracepoint actions.
45370 The @samp{@var{action}@dots{}} portion of the packet is a series of
45371 actions, concatenated without separators. Each action has one of the
45377 Collect the registers whose bits are set in @var{mask},
45378 a hexadecimal number whose @var{i}'th bit is set if register number
45379 @var{i} should be collected. (The least significant bit is numbered
45380 zero.) Note that @var{mask} may be any number of digits long; it may
45381 not fit in a 32-bit word.
45383 @item M @var{basereg},@var{offset},@var{len}
45384 Collect @var{len} bytes of memory starting at the address in register
45385 number @var{basereg}, plus @var{offset}. If @var{basereg} is
45386 @samp{-1}, then the range has a fixed address: @var{offset} is the
45387 address of the lowest byte to collect. The @var{basereg},
45388 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
45389 values (the @samp{-1} value for @var{basereg} is a special case).
45391 @item X @var{len},@var{expr}
45392 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
45393 it directs. The agent expression @var{expr} is as described in
45394 @ref{Agent Expressions}. Each byte of the expression is encoded as a
45395 two-digit hex number in the packet; @var{len} is the number of bytes
45396 in the expression (and thus one-half the number of hex digits in the
45401 Any number of actions may be packed together in a single @samp{QTDP}
45402 packet, as long as the packet does not exceed the maximum packet
45403 length (400 bytes, for many stubs). There may be only one @samp{R}
45404 action per tracepoint, and it must precede any @samp{M} or @samp{X}
45405 actions. Any registers referred to by @samp{M} and @samp{X} actions
45406 must be collected by a preceding @samp{R} action. (The
45407 ``while-stepping'' actions are treated as if they were attached to a
45408 separate tracepoint, as far as these restrictions are concerned.)
45413 The packet was understood and carried out.
45415 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
45417 The packet was not recognized.
45420 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
45421 @cindex @samp{QTDPsrc} packet
45422 Specify a source string of tracepoint @var{n} at address @var{addr}.
45423 This is useful to get accurate reproduction of the tracepoints
45424 originally downloaded at the beginning of the trace run. The @var{type}
45425 is the name of the tracepoint part, such as @samp{cond} for the
45426 tracepoint's conditional expression (see below for a list of types), while
45427 @var{bytes} is the string, encoded in hexadecimal.
45429 @var{start} is the offset of the @var{bytes} within the overall source
45430 string, while @var{slen} is the total length of the source string.
45431 This is intended for handling source strings that are longer than will
45432 fit in a single packet.
45433 @c Add detailed example when this info is moved into a dedicated
45434 @c tracepoint descriptions section.
45436 The available string types are @samp{at} for the location,
45437 @samp{cond} for the conditional, and @samp{cmd} for an action command.
45438 @value{GDBN} sends a separate packet for each command in the action
45439 list, in the same order in which the commands are stored in the list.
45441 The target does not need to do anything with source strings except
45442 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
45445 Although this packet is optional, and @value{GDBN} will only send it
45446 if the target replies with @samp{TracepointSource} @xref{General
45447 Query Packets}, it makes both disconnected tracing and trace files
45448 much easier to use. Otherwise the user must be careful that the
45449 tracepoints in effect while looking at trace frames are identical to
45450 the ones in effect during the trace run; even a small discrepancy
45451 could cause @samp{tdump} not to work, or a particular trace frame not
45454 @item QTDV:@var{n}:@var{value}:@var{builtin}:@var{name}
45455 @cindex define trace state variable, remote request
45456 @cindex @samp{QTDV} packet
45457 Create a new trace state variable, number @var{n}, with an initial
45458 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
45459 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
45460 the option of not using this packet for initial values of zero; the
45461 target should simply create the trace state variables as they are
45462 mentioned in expressions. The value @var{builtin} should be 1 (one)
45463 if the trace state variable is builtin and 0 (zero) if it is not builtin.
45464 @value{GDBN} only sets @var{builtin} to 1 if a previous @samp{qTfV} or
45465 @samp{qTsV} packet had it set. The contents of @var{name} is the
45466 hex-encoded name (without the leading @samp{$}) of the trace state
45469 @item QTFrame:@var{n}
45470 @cindex @samp{QTFrame} packet
45471 Select the @var{n}'th tracepoint frame from the buffer, and use the
45472 register and memory contents recorded there to answer subsequent
45473 request packets from @value{GDBN}.
45475 A successful reply from the stub indicates that the stub has found the
45476 requested frame. The response is a series of parts, concatenated
45477 without separators, describing the frame we selected. Each part has
45478 one of the following forms:
45482 The selected frame is number @var{n} in the trace frame buffer;
45483 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
45484 was no frame matching the criteria in the request packet.
45487 The selected trace frame records a hit of tracepoint number @var{t};
45488 @var{t} is a hexadecimal number.
45492 @item QTFrame:pc:@var{addr}
45493 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
45494 currently selected frame whose PC is @var{addr};
45495 @var{addr} is a hexadecimal number.
45497 @item QTFrame:tdp:@var{t}
45498 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
45499 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
45500 is a hexadecimal number.
45502 @item QTFrame:range:@var{start}:@var{end}
45503 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
45504 currently selected frame whose PC is between @var{start} (inclusive)
45505 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
45508 @item QTFrame:outside:@var{start}:@var{end}
45509 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
45510 frame @emph{outside} the given range of addresses (exclusive).
45513 @cindex @samp{qTMinFTPILen} packet
45514 This packet requests the minimum length of instruction at which a fast
45515 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
45516 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
45517 it depends on the target system being able to create trampolines in
45518 the first 64K of memory, which might or might not be possible for that
45519 system. So the reply to this packet will be 4 if it is able to
45526 The minimum instruction length is currently unknown.
45528 The minimum instruction length is @var{length}, where @var{length}
45529 is a hexadecimal number greater or equal to 1. A reply
45530 of 1 means that a fast tracepoint may be placed on any instruction
45531 regardless of size.
45533 An error has occurred.
45535 An empty reply indicates that the request is not supported by the stub.
45539 @cindex @samp{QTStart} packet
45540 Begin the tracepoint experiment. Begin collecting data from
45541 tracepoint hits in the trace frame buffer. This packet supports the
45542 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
45543 instruction reply packet}).
45546 @cindex @samp{QTStop} packet
45547 End the tracepoint experiment. Stop collecting trace frames.
45549 @item QTEnable:@var{n}:@var{addr}
45551 @cindex @samp{QTEnable} packet
45552 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
45553 experiment. If the tracepoint was previously disabled, then collection
45554 of data from it will resume.
45556 @item QTDisable:@var{n}:@var{addr}
45558 @cindex @samp{QTDisable} packet
45559 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
45560 experiment. No more data will be collected from the tracepoint unless
45561 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
45564 @cindex @samp{QTinit} packet
45565 Clear the table of tracepoints, and empty the trace frame buffer.
45567 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
45568 @cindex @samp{QTro} packet
45569 Establish the given ranges of memory as ``transparent''. The stub
45570 will answer requests for these ranges from memory's current contents,
45571 if they were not collected as part of the tracepoint hit.
45573 @value{GDBN} uses this to mark read-only regions of memory, like those
45574 containing program code. Since these areas never change, they should
45575 still have the same contents they did when the tracepoint was hit, so
45576 there's no reason for the stub to refuse to provide their contents.
45578 @item QTDisconnected:@var{value}
45579 @cindex @samp{QTDisconnected} packet
45580 Set the choice to what to do with the tracing run when @value{GDBN}
45581 disconnects from the target. A @var{value} of 1 directs the target to
45582 continue the tracing run, while 0 tells the target to stop tracing if
45583 @value{GDBN} is no longer in the picture.
45586 @cindex @samp{qTStatus} packet
45587 Ask the stub if there is a trace experiment running right now.
45589 The reply has the form:
45593 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
45594 @var{running} is a single digit @code{1} if the trace is presently
45595 running, or @code{0} if not. It is followed by semicolon-separated
45596 optional fields that an agent may use to report additional status.
45600 If the trace is not running, the agent may report any of several
45601 explanations as one of the optional fields:
45606 No trace has been run yet.
45608 @item tstop[:@var{text}]:0
45609 The trace was stopped by a user-originated stop command. The optional
45610 @var{text} field is a user-supplied string supplied as part of the
45611 stop command (for instance, an explanation of why the trace was
45612 stopped manually). It is hex-encoded.
45615 The trace stopped because the trace buffer filled up.
45617 @item tdisconnected:0
45618 The trace stopped because @value{GDBN} disconnected from the target.
45620 @item tpasscount:@var{tpnum}
45621 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
45623 @item terror:@var{text}:@var{tpnum}
45624 The trace stopped because tracepoint @var{tpnum} had an error. The
45625 string @var{text} is available to describe the nature of the error
45626 (for instance, a divide by zero in the condition expression); it
45630 The trace stopped for some other reason.
45634 Additional optional fields supply statistical and other information.
45635 Although not required, they are extremely useful for users monitoring
45636 the progress of a trace run. If a trace has stopped, and these
45637 numbers are reported, they must reflect the state of the just-stopped
45642 @item tframes:@var{n}
45643 The number of trace frames in the buffer.
45645 @item tcreated:@var{n}
45646 The total number of trace frames created during the run. This may
45647 be larger than the trace frame count, if the buffer is circular.
45649 @item tsize:@var{n}
45650 The total size of the trace buffer, in bytes.
45652 @item tfree:@var{n}
45653 The number of bytes still unused in the buffer.
45655 @item circular:@var{n}
45656 The value of the circular trace buffer flag. @code{1} means that the
45657 trace buffer is circular and old trace frames will be discarded if
45658 necessary to make room, @code{0} means that the trace buffer is linear
45661 @item disconn:@var{n}
45662 The value of the disconnected tracing flag. @code{1} means that
45663 tracing will continue after @value{GDBN} disconnects, @code{0} means
45664 that the trace run will stop.
45668 @item qTP:@var{tp}:@var{addr}
45669 @cindex tracepoint status, remote request
45670 @cindex @samp{qTP} packet
45671 Ask the stub for the current state of tracepoint number @var{tp} at
45672 address @var{addr}.
45676 @item V@var{hits}:@var{usage}
45677 The tracepoint has been hit @var{hits} times so far during the trace
45678 run, and accounts for @var{usage} in the trace buffer. Note that
45679 @code{while-stepping} steps are not counted as separate hits, but the
45680 steps' space consumption is added into the usage number.
45684 @item qTV:@var{var}
45685 @cindex trace state variable value, remote request
45686 @cindex @samp{qTV} packet
45687 Ask the stub for the value of the trace state variable number @var{var}.
45692 The value of the variable is @var{value}. This will be the current
45693 value of the variable if the user is examining a running target, or a
45694 saved value if the variable was collected in the trace frame that the
45695 user is looking at. Note that multiple requests may result in
45696 different reply values, such as when requesting values while the
45697 program is running.
45700 The value of the variable is unknown. This would occur, for example,
45701 if the user is examining a trace frame in which the requested variable
45706 @cindex @samp{qTfP} packet
45708 @cindex @samp{qTsP} packet
45709 These packets request data about tracepoints that are being used by
45710 the target. @value{GDBN} sends @code{qTfP} to get the first piece
45711 of data, and multiple @code{qTsP} to get additional pieces. Replies
45712 to these packets generally take the form of the @code{QTDP} packets
45713 that define tracepoints. (FIXME add detailed syntax)
45716 @cindex @samp{qTfV} packet
45718 @cindex @samp{qTsV} packet
45719 These packets request data about trace state variables that are on the
45720 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
45721 and multiple @code{qTsV} to get additional variables. Replies to
45722 these packets follow the syntax of the @code{QTDV} packets that define
45723 trace state variables.
45729 @cindex @samp{qTfSTM} packet
45730 @cindex @samp{qTsSTM} packet
45731 These packets request data about static tracepoint markers that exist
45732 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
45733 first piece of data, and multiple @code{qTsSTM} to get additional
45734 pieces. Replies to these packets take the following form:
45738 @item m @var{address}:@var{id}:@var{extra}
45740 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
45741 a comma-separated list of markers
45743 (lower case letter @samp{L}) denotes end of list.
45745 An error occurred. The error number @var{nn} is given as hex digits.
45747 An empty reply indicates that the request is not supported by the
45751 The @var{address} is encoded in hex;
45752 @var{id} and @var{extra} are strings encoded in hex.
45754 In response to each query, the target will reply with a list of one or
45755 more markers, separated by commas. @value{GDBN} will respond to each
45756 reply with a request for more markers (using the @samp{qs} form of the
45757 query), until the target responds with @samp{l} (lower-case ell, for
45760 @item qTSTMat:@var{address}
45762 @cindex @samp{qTSTMat} packet
45763 This packets requests data about static tracepoint markers in the
45764 target program at @var{address}. Replies to this packet follow the
45765 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
45766 tracepoint markers.
45768 @item QTSave:@var{filename}
45769 @cindex @samp{QTSave} packet
45770 This packet directs the target to save trace data to the file name
45771 @var{filename} in the target's filesystem. The @var{filename} is encoded
45772 as a hex string; the interpretation of the file name (relative vs
45773 absolute, wild cards, etc) is up to the target.
45775 @item qTBuffer:@var{offset},@var{len}
45776 @cindex @samp{qTBuffer} packet
45777 Return up to @var{len} bytes of the current contents of trace buffer,
45778 starting at @var{offset}. The trace buffer is treated as if it were
45779 a contiguous collection of traceframes, as per the trace file format.
45780 The reply consists as many hex-encoded bytes as the target can deliver
45781 in a packet; it is not an error to return fewer than were asked for.
45782 A reply consisting of just @code{l} indicates that no bytes are
45785 @item QTBuffer:circular:@var{value}
45786 This packet directs the target to use a circular trace buffer if
45787 @var{value} is 1, or a linear buffer if the value is 0.
45789 @item QTBuffer:size:@var{size}
45790 @anchor{QTBuffer-size}
45791 @cindex @samp{QTBuffer size} packet
45792 This packet directs the target to make the trace buffer be of size
45793 @var{size} if possible. A value of @code{-1} tells the target to
45794 use whatever size it prefers.
45796 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
45797 @cindex @samp{QTNotes} packet
45798 This packet adds optional textual notes to the trace run. Allowable
45799 types include @code{user}, @code{notes}, and @code{tstop}, the
45800 @var{text} fields are arbitrary strings, hex-encoded.
45804 @subsection Relocate instruction reply packet
45805 When installing fast tracepoints in memory, the target may need to
45806 relocate the instruction currently at the tracepoint address to a
45807 different address in memory. For most instructions, a simple copy is
45808 enough, but, for example, call instructions that implicitly push the
45809 return address on the stack, and relative branches or other
45810 PC-relative instructions require offset adjustment, so that the effect
45811 of executing the instruction at a different address is the same as if
45812 it had executed in the original location.
45814 In response to several of the tracepoint packets, the target may also
45815 respond with a number of intermediate @samp{qRelocInsn} request
45816 packets before the final result packet, to have @value{GDBN} handle
45817 this relocation operation. If a packet supports this mechanism, its
45818 documentation will explicitly say so. See for example the above
45819 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
45820 format of the request is:
45823 @item qRelocInsn:@var{from};@var{to}
45825 This requests @value{GDBN} to copy instruction at address @var{from}
45826 to address @var{to}, possibly adjusted so that executing the
45827 instruction at @var{to} has the same effect as executing it at
45828 @var{from}. @value{GDBN} writes the adjusted instruction to target
45829 memory starting at @var{to}.
45834 @item qRelocInsn:@var{adjusted_size}
45835 Informs the stub the relocation is complete. The @var{adjusted_size} is
45836 the length in bytes of resulting relocated instruction sequence.
45838 A badly formed request was detected, or an error was encountered while
45839 relocating the instruction.
45842 @node Host I/O Packets
45843 @section Host I/O Packets
45844 @cindex Host I/O, remote protocol
45845 @cindex file transfer, remote protocol
45847 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
45848 operations on the far side of a remote link. For example, Host I/O is
45849 used to upload and download files to a remote target with its own
45850 filesystem. Host I/O uses the same constant values and data structure
45851 layout as the target-initiated File-I/O protocol. However, the
45852 Host I/O packets are structured differently. The target-initiated
45853 protocol relies on target memory to store parameters and buffers.
45854 Host I/O requests are initiated by @value{GDBN}, and the
45855 target's memory is not involved. @xref{File-I/O Remote Protocol
45856 Extension}, for more details on the target-initiated protocol.
45858 The Host I/O request packets all encode a single operation along with
45859 its arguments. They have this format:
45863 @item vFile:@var{operation}: @var{parameter}@dots{}
45864 @var{operation} is the name of the particular request; the target
45865 should compare the entire packet name up to the second colon when checking
45866 for a supported operation. The format of @var{parameter} depends on
45867 the operation. Numbers are always passed in hexadecimal. Negative
45868 numbers have an explicit minus sign (i.e.@: two's complement is not
45869 used). Strings (e.g.@: filenames) are encoded as a series of
45870 hexadecimal bytes. The last argument to a system call may be a
45871 buffer of escaped binary data (@pxref{Binary Data}).
45875 The valid responses to Host I/O packets are:
45879 @item F @var{result} [, @var{errno}] [; @var{attachment}]
45880 @var{result} is the integer value returned by this operation, usually
45881 non-negative for success and -1 for errors. If an error has occurred,
45882 @var{errno} will be included in the result specifying a
45883 value defined by the File-I/O protocol (@pxref{Errno Values}). For
45884 operations which return data, @var{attachment} supplies the data as a
45885 binary buffer. Binary buffers in response packets are escaped in the
45886 normal way (@pxref{Binary Data}). See the individual packet
45887 documentation for the interpretation of @var{result} and
45891 An empty response indicates that this operation is not recognized.
45895 These are the supported Host I/O operations:
45898 @item vFile:open: @var{filename}, @var{flags}, @var{mode}
45899 Open a file at @var{filename} and return a file descriptor for it, or
45900 return -1 if an error occurs. The @var{filename} is a string,
45901 @var{flags} is an integer indicating a mask of open flags
45902 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
45903 of mode bits to use if the file is created (@pxref{mode_t Values}).
45904 @xref{open}, for details of the open flags and mode values.
45906 @item vFile:close: @var{fd}
45907 Close the open file corresponding to @var{fd} and return 0, or
45908 -1 if an error occurs.
45910 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
45911 Read data from the open file corresponding to @var{fd}. Up to
45912 @var{count} bytes will be read from the file, starting at @var{offset}
45913 relative to the start of the file. The target may read fewer bytes;
45914 common reasons include packet size limits and an end-of-file
45915 condition. The number of bytes read is returned. Zero should only be
45916 returned for a successful read at the end of the file, or if
45917 @var{count} was zero.
45919 The data read should be returned as a binary attachment on success.
45920 If zero bytes were read, the response should include an empty binary
45921 attachment (i.e.@: a trailing semicolon). The return value is the
45922 number of target bytes read; the binary attachment may be longer if
45923 some characters were escaped.
45925 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
45926 Write @var{data} (a binary buffer) to the open file corresponding
45927 to @var{fd}. Start the write at @var{offset} from the start of the
45928 file. Unlike many @code{write} system calls, there is no
45929 separate @var{count} argument; the length of @var{data} in the
45930 packet is used. @samp{vFile:pwrite} returns the number of bytes written,
45931 which may be shorter than the length of @var{data}, or -1 if an
45934 @item vFile:fstat: @var{fd}
45935 Get information about the open file corresponding to @var{fd}.
45936 On success the information is returned as a binary attachment
45937 and the return value is the size of this attachment in bytes.
45938 If an error occurs the return value is -1. The format of the
45939 returned binary attachment is as described in @ref{struct stat}.
45941 @item vFile:unlink: @var{filename}
45942 Delete the file at @var{filename} on the target. Return 0,
45943 or -1 if an error occurs. The @var{filename} is a string.
45945 @item vFile:readlink: @var{filename}
45946 Read value of symbolic link @var{filename} on the target. Return
45947 the number of bytes read, or -1 if an error occurs.
45949 The data read should be returned as a binary attachment on success.
45950 If zero bytes were read, the response should include an empty binary
45951 attachment (i.e.@: a trailing semicolon). The return value is the
45952 number of target bytes read; the binary attachment may be longer if
45953 some characters were escaped.
45955 @item vFile:setfs: @var{pid}
45956 Select the filesystem on which @code{vFile} operations with
45957 @var{filename} arguments will operate. This is required for
45958 @value{GDBN} to be able to access files on remote targets where
45959 the remote stub does not share a common filesystem with the
45962 If @var{pid} is nonzero, select the filesystem as seen by process
45963 @var{pid}. If @var{pid} is zero, select the filesystem as seen by
45964 the remote stub. Return 0 on success, or -1 if an error occurs.
45965 If @code{vFile:setfs:} indicates success, the selected filesystem
45966 remains selected until the next successful @code{vFile:setfs:}
45972 @section Interrupts
45973 @cindex interrupts (remote protocol)
45974 @anchor{interrupting remote targets}
45976 In all-stop mode, when a program on the remote target is running,
45977 @value{GDBN} may attempt to interrupt it by sending a @samp{Ctrl-C},
45978 @code{BREAK} or a @code{BREAK} followed by @code{g}, control of which
45979 is specified via @value{GDBN}'s @samp{interrupt-sequence}.
45981 The precise meaning of @code{BREAK} is defined by the transport
45982 mechanism and may, in fact, be undefined. @value{GDBN} does not
45983 currently define a @code{BREAK} mechanism for any of the network
45984 interfaces except for TCP, in which case @value{GDBN} sends the
45985 @code{telnet} BREAK sequence.
45987 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
45988 transport mechanisms. It is represented by sending the single byte
45989 @code{0x03} without any of the usual packet overhead described in
45990 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
45991 transmitted as part of a packet, it is considered to be packet data
45992 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
45993 (@pxref{X packet}), used for binary downloads, may include an unescaped
45994 @code{0x03} as part of its packet.
45996 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
45997 When Linux kernel receives this sequence from serial port,
45998 it stops execution and connects to gdb.
46000 In non-stop mode, because packet resumptions are asynchronous
46001 (@pxref{vCont packet}), @value{GDBN} is always free to send a remote
46002 command to the remote stub, even when the target is running. For that
46003 reason, @value{GDBN} instead sends a regular packet (@pxref{vCtrlC
46004 packet}) with the usual packet framing instead of the single byte
46007 Stubs are not required to recognize these interrupt mechanisms and the
46008 precise meaning associated with receipt of the interrupt is
46009 implementation defined. If the target supports debugging of multiple
46010 threads and/or processes, it should attempt to interrupt all
46011 currently-executing threads and processes.
46012 If the stub is successful at interrupting the
46013 running program, it should send one of the stop
46014 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
46015 of successfully stopping the program in all-stop mode, and a stop reply
46016 for each stopped thread in non-stop mode.
46017 Interrupts received while the
46018 program is stopped are queued and the program will be interrupted when
46019 it is resumed next time.
46021 @node Notification Packets
46022 @section Notification Packets
46023 @cindex notification packets
46024 @cindex packets, notification
46026 The @value{GDBN} remote serial protocol includes @dfn{notifications},
46027 packets that require no acknowledgment. Both the GDB and the stub
46028 may send notifications (although the only notifications defined at
46029 present are sent by the stub). Notifications carry information
46030 without incurring the round-trip latency of an acknowledgment, and so
46031 are useful for low-impact communications where occasional packet loss
46034 A notification packet has the form @samp{% @var{data} #
46035 @var{checksum}}, where @var{data} is the content of the notification,
46036 and @var{checksum} is a checksum of @var{data}, computed and formatted
46037 as for ordinary @value{GDBN} packets. A notification's @var{data}
46038 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
46039 receiving a notification, the recipient sends no @samp{+} or @samp{-}
46040 to acknowledge the notification's receipt or to report its corruption.
46042 Every notification's @var{data} begins with a name, which contains no
46043 colon characters, followed by a colon character.
46045 Recipients should silently ignore corrupted notifications and
46046 notifications they do not understand. Recipients should restart
46047 timeout periods on receipt of a well-formed notification, whether or
46048 not they understand it.
46050 Senders should only send the notifications described here when this
46051 protocol description specifies that they are permitted. In the
46052 future, we may extend the protocol to permit existing notifications in
46053 new contexts; this rule helps older senders avoid confusing newer
46056 (Older versions of @value{GDBN} ignore bytes received until they see
46057 the @samp{$} byte that begins an ordinary packet, so new stubs may
46058 transmit notifications without fear of confusing older clients. There
46059 are no notifications defined for @value{GDBN} to send at the moment, but we
46060 assume that most older stubs would ignore them, as well.)
46062 Each notification is comprised of three parts:
46064 @item @var{name}:@var{event}
46065 The notification packet is sent by the side that initiates the
46066 exchange (currently, only the stub does that), with @var{event}
46067 carrying the specific information about the notification, and
46068 @var{name} specifying the name of the notification.
46070 The acknowledge sent by the other side, usually @value{GDBN}, to
46071 acknowledge the exchange and request the event.
46074 The purpose of an asynchronous notification mechanism is to report to
46075 @value{GDBN} that something interesting happened in the remote stub.
46077 The remote stub may send notification @var{name}:@var{event}
46078 at any time, but @value{GDBN} acknowledges the notification when
46079 appropriate. The notification event is pending before @value{GDBN}
46080 acknowledges. Only one notification at a time may be pending; if
46081 additional events occur before @value{GDBN} has acknowledged the
46082 previous notification, they must be queued by the stub for later
46083 synchronous transmission in response to @var{ack} packets from
46084 @value{GDBN}. Because the notification mechanism is unreliable,
46085 the stub is permitted to resend a notification if it believes
46086 @value{GDBN} may not have received it.
46088 Specifically, notifications may appear when @value{GDBN} is not
46089 otherwise reading input from the stub, or when @value{GDBN} is
46090 expecting to read a normal synchronous response or a
46091 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
46092 Notification packets are distinct from any other communication from
46093 the stub so there is no ambiguity.
46095 After receiving a notification, @value{GDBN} shall acknowledge it by
46096 sending a @var{ack} packet as a regular, synchronous request to the
46097 stub. Such acknowledgment is not required to happen immediately, as
46098 @value{GDBN} is permitted to send other, unrelated packets to the
46099 stub first, which the stub should process normally.
46101 Upon receiving a @var{ack} packet, if the stub has other queued
46102 events to report to @value{GDBN}, it shall respond by sending a
46103 normal @var{event}. @value{GDBN} shall then send another @var{ack}
46104 packet to solicit further responses; again, it is permitted to send
46105 other, unrelated packets as well which the stub should process
46108 If the stub receives a @var{ack} packet and there are no additional
46109 @var{event} to report, the stub shall return an @samp{OK} response.
46110 At this point, @value{GDBN} has finished processing a notification
46111 and the stub has completed sending any queued events. @value{GDBN}
46112 won't accept any new notifications until the final @samp{OK} is
46113 received . If further notification events occur, the stub shall send
46114 a new notification, @value{GDBN} shall accept the notification, and
46115 the process shall be repeated.
46117 The process of asynchronous notification can be illustrated by the
46120 <- @code{%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
46123 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
46125 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
46130 The following notifications are defined:
46131 @multitable @columnfractions 0.12 0.12 0.38 0.38
46140 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
46141 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
46142 for information on how these notifications are acknowledged by
46144 @tab Report an asynchronous stop event in non-stop mode.
46148 @node Remote Non-Stop
46149 @section Remote Protocol Support for Non-Stop Mode
46151 @value{GDBN}'s remote protocol supports non-stop debugging of
46152 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
46153 supports non-stop mode, it should report that to @value{GDBN} by including
46154 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
46156 @value{GDBN} typically sends a @samp{QNonStop} packet only when
46157 establishing a new connection with the stub. Entering non-stop mode
46158 does not alter the state of any currently-running threads, but targets
46159 must stop all threads in any already-attached processes when entering
46160 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
46161 probe the target state after a mode change.
46163 In non-stop mode, when an attached process encounters an event that
46164 would otherwise be reported with a stop reply, it uses the
46165 asynchronous notification mechanism (@pxref{Notification Packets}) to
46166 inform @value{GDBN}. In contrast to all-stop mode, where all threads
46167 in all processes are stopped when a stop reply is sent, in non-stop
46168 mode only the thread reporting the stop event is stopped. That is,
46169 when reporting a @samp{S} or @samp{T} response to indicate completion
46170 of a step operation, hitting a breakpoint, or a fault, only the
46171 affected thread is stopped; any other still-running threads continue
46172 to run. When reporting a @samp{W} or @samp{X} response, all running
46173 threads belonging to other attached processes continue to run.
46175 In non-stop mode, the target shall respond to the @samp{?} packet as
46176 follows. First, any incomplete stop reply notification/@samp{vStopped}
46177 sequence in progress is abandoned. The target must begin a new
46178 sequence reporting stop events for all stopped threads, whether or not
46179 it has previously reported those events to @value{GDBN}. The first
46180 stop reply is sent as a synchronous reply to the @samp{?} packet, and
46181 subsequent stop replies are sent as responses to @samp{vStopped} packets
46182 using the mechanism described above. The target must not send
46183 asynchronous stop reply notifications until the sequence is complete.
46184 If all threads are running when the target receives the @samp{?} packet,
46185 or if the target is not attached to any process, it shall respond
46188 If the stub supports non-stop mode, it should also support the
46189 @samp{swbreak} stop reason if software breakpoints are supported, and
46190 the @samp{hwbreak} stop reason if hardware breakpoints are supported
46191 (@pxref{swbreak stop reason}). This is because given the asynchronous
46192 nature of non-stop mode, between the time a thread hits a breakpoint
46193 and the time the event is finally processed by @value{GDBN}, the
46194 breakpoint may have already been removed from the target. Due to
46195 this, @value{GDBN} needs to be able to tell whether a trap stop was
46196 caused by a delayed breakpoint event, which should be ignored, as
46197 opposed to a random trap signal, which should be reported to the user.
46198 Note the @samp{swbreak} feature implies that the target is responsible
46199 for adjusting the PC when a software breakpoint triggers, if
46200 necessary, such as on the x86 architecture.
46202 @node Packet Acknowledgment
46203 @section Packet Acknowledgment
46205 @cindex acknowledgment, for @value{GDBN} remote
46206 @cindex packet acknowledgment, for @value{GDBN} remote
46207 By default, when either the host or the target machine receives a packet,
46208 the first response expected is an acknowledgment: either @samp{+} (to indicate
46209 the package was received correctly) or @samp{-} (to request retransmission).
46210 This mechanism allows the @value{GDBN} remote protocol to operate over
46211 unreliable transport mechanisms, such as a serial line.
46213 In cases where the transport mechanism is itself reliable (such as a pipe or
46214 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
46215 It may be desirable to disable them in that case to reduce communication
46216 overhead, or for other reasons. This can be accomplished by means of the
46217 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
46219 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
46220 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
46221 and response format still includes the normal checksum, as described in
46222 @ref{Overview}, but the checksum may be ignored by the receiver.
46224 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
46225 no-acknowledgment mode, it should report that to @value{GDBN}
46226 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
46227 @pxref{qSupported}.
46228 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
46229 disabled via the @code{set remote noack-packet off} command
46230 (@pxref{Remote Configuration}),
46231 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
46232 Only then may the stub actually turn off packet acknowledgments.
46233 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
46234 response, which can be safely ignored by the stub.
46236 Note that @code{set remote noack-packet} command only affects negotiation
46237 between @value{GDBN} and the stub when subsequent connections are made;
46238 it does not affect the protocol acknowledgment state for any current
46240 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
46241 new connection is established,
46242 there is also no protocol request to re-enable the acknowledgments
46243 for the current connection, once disabled.
46248 Example sequence of a target being re-started. Notice how the restart
46249 does not get any direct output:
46254 @emph{target restarts}
46257 <- @code{T001:1234123412341234}
46261 Example sequence of a target being stepped by a single instruction:
46264 -> @code{G1445@dots{}}
46269 <- @code{T001:1234123412341234}
46273 <- @code{1455@dots{}}
46277 @node File-I/O Remote Protocol Extension
46278 @section File-I/O Remote Protocol Extension
46279 @cindex File-I/O remote protocol extension
46282 * File-I/O Overview::
46283 * Protocol Basics::
46284 * The F Request Packet::
46285 * The F Reply Packet::
46286 * The Ctrl-C Message::
46288 * List of Supported Calls::
46289 * Protocol-specific Representation of Datatypes::
46291 * File-I/O Examples::
46294 @node File-I/O Overview
46295 @subsection File-I/O Overview
46296 @cindex file-i/o overview
46298 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
46299 target to use the host's file system and console I/O to perform various
46300 system calls. System calls on the target system are translated into a
46301 remote protocol packet to the host system, which then performs the needed
46302 actions and returns a response packet to the target system.
46303 This simulates file system operations even on targets that lack file systems.
46305 The protocol is defined to be independent of both the host and target systems.
46306 It uses its own internal representation of datatypes and values. Both
46307 @value{GDBN} and the target's @value{GDBN} stub are responsible for
46308 translating the system-dependent value representations into the internal
46309 protocol representations when data is transmitted.
46311 The communication is synchronous. A system call is possible only when
46312 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
46313 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
46314 the target is stopped to allow deterministic access to the target's
46315 memory. Therefore File-I/O is not interruptible by target signals. On
46316 the other hand, it is possible to interrupt File-I/O by a user interrupt
46317 (@samp{Ctrl-C}) within @value{GDBN}.
46319 The target's request to perform a host system call does not finish
46320 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
46321 after finishing the system call, the target returns to continuing the
46322 previous activity (continue, step). No additional continue or step
46323 request from @value{GDBN} is required.
46326 (@value{GDBP}) continue
46327 <- target requests 'system call X'
46328 target is stopped, @value{GDBN} executes system call
46329 -> @value{GDBN} returns result
46330 ... target continues, @value{GDBN} returns to wait for the target
46331 <- target hits breakpoint and sends a Txx packet
46334 The protocol only supports I/O on the console and to regular files on
46335 the host file system. Character or block special devices, pipes,
46336 named pipes, sockets or any other communication method on the host
46337 system are not supported by this protocol.
46339 File I/O is not supported in non-stop mode.
46341 @node Protocol Basics
46342 @subsection Protocol Basics
46343 @cindex protocol basics, file-i/o
46345 The File-I/O protocol uses the @code{F} packet as the request as well
46346 as reply packet. Since a File-I/O system call can only occur when
46347 @value{GDBN} is waiting for a response from the continuing or stepping target,
46348 the File-I/O request is a reply that @value{GDBN} has to expect as a result
46349 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
46350 This @code{F} packet contains all information needed to allow @value{GDBN}
46351 to call the appropriate host system call:
46355 A unique identifier for the requested system call.
46358 All parameters to the system call. Pointers are given as addresses
46359 in the target memory address space. Pointers to strings are given as
46360 pointer/length pair. Numerical values are given as they are.
46361 Numerical control flags are given in a protocol-specific representation.
46365 At this point, @value{GDBN} has to perform the following actions.
46369 If the parameters include pointer values to data needed as input to a
46370 system call, @value{GDBN} requests this data from the target with a
46371 standard @code{m} packet request. This additional communication has to be
46372 expected by the target implementation and is handled as any other @code{m}
46376 @value{GDBN} translates all value from protocol representation to host
46377 representation as needed. Datatypes are coerced into the host types.
46380 @value{GDBN} calls the system call.
46383 It then coerces datatypes back to protocol representation.
46386 If the system call is expected to return data in buffer space specified
46387 by pointer parameters to the call, the data is transmitted to the
46388 target using a @code{M} or @code{X} packet. This packet has to be expected
46389 by the target implementation and is handled as any other @code{M} or @code{X}
46394 Eventually @value{GDBN} replies with another @code{F} packet which contains all
46395 necessary information for the target to continue. This at least contains
46402 @code{errno}, if has been changed by the system call.
46409 After having done the needed type and value coercion, the target continues
46410 the latest continue or step action.
46412 @node The F Request Packet
46413 @subsection The @code{F} Request Packet
46414 @cindex file-i/o request packet
46415 @cindex @code{F} request packet
46417 The @code{F} request packet has the following format:
46420 @item F@var{call-id},@var{parameter@dots{}}
46422 @var{call-id} is the identifier to indicate the host system call to be called.
46423 This is just the name of the function.
46425 @var{parameter@dots{}} are the parameters to the system call.
46426 Parameters are hexadecimal integer values, either the actual values in case
46427 of scalar datatypes, pointers to target buffer space in case of compound
46428 datatypes and unspecified memory areas, or pointer/length pairs in case
46429 of string parameters. These are appended to the @var{call-id} as a
46430 comma-delimited list. All values are transmitted in ASCII
46431 string representation, pointer/length pairs separated by a slash.
46437 @node The F Reply Packet
46438 @subsection The @code{F} Reply Packet
46439 @cindex file-i/o reply packet
46440 @cindex @code{F} reply packet
46442 The @code{F} reply packet has the following format:
46446 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
46448 @var{retcode} is the return code of the system call as hexadecimal value.
46450 @var{errno} is the @code{errno} set by the call, in protocol-specific
46452 This parameter can be omitted if the call was successful.
46454 @var{Ctrl-C flag} is only sent if the user requested a break. In this
46455 case, @var{errno} must be sent as well, even if the call was successful.
46456 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
46463 or, if the call was interrupted before the host call has been performed:
46470 assuming 4 is the protocol-specific representation of @code{EINTR}.
46475 @node The Ctrl-C Message
46476 @subsection The @samp{Ctrl-C} Message
46477 @cindex ctrl-c message, in file-i/o protocol
46479 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
46480 reply packet (@pxref{The F Reply Packet}),
46481 the target should behave as if it had
46482 gotten a break message. The meaning for the target is ``system call
46483 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
46484 (as with a break message) and return to @value{GDBN} with a @code{T02}
46487 It's important for the target to know in which
46488 state the system call was interrupted. There are two possible cases:
46492 The system call hasn't been performed on the host yet.
46495 The system call on the host has been finished.
46499 These two states can be distinguished by the target by the value of the
46500 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
46501 call hasn't been performed. This is equivalent to the @code{EINTR} handling
46502 on POSIX systems. In any other case, the target may presume that the
46503 system call has been finished --- successfully or not --- and should behave
46504 as if the break message arrived right after the system call.
46506 @value{GDBN} must behave reliably. If the system call has not been called
46507 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
46508 @code{errno} in the packet. If the system call on the host has been finished
46509 before the user requests a break, the full action must be finished by
46510 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
46511 The @code{F} packet may only be sent when either nothing has happened
46512 or the full action has been completed.
46515 @subsection Console I/O
46516 @cindex console i/o as part of file-i/o
46518 By default and if not explicitly closed by the target system, the file
46519 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
46520 on the @value{GDBN} console is handled as any other file output operation
46521 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
46522 by @value{GDBN} so that after the target read request from file descriptor
46523 0 all following typing is buffered until either one of the following
46528 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
46530 system call is treated as finished.
46533 The user presses @key{RET}. This is treated as end of input with a trailing
46537 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
46538 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
46542 If the user has typed more characters than fit in the buffer given to
46543 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
46544 either another @code{read(0, @dots{})} is requested by the target, or debugging
46545 is stopped at the user's request.
46548 @node List of Supported Calls
46549 @subsection List of Supported Calls
46550 @cindex list of supported file-i/o calls
46567 @unnumberedsubsubsec open
46568 @cindex open, file-i/o system call
46573 int open(const char *pathname, int flags);
46574 int open(const char *pathname, int flags, mode_t mode);
46578 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
46581 @var{flags} is the bitwise @code{OR} of the following values:
46585 If the file does not exist it will be created. The host
46586 rules apply as far as file ownership and time stamps
46590 When used with @code{O_CREAT}, if the file already exists it is
46591 an error and open() fails.
46594 If the file already exists and the open mode allows
46595 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
46596 truncated to zero length.
46599 The file is opened in append mode.
46602 The file is opened for reading only.
46605 The file is opened for writing only.
46608 The file is opened for reading and writing.
46612 Other bits are silently ignored.
46616 @var{mode} is the bitwise @code{OR} of the following values:
46620 User has read permission.
46623 User has write permission.
46626 Group has read permission.
46629 Group has write permission.
46632 Others have read permission.
46635 Others have write permission.
46639 Other bits are silently ignored.
46642 @item Return value:
46643 @code{open} returns the new file descriptor or -1 if an error
46650 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
46653 @var{pathname} refers to a directory.
46656 The requested access is not allowed.
46659 @var{pathname} was too long.
46662 A directory component in @var{pathname} does not exist.
46665 @var{pathname} refers to a device, pipe, named pipe or socket.
46668 @var{pathname} refers to a file on a read-only filesystem and
46669 write access was requested.
46672 @var{pathname} is an invalid pointer value.
46675 No space on device to create the file.
46678 The process already has the maximum number of files open.
46681 The limit on the total number of files open on the system
46685 The call was interrupted by the user.
46691 @unnumberedsubsubsec close
46692 @cindex close, file-i/o system call
46701 @samp{Fclose,@var{fd}}
46703 @item Return value:
46704 @code{close} returns zero on success, or -1 if an error occurred.
46710 @var{fd} isn't a valid open file descriptor.
46713 The call was interrupted by the user.
46719 @unnumberedsubsubsec read
46720 @cindex read, file-i/o system call
46725 int read(int fd, void *buf, unsigned int count);
46729 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
46731 @item Return value:
46732 On success, the number of bytes read is returned.
46733 Zero indicates end of file. If count is zero, read
46734 returns zero as well. On error, -1 is returned.
46740 @var{fd} is not a valid file descriptor or is not open for
46744 @var{bufptr} is an invalid pointer value.
46747 The call was interrupted by the user.
46753 @unnumberedsubsubsec write
46754 @cindex write, file-i/o system call
46759 int write(int fd, const void *buf, unsigned int count);
46763 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
46765 @item Return value:
46766 On success, the number of bytes written are returned.
46767 Zero indicates nothing was written. On error, -1
46774 @var{fd} is not a valid file descriptor or is not open for
46778 @var{bufptr} is an invalid pointer value.
46781 An attempt was made to write a file that exceeds the
46782 host-specific maximum file size allowed.
46785 No space on device to write the data.
46788 The call was interrupted by the user.
46794 @unnumberedsubsubsec lseek
46795 @cindex lseek, file-i/o system call
46800 long lseek (int fd, long offset, int flag);
46804 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
46806 @var{flag} is one of:
46810 The offset is set to @var{offset} bytes.
46813 The offset is set to its current location plus @var{offset}
46817 The offset is set to the size of the file plus @var{offset}
46821 @item Return value:
46822 On success, the resulting unsigned offset in bytes from
46823 the beginning of the file is returned. Otherwise, a
46824 value of -1 is returned.
46830 @var{fd} is not a valid open file descriptor.
46833 @var{fd} is associated with the @value{GDBN} console.
46836 @var{flag} is not a proper value.
46839 The call was interrupted by the user.
46845 @unnumberedsubsubsec rename
46846 @cindex rename, file-i/o system call
46851 int rename(const char *oldpath, const char *newpath);
46855 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
46857 @item Return value:
46858 On success, zero is returned. On error, -1 is returned.
46864 @var{newpath} is an existing directory, but @var{oldpath} is not a
46868 @var{newpath} is a non-empty directory.
46871 @var{oldpath} or @var{newpath} is a directory that is in use by some
46875 An attempt was made to make a directory a subdirectory
46879 A component used as a directory in @var{oldpath} or new
46880 path is not a directory. Or @var{oldpath} is a directory
46881 and @var{newpath} exists but is not a directory.
46884 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
46887 No access to the file or the path of the file.
46891 @var{oldpath} or @var{newpath} was too long.
46894 A directory component in @var{oldpath} or @var{newpath} does not exist.
46897 The file is on a read-only filesystem.
46900 The device containing the file has no room for the new
46904 The call was interrupted by the user.
46910 @unnumberedsubsubsec unlink
46911 @cindex unlink, file-i/o system call
46916 int unlink(const char *pathname);
46920 @samp{Funlink,@var{pathnameptr}/@var{len}}
46922 @item Return value:
46923 On success, zero is returned. On error, -1 is returned.
46929 No access to the file or the path of the file.
46932 The system does not allow unlinking of directories.
46935 The file @var{pathname} cannot be unlinked because it's
46936 being used by another process.
46939 @var{pathnameptr} is an invalid pointer value.
46942 @var{pathname} was too long.
46945 A directory component in @var{pathname} does not exist.
46948 A component of the path is not a directory.
46951 The file is on a read-only filesystem.
46954 The call was interrupted by the user.
46960 @unnumberedsubsubsec stat/fstat
46961 @cindex fstat, file-i/o system call
46962 @cindex stat, file-i/o system call
46967 int stat(const char *pathname, struct stat *buf);
46968 int fstat(int fd, struct stat *buf);
46972 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
46973 @samp{Ffstat,@var{fd},@var{bufptr}}
46975 @item Return value:
46976 On success, zero is returned. On error, -1 is returned.
46982 @var{fd} is not a valid open file.
46985 A directory component in @var{pathname} does not exist or the
46986 path is an empty string.
46989 A component of the path is not a directory.
46992 @var{pathnameptr} is an invalid pointer value.
46995 No access to the file or the path of the file.
46998 @var{pathname} was too long.
47001 The call was interrupted by the user.
47007 @unnumberedsubsubsec gettimeofday
47008 @cindex gettimeofday, file-i/o system call
47013 int gettimeofday(struct timeval *tv, void *tz);
47017 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
47019 @item Return value:
47020 On success, 0 is returned, -1 otherwise.
47026 @var{tz} is a non-NULL pointer.
47029 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
47035 @unnumberedsubsubsec isatty
47036 @cindex isatty, file-i/o system call
47041 int isatty(int fd);
47045 @samp{Fisatty,@var{fd}}
47047 @item Return value:
47048 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
47054 The call was interrupted by the user.
47059 Note that the @code{isatty} call is treated as a special case: it returns
47060 1 to the target if the file descriptor is attached
47061 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
47062 would require implementing @code{ioctl} and would be more complex than
47067 @unnumberedsubsubsec system
47068 @cindex system, file-i/o system call
47073 int system(const char *command);
47077 @samp{Fsystem,@var{commandptr}/@var{len}}
47079 @item Return value:
47080 If @var{len} is zero, the return value indicates whether a shell is
47081 available. A zero return value indicates a shell is not available.
47082 For non-zero @var{len}, the value returned is -1 on error and the
47083 return status of the command otherwise. Only the exit status of the
47084 command is returned, which is extracted from the host's @code{system}
47085 return value by calling @code{WEXITSTATUS(retval)}. In case
47086 @file{/bin/sh} could not be executed, 127 is returned.
47092 The call was interrupted by the user.
47097 @value{GDBN} takes over the full task of calling the necessary host calls
47098 to perform the @code{system} call. The return value of @code{system} on
47099 the host is simplified before it's returned
47100 to the target. Any termination signal information from the child process
47101 is discarded, and the return value consists
47102 entirely of the exit status of the called command.
47104 Due to security concerns, the @code{system} call is by default refused
47105 by @value{GDBN}. The user has to allow this call explicitly with the
47106 @code{set remote system-call-allowed 1} command.
47109 @item set remote system-call-allowed
47110 @kindex set remote system-call-allowed
47111 Control whether to allow the @code{system} calls in the File I/O
47112 protocol for the remote target. The default is zero (disabled).
47114 @item show remote system-call-allowed
47115 @kindex show remote system-call-allowed
47116 Show whether the @code{system} calls are allowed in the File I/O
47120 @node Protocol-specific Representation of Datatypes
47121 @subsection Protocol-specific Representation of Datatypes
47122 @cindex protocol-specific representation of datatypes, in file-i/o protocol
47125 * Integral Datatypes::
47127 * Memory Transfer::
47132 @node Integral Datatypes
47133 @unnumberedsubsubsec Integral Datatypes
47134 @cindex integral datatypes, in file-i/o protocol
47136 The integral datatypes used in the system calls are @code{int},
47137 @code{unsigned int}, @code{long}, @code{unsigned long},
47138 @code{mode_t}, and @code{time_t}.
47140 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
47141 implemented as 32 bit values in this protocol.
47143 @code{long} and @code{unsigned long} are implemented as 64 bit types.
47145 @xref{Limits}, for corresponding MIN and MAX values (similar to those
47146 in @file{limits.h}) to allow range checking on host and target.
47148 @code{time_t} datatypes are defined as seconds since the Epoch.
47150 All integral datatypes transferred as part of a memory read or write of a
47151 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
47154 @node Pointer Values
47155 @unnumberedsubsubsec Pointer Values
47156 @cindex pointer values, in file-i/o protocol
47158 Pointers to target data are transmitted as they are. An exception
47159 is made for pointers to buffers for which the length isn't
47160 transmitted as part of the function call, namely strings. Strings
47161 are transmitted as a pointer/length pair, both as hex values, e.g.@:
47168 which is a pointer to data of length 18 bytes at position 0x1aaf.
47169 The length is defined as the full string length in bytes, including
47170 the trailing null byte. For example, the string @code{"hello world"}
47171 at address 0x123456 is transmitted as
47177 @node Memory Transfer
47178 @unnumberedsubsubsec Memory Transfer
47179 @cindex memory transfer, in file-i/o protocol
47181 Structured data which is transferred using a memory read or write (for
47182 example, a @code{struct stat}) is expected to be in a protocol-specific format
47183 with all scalar multibyte datatypes being big endian. Translation to
47184 this representation needs to be done both by the target before the @code{F}
47185 packet is sent, and by @value{GDBN} before
47186 it transfers memory to the target. Transferred pointers to structured
47187 data should point to the already-coerced data at any time.
47191 @unnumberedsubsubsec struct stat
47192 @cindex struct stat, in file-i/o protocol
47194 The buffer of type @code{struct stat} used by the target and @value{GDBN}
47195 is defined as follows:
47199 unsigned int st_dev; /* device */
47200 unsigned int st_ino; /* inode */
47201 mode_t st_mode; /* protection */
47202 unsigned int st_nlink; /* number of hard links */
47203 unsigned int st_uid; /* user ID of owner */
47204 unsigned int st_gid; /* group ID of owner */
47205 unsigned int st_rdev; /* device type (if inode device) */
47206 unsigned long st_size; /* total size, in bytes */
47207 unsigned long st_blksize; /* blocksize for filesystem I/O */
47208 unsigned long st_blocks; /* number of blocks allocated */
47209 time_t st_atime; /* time of last access */
47210 time_t st_mtime; /* time of last modification */
47211 time_t st_ctime; /* time of last change */
47215 The integral datatypes conform to the definitions given in the
47216 appropriate section (see @ref{Integral Datatypes}, for details) so this
47217 structure is of size 64 bytes.
47219 The values of several fields have a restricted meaning and/or
47225 A value of 0 represents a file, 1 the console.
47228 No valid meaning for the target. Transmitted unchanged.
47231 Valid mode bits are described in @ref{Constants}. Any other
47232 bits have currently no meaning for the target.
47237 No valid meaning for the target. Transmitted unchanged.
47242 These values have a host and file system dependent
47243 accuracy. Especially on Windows hosts, the file system may not
47244 support exact timing values.
47247 The target gets a @code{struct stat} of the above representation and is
47248 responsible for coercing it to the target representation before
47251 Note that due to size differences between the host, target, and protocol
47252 representations of @code{struct stat} members, these members could eventually
47253 get truncated on the target.
47255 @node struct timeval
47256 @unnumberedsubsubsec struct timeval
47257 @cindex struct timeval, in file-i/o protocol
47259 The buffer of type @code{struct timeval} used by the File-I/O protocol
47260 is defined as follows:
47264 time_t tv_sec; /* second */
47265 long tv_usec; /* microsecond */
47269 The integral datatypes conform to the definitions given in the
47270 appropriate section (see @ref{Integral Datatypes}, for details) so this
47271 structure is of size 8 bytes.
47274 @subsection Constants
47275 @cindex constants, in file-i/o protocol
47277 The following values are used for the constants inside of the
47278 protocol. @value{GDBN} and target are responsible for translating these
47279 values before and after the call as needed.
47290 @unnumberedsubsubsec Open Flags
47291 @cindex open flags, in file-i/o protocol
47293 All values are given in hexadecimal representation.
47305 @node mode_t Values
47306 @unnumberedsubsubsec mode_t Values
47307 @cindex mode_t values, in file-i/o protocol
47309 All values are given in octal representation.
47326 @unnumberedsubsubsec Errno Values
47327 @cindex errno values, in file-i/o protocol
47329 All values are given in decimal representation.
47354 @code{EUNKNOWN} is used as a fallback error value if a host system returns
47355 any error value not in the list of supported error numbers.
47358 @unnumberedsubsubsec Lseek Flags
47359 @cindex lseek flags, in file-i/o protocol
47368 @unnumberedsubsubsec Limits
47369 @cindex limits, in file-i/o protocol
47371 All values are given in decimal representation.
47374 INT_MIN -2147483648
47376 UINT_MAX 4294967295
47377 LONG_MIN -9223372036854775808
47378 LONG_MAX 9223372036854775807
47379 ULONG_MAX 18446744073709551615
47382 @node File-I/O Examples
47383 @subsection File-I/O Examples
47384 @cindex file-i/o examples
47386 Example sequence of a write call, file descriptor 3, buffer is at target
47387 address 0x1234, 6 bytes should be written:
47390 <- @code{Fwrite,3,1234,6}
47391 @emph{request memory read from target}
47394 @emph{return "6 bytes written"}
47398 Example sequence of a read call, file descriptor 3, buffer is at target
47399 address 0x1234, 6 bytes should be read:
47402 <- @code{Fread,3,1234,6}
47403 @emph{request memory write to target}
47404 -> @code{X1234,6:XXXXXX}
47405 @emph{return "6 bytes read"}
47409 Example sequence of a read call, call fails on the host due to invalid
47410 file descriptor (@code{EBADF}):
47413 <- @code{Fread,3,1234,6}
47417 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
47421 <- @code{Fread,3,1234,6}
47426 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
47430 <- @code{Fread,3,1234,6}
47431 -> @code{X1234,6:XXXXXX}
47435 @node Library List Format
47436 @section Library List Format
47437 @cindex library list format, remote protocol
47439 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
47440 same process as your application to manage libraries. In this case,
47441 @value{GDBN} can use the loader's symbol table and normal memory
47442 operations to maintain a list of shared libraries. On other
47443 platforms, the operating system manages loaded libraries.
47444 @value{GDBN} can not retrieve the list of currently loaded libraries
47445 through memory operations, so it uses the @samp{qXfer:libraries:read}
47446 packet (@pxref{qXfer library list read}) instead. The remote stub
47447 queries the target's operating system and reports which libraries
47450 The @samp{qXfer:libraries:read} packet returns an XML document which
47451 lists loaded libraries and their offsets. Each library has an
47452 associated name and one or more segment or section base addresses,
47453 which report where the library was loaded in memory.
47455 For the common case of libraries that are fully linked binaries, the
47456 library should have a list of segments. If the target supports
47457 dynamic linking of a relocatable object file, its library XML element
47458 should instead include a list of allocated sections. The segment or
47459 section bases are start addresses, not relocation offsets; they do not
47460 depend on the library's link-time base addresses.
47462 @value{GDBN} must be linked with the Expat library to support XML
47463 library lists. @xref{Expat}.
47465 A simple memory map, with one loaded library relocated by a single
47466 offset, looks like this:
47470 <library name="/lib/libc.so.6">
47471 <segment address="0x10000000"/>
47476 Another simple memory map, with one loaded library with three
47477 allocated sections (.text, .data, .bss), looks like this:
47481 <library name="sharedlib.o">
47482 <section address="0x10000000"/>
47483 <section address="0x20000000"/>
47484 <section address="0x30000000"/>
47489 The format of a library list is described by this DTD:
47492 <!-- library-list: Root element with versioning -->
47493 <!ELEMENT library-list (library)*>
47494 <!ATTLIST library-list version CDATA #FIXED "1.0">
47495 <!ELEMENT library (segment*, section*)>
47496 <!ATTLIST library name CDATA #REQUIRED>
47497 <!ELEMENT segment EMPTY>
47498 <!ATTLIST segment address CDATA #REQUIRED>
47499 <!ELEMENT section EMPTY>
47500 <!ATTLIST section address CDATA #REQUIRED>
47503 In addition, segments and section descriptors cannot be mixed within a
47504 single library element, and you must supply at least one segment or
47505 section for each library.
47507 @node Library List Format for SVR4 Targets
47508 @section Library List Format for SVR4 Targets
47509 @cindex library list format, remote protocol
47511 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
47512 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
47513 shared libraries. Still a special library list provided by this packet is
47514 more efficient for the @value{GDBN} remote protocol.
47516 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
47517 loaded libraries and their SVR4 linker parameters. For each library on SVR4
47518 target, the following parameters are reported:
47522 @code{name}, the absolute file name from the @code{l_name} field of
47523 @code{struct link_map}.
47525 @code{lm} with address of @code{struct link_map} used for TLS
47526 (Thread Local Storage) access.
47528 @code{l_addr}, the displacement as read from the field @code{l_addr} of
47529 @code{struct link_map}. For prelinked libraries this is not an absolute
47530 memory address. It is a displacement of absolute memory address against
47531 address the file was prelinked to during the library load.
47533 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
47535 @code{lmid}, which is an identifier for a linker namespace, such as
47536 the memory address of the @code{r_debug} object that contains this
47537 namespace's load map or the namespace identifier returned by
47541 Additionally the single @code{main-lm} attribute specifies address of
47542 @code{struct link_map} used for the main executable. This parameter is used
47543 for TLS access and its presence is optional.
47545 @value{GDBN} must be linked with the Expat library to support XML
47546 SVR4 library lists. @xref{Expat}.
47548 A simple memory map, with two loaded libraries (which do not use prelink),
47552 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
47553 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
47554 l_ld="0xe4eefc" lmid="0xfffe0"/>
47555 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
47556 l_ld="0x152350" lmid="0xfffe0"/>
47557 </library-list-svr>
47560 The format of an SVR4 library list is described by this DTD:
47563 <!-- library-list-svr4: Root element with versioning -->
47564 <!ELEMENT library-list-svr4 (library)*>
47565 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
47566 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
47567 <!ELEMENT library EMPTY>
47568 <!ATTLIST library name CDATA #REQUIRED>
47569 <!ATTLIST library lm CDATA #REQUIRED>
47570 <!ATTLIST library l_addr CDATA #REQUIRED>
47571 <!ATTLIST library l_ld CDATA #REQUIRED>
47572 <!ATTLIST library lmid CDATA #IMPLIED>
47575 @node Memory Map Format
47576 @section Memory Map Format
47577 @cindex memory map format
47579 To be able to write into flash memory, @value{GDBN} needs to obtain a
47580 memory map from the target. This section describes the format of the
47583 The memory map is obtained using the @samp{qXfer:memory-map:read}
47584 (@pxref{qXfer memory map read}) packet and is an XML document that
47585 lists memory regions.
47587 @value{GDBN} must be linked with the Expat library to support XML
47588 memory maps. @xref{Expat}.
47590 The top-level structure of the document is shown below:
47593 <?xml version="1.0"?>
47594 <!DOCTYPE memory-map
47595 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
47596 "http://sourceware.org/gdb/gdb-memory-map.dtd">
47602 Each region can be either:
47607 A region of RAM starting at @var{addr} and extending for @var{length}
47611 <memory type="ram" start="@var{addr}" length="@var{length}"/>
47616 A region of read-only memory:
47619 <memory type="rom" start="@var{addr}" length="@var{length}"/>
47624 A region of flash memory, with erasure blocks @var{blocksize}
47628 <memory type="flash" start="@var{addr}" length="@var{length}">
47629 <property name="blocksize">@var{blocksize}</property>
47635 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
47636 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
47637 packets to write to addresses in such ranges.
47639 The formal DTD for memory map format is given below:
47642 <!-- ................................................... -->
47643 <!-- Memory Map XML DTD ................................ -->
47644 <!-- File: memory-map.dtd .............................. -->
47645 <!-- .................................... .............. -->
47646 <!-- memory-map.dtd -->
47647 <!-- memory-map: Root element with versioning -->
47648 <!ELEMENT memory-map (memory)*>
47649 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
47650 <!ELEMENT memory (property)*>
47651 <!-- memory: Specifies a memory region,
47652 and its type, or device. -->
47653 <!ATTLIST memory type (ram|rom|flash) #REQUIRED
47654 start CDATA #REQUIRED
47655 length CDATA #REQUIRED>
47656 <!-- property: Generic attribute tag -->
47657 <!ELEMENT property (#PCDATA | property)*>
47658 <!ATTLIST property name (blocksize) #REQUIRED>
47661 @node Thread List Format
47662 @section Thread List Format
47663 @cindex thread list format
47665 To efficiently update the list of threads and their attributes,
47666 @value{GDBN} issues the @samp{qXfer:threads:read} packet
47667 (@pxref{qXfer threads read}) and obtains the XML document with
47668 the following structure:
47671 <?xml version="1.0"?>
47673 <thread id="id" core="0" name="name">
47674 ... description ...
47679 Each @samp{thread} element must have the @samp{id} attribute that
47680 identifies the thread (@pxref{thread-id syntax}). The
47681 @samp{core} attribute, if present, specifies which processor core
47682 the thread was last executing on. The @samp{name} attribute, if
47683 present, specifies the human-readable name of the thread. The content
47684 of the of @samp{thread} element is interpreted as human-readable
47685 auxiliary information. The @samp{handle} attribute, if present,
47686 is a hex encoded representation of the thread handle.
47689 @node Traceframe Info Format
47690 @section Traceframe Info Format
47691 @cindex traceframe info format
47693 To be able to know which objects in the inferior can be examined when
47694 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
47695 memory ranges, registers and trace state variables that have been
47696 collected in a traceframe.
47698 This list is obtained using the @samp{qXfer:traceframe-info:read}
47699 (@pxref{qXfer traceframe info read}) packet and is an XML document.
47701 @value{GDBN} must be linked with the Expat library to support XML
47702 traceframe info discovery. @xref{Expat}.
47704 The top-level structure of the document is shown below:
47707 <?xml version="1.0"?>
47708 <!DOCTYPE traceframe-info
47709 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
47710 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
47716 Each traceframe block can be either:
47721 A region of collected memory starting at @var{addr} and extending for
47722 @var{length} bytes from there:
47725 <memory start="@var{addr}" length="@var{length}"/>
47729 A block indicating trace state variable numbered @var{number} has been
47733 <tvar id="@var{number}"/>
47738 The formal DTD for the traceframe info format is given below:
47741 <!ELEMENT traceframe-info (memory | tvar)* >
47742 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
47744 <!ELEMENT memory EMPTY>
47745 <!ATTLIST memory start CDATA #REQUIRED
47746 length CDATA #REQUIRED>
47748 <!ATTLIST tvar id CDATA #REQUIRED>
47751 @node Branch Trace Format
47752 @section Branch Trace Format
47753 @cindex branch trace format
47755 In order to display the branch trace of an inferior thread,
47756 @value{GDBN} needs to obtain the list of branches. This list is
47757 represented as list of sequential code blocks that are connected via
47758 branches. The code in each block has been executed sequentially.
47760 This list is obtained using the @samp{qXfer:btrace:read}
47761 (@pxref{qXfer btrace read}) packet and is an XML document.
47763 @value{GDBN} must be linked with the Expat library to support XML
47764 traceframe info discovery. @xref{Expat}.
47766 The top-level structure of the document is shown below:
47769 <?xml version="1.0"?>
47771 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
47772 "http://sourceware.org/gdb/gdb-btrace.dtd">
47781 A block of sequentially executed instructions starting at @var{begin}
47782 and ending at @var{end}:
47785 <block begin="@var{begin}" end="@var{end}"/>
47790 The formal DTD for the branch trace format is given below:
47793 <!ELEMENT btrace (block* | pt) >
47794 <!ATTLIST btrace version CDATA #FIXED "1.0">
47796 <!ELEMENT block EMPTY>
47797 <!ATTLIST block begin CDATA #REQUIRED
47798 end CDATA #REQUIRED>
47800 <!ELEMENT pt (pt-config?, raw?)>
47802 <!ELEMENT pt-config (cpu?)>
47804 <!ELEMENT cpu EMPTY>
47805 <!ATTLIST cpu vendor CDATA #REQUIRED
47806 family CDATA #REQUIRED
47807 model CDATA #REQUIRED
47808 stepping CDATA #REQUIRED>
47810 <!ELEMENT raw (#PCDATA)>
47813 @node Branch Trace Configuration Format
47814 @section Branch Trace Configuration Format
47815 @cindex branch trace configuration format
47817 For each inferior thread, @value{GDBN} can obtain the branch trace
47818 configuration using the @samp{qXfer:btrace-conf:read}
47819 (@pxref{qXfer btrace-conf read}) packet.
47821 The configuration describes the branch trace format and configuration
47822 settings for that format. The following information is described:
47826 This thread uses the @dfn{Branch Trace Store} (@acronym{BTS}) format.
47829 The size of the @acronym{BTS} ring buffer in bytes.
47832 This thread uses the @dfn{Intel Processor Trace} (@acronym{Intel
47836 The size of the @acronym{Intel PT} ring buffer in bytes.
47840 @value{GDBN} must be linked with the Expat library to support XML
47841 branch trace configuration discovery. @xref{Expat}.
47843 The formal DTD for the branch trace configuration format is given below:
47846 <!ELEMENT btrace-conf (bts?, pt?)>
47847 <!ATTLIST btrace-conf version CDATA #FIXED "1.0">
47849 <!ELEMENT bts EMPTY>
47850 <!ATTLIST bts size CDATA #IMPLIED>
47852 <!ELEMENT pt EMPTY>
47853 <!ATTLIST pt size CDATA #IMPLIED>
47856 @include agentexpr.texi
47858 @node Target Descriptions
47859 @appendix Target Descriptions
47860 @cindex target descriptions
47862 One of the challenges of using @value{GDBN} to debug embedded systems
47863 is that there are so many minor variants of each processor
47864 architecture in use. It is common practice for vendors to start with
47865 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
47866 and then make changes to adapt it to a particular market niche. Some
47867 architectures have hundreds of variants, available from dozens of
47868 vendors. This leads to a number of problems:
47872 With so many different customized processors, it is difficult for
47873 the @value{GDBN} maintainers to keep up with the changes.
47875 Since individual variants may have short lifetimes or limited
47876 audiences, it may not be worthwhile to carry information about every
47877 variant in the @value{GDBN} source tree.
47879 When @value{GDBN} does support the architecture of the embedded system
47880 at hand, the task of finding the correct architecture name to give the
47881 @command{set architecture} command can be error-prone.
47884 To address these problems, the @value{GDBN} remote protocol allows a
47885 target system to not only identify itself to @value{GDBN}, but to
47886 actually describe its own features. This lets @value{GDBN} support
47887 processor variants it has never seen before --- to the extent that the
47888 descriptions are accurate, and that @value{GDBN} understands them.
47890 @value{GDBN} must be linked with the Expat library to support XML
47891 target descriptions. @xref{Expat}.
47894 * Retrieving Descriptions:: How descriptions are fetched from a target.
47895 * Target Description Format:: The contents of a target description.
47896 * Predefined Target Types:: Standard types available for target
47898 * Enum Target Types:: How to define enum target types.
47899 * Standard Target Features:: Features @value{GDBN} knows about.
47902 @node Retrieving Descriptions
47903 @section Retrieving Descriptions
47905 Target descriptions can be read from the target automatically, or
47906 specified by the user manually. The default behavior is to read the
47907 description from the target. @value{GDBN} retrieves it via the remote
47908 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
47909 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
47910 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
47911 XML document, of the form described in @ref{Target Description
47914 Alternatively, you can specify a file to read for the target description.
47915 If a file is set, the target will not be queried. The commands to
47916 specify a file are:
47919 @cindex set tdesc filename
47920 @item set tdesc filename @var{path}
47921 Read the target description from @var{path}.
47923 @cindex unset tdesc filename
47924 @item unset tdesc filename
47925 Do not read the XML target description from a file. @value{GDBN}
47926 will use the description supplied by the current target.
47928 @cindex show tdesc filename
47929 @item show tdesc filename
47930 Show the filename to read for a target description, if any.
47934 @node Target Description Format
47935 @section Target Description Format
47936 @cindex target descriptions, XML format
47938 A target description annex is an @uref{http://www.w3.org/XML/, XML}
47939 document which complies with the Document Type Definition provided in
47940 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
47941 means you can use generally available tools like @command{xmllint} to
47942 check that your feature descriptions are well-formed and valid.
47943 However, to help people unfamiliar with XML write descriptions for
47944 their targets, we also describe the grammar here.
47946 Target descriptions can identify the architecture of the remote target
47947 and (for some architectures) provide information about custom register
47948 sets. They can also identify the OS ABI of the remote target.
47949 @value{GDBN} can use this information to autoconfigure for your
47950 target, or to warn you if you connect to an unsupported target.
47952 Here is a simple target description:
47955 <target version="1.0">
47956 <architecture>i386:x86-64</architecture>
47961 This minimal description only says that the target uses
47962 the x86-64 architecture.
47964 A target description has the following overall form, with [ ] marking
47965 optional elements and @dots{} marking repeatable elements. The elements
47966 are explained further below.
47969 <?xml version="1.0"?>
47970 <!DOCTYPE target SYSTEM "gdb-target.dtd">
47971 <target version="1.0">
47972 @r{[}@var{architecture}@r{]}
47973 @r{[}@var{osabi}@r{]}
47974 @r{[}@var{compatible}@r{]}
47975 @r{[}@var{feature}@dots{}@r{]}
47980 The description is generally insensitive to whitespace and line
47981 breaks, under the usual common-sense rules. The XML version
47982 declaration and document type declaration can generally be omitted
47983 (@value{GDBN} does not require them), but specifying them may be
47984 useful for XML validation tools. The @samp{version} attribute for
47985 @samp{<target>} may also be omitted, but we recommend
47986 including it; if future versions of @value{GDBN} use an incompatible
47987 revision of @file{gdb-target.dtd}, they will detect and report
47988 the version mismatch.
47990 @subsection Inclusion
47991 @cindex target descriptions, inclusion
47994 @cindex <xi:include>
47997 It can sometimes be valuable to split a target description up into
47998 several different annexes, either for organizational purposes, or to
47999 share files between different possible target descriptions. You can
48000 divide a description into multiple files by replacing any element of
48001 the target description with an inclusion directive of the form:
48004 <xi:include href="@var{document}"/>
48008 When @value{GDBN} encounters an element of this form, it will retrieve
48009 the named XML @var{document}, and replace the inclusion directive with
48010 the contents of that document. If the current description was read
48011 using @samp{qXfer}, then so will be the included document;
48012 @var{document} will be interpreted as the name of an annex. If the
48013 current description was read from a file, @value{GDBN} will look for
48014 @var{document} as a file in the same directory where it found the
48015 original description.
48017 @subsection Architecture
48018 @cindex <architecture>
48020 An @samp{<architecture>} element has this form:
48023 <architecture>@var{arch}</architecture>
48026 @var{arch} is one of the architectures from the set accepted by
48027 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
48030 @cindex @code{<osabi>}
48032 This optional field was introduced in @value{GDBN} version 7.0.
48033 Previous versions of @value{GDBN} ignore it.
48035 An @samp{<osabi>} element has this form:
48038 <osabi>@var{abi-name}</osabi>
48041 @var{abi-name} is an OS ABI name from the same selection accepted by
48042 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
48044 @subsection Compatible Architecture
48045 @cindex @code{<compatible>}
48047 This optional field was introduced in @value{GDBN} version 7.0.
48048 Previous versions of @value{GDBN} ignore it.
48050 A @samp{<compatible>} element has this form:
48053 <compatible>@var{arch}</compatible>
48056 @var{arch} is one of the architectures from the set accepted by
48057 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
48059 A @samp{<compatible>} element is used to specify that the target
48060 is able to run binaries in some other than the main target architecture
48061 given by the @samp{<architecture>} element. For example, on the
48062 Cell Broadband Engine, the main architecture is @code{powerpc:common}
48063 or @code{powerpc:common64}, but the system is able to run binaries
48064 in the @code{spu} architecture as well. The way to describe this
48065 capability with @samp{<compatible>} is as follows:
48068 <architecture>powerpc:common</architecture>
48069 <compatible>spu</compatible>
48072 @subsection Features
48075 Each @samp{<feature>} describes some logical portion of the target
48076 system. Features are currently used to describe available CPU
48077 registers and the types of their contents. A @samp{<feature>} element
48081 <feature name="@var{name}">
48082 @r{[}@var{type}@dots{}@r{]}
48088 Each feature's name should be unique within the description. The name
48089 of a feature does not matter unless @value{GDBN} has some special
48090 knowledge of the contents of that feature; if it does, the feature
48091 should have its standard name. @xref{Standard Target Features}.
48095 Any register's value is a collection of bits which @value{GDBN} must
48096 interpret. The default interpretation is a two's complement integer,
48097 but other types can be requested by name in the register description.
48098 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
48099 Target Types}), and the description can define additional composite
48102 Each type element must have an @samp{id} attribute, which gives
48103 a unique (within the containing @samp{<feature>}) name to the type.
48104 Types must be defined before they are used.
48107 Some targets offer vector registers, which can be treated as arrays
48108 of scalar elements. These types are written as @samp{<vector>} elements,
48109 specifying the array element type, @var{type}, and the number of elements,
48113 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
48117 If a register's value is usefully viewed in multiple ways, define it
48118 with a union type containing the useful representations. The
48119 @samp{<union>} element contains one or more @samp{<field>} elements,
48120 each of which has a @var{name} and a @var{type}:
48123 <union id="@var{id}">
48124 <field name="@var{name}" type="@var{type}"/>
48131 If a register's value is composed from several separate values, define
48132 it with either a structure type or a flags type.
48133 A flags type may only contain bitfields.
48134 A structure type may either contain only bitfields or contain no bitfields.
48135 If the value contains only bitfields, its total size in bytes must be
48138 Non-bitfield values have a @var{name} and @var{type}.
48141 <struct id="@var{id}">
48142 <field name="@var{name}" type="@var{type}"/>
48147 Both @var{name} and @var{type} values are required.
48148 No implicit padding is added.
48150 Bitfield values have a @var{name}, @var{start}, @var{end} and @var{type}.
48153 <struct id="@var{id}" size="@var{size}">
48154 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
48160 <flags id="@var{id}" size="@var{size}">
48161 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
48166 The @var{name} value is required.
48167 Bitfield values may be named with the empty string, @samp{""},
48168 in which case the field is ``filler'' and its value is not printed.
48169 Not all bits need to be specified, so ``filler'' fields are optional.
48171 The @var{start} and @var{end} values are required, and @var{type}
48173 The field's @var{start} must be less than or equal to its @var{end},
48174 and zero represents the least significant bit.
48176 The default value of @var{type} is @code{bool} for single bit fields,
48177 and an unsigned integer otherwise.
48179 Which to choose? Structures or flags?
48181 Registers defined with @samp{flags} have these advantages over
48182 defining them with @samp{struct}:
48186 Arithmetic may be performed on them as if they were integers.
48188 They are printed in a more readable fashion.
48191 Registers defined with @samp{struct} have one advantage over
48192 defining them with @samp{flags}:
48196 One can fetch individual fields like in @samp{C}.
48199 (@value{GDBP}) print $my_struct_reg.field3
48205 @subsection Registers
48208 Each register is represented as an element with this form:
48211 <reg name="@var{name}"
48212 bitsize="@var{size}"
48213 @r{[}regnum="@var{num}"@r{]}
48214 @r{[}save-restore="@var{save-restore}"@r{]}
48215 @r{[}type="@var{type}"@r{]}
48216 @r{[}group="@var{group}"@r{]}/>
48220 The components are as follows:
48225 The register's name; it must be unique within the target description.
48228 The register's size, in bits.
48231 The register's number. If omitted, a register's number is one greater
48232 than that of the previous register (either in the current feature or in
48233 a preceding feature); the first register in the target description
48234 defaults to zero. This register number is used to read or write
48235 the register; e.g.@: it is used in the remote @code{p} and @code{P}
48236 packets, and registers appear in the @code{g} and @code{G} packets
48237 in order of increasing register number.
48240 Whether the register should be preserved across inferior function
48241 calls; this must be either @code{yes} or @code{no}. The default is
48242 @code{yes}, which is appropriate for most registers except for
48243 some system control registers; this is not related to the target's
48247 The type of the register. It may be a predefined type, a type
48248 defined in the current feature, or one of the special types @code{int}
48249 and @code{float}. @code{int} is an integer type of the correct size
48250 for @var{bitsize}, and @code{float} is a floating point type (in the
48251 architecture's normal floating point format) of the correct size for
48252 @var{bitsize}. The default is @code{int}.
48255 The register group to which this register belongs. It can be one of the
48256 standard register groups @code{general}, @code{float}, @code{vector} or an
48257 arbitrary string. Group names should be limited to alphanumeric characters.
48258 If a group name is made up of multiple words the words may be separated by
48259 hyphens; e.g.@: @code{special-group} or @code{ultra-special-group}. If no
48260 @var{group} is specified, @value{GDBN} will not display the register in
48261 @code{info registers}.
48265 @node Predefined Target Types
48266 @section Predefined Target Types
48267 @cindex target descriptions, predefined types
48269 Type definitions in the self-description can build up composite types
48270 from basic building blocks, but can not define fundamental types. Instead,
48271 standard identifiers are provided by @value{GDBN} for the fundamental
48272 types. The currently supported types are:
48277 Boolean type, occupying a single bit.
48285 Signed integer types holding the specified number of bits.
48293 Unsigned integer types holding the specified number of bits.
48297 Pointers to unspecified code and data. The program counter and
48298 any dedicated return address register may be marked as code
48299 pointers; printing a code pointer converts it into a symbolic
48300 address. The stack pointer and any dedicated address registers
48301 may be marked as data pointers.
48304 Half precision IEEE floating point.
48307 Single precision IEEE floating point.
48310 Double precision IEEE floating point.
48313 The 16-bit @dfn{brain floating point} format used e.g.@: by x86 and ARM.
48316 The 12-byte extended precision format used by ARM FPA registers.
48319 The 10-byte extended precision format used by x87 registers.
48322 32bit @sc{eflags} register used by x86.
48325 32bit @sc{mxcsr} register used by x86.
48329 @node Enum Target Types
48330 @section Enum Target Types
48331 @cindex target descriptions, enum types
48333 Enum target types are useful in @samp{struct} and @samp{flags}
48334 register descriptions. @xref{Target Description Format}.
48336 Enum types have a name, size and a list of name/value pairs.
48339 <enum id="@var{id}" size="@var{size}">
48340 <evalue name="@var{name}" value="@var{value}"/>
48345 Enums must be defined before they are used.
48348 <enum id="levels_type" size="4">
48349 <evalue name="low" value="0"/>
48350 <evalue name="high" value="1"/>
48352 <flags id="flags_type" size="4">
48353 <field name="X" start="0"/>
48354 <field name="LEVEL" start="1" end="1" type="levels_type"/>
48356 <reg name="flags" bitsize="32" type="flags_type"/>
48359 Given that description, a value of 3 for the @samp{flags} register
48360 would be printed as:
48363 (@value{GDBP}) info register flags
48364 flags 0x3 [ X LEVEL=high ]
48367 @node Standard Target Features
48368 @section Standard Target Features
48369 @cindex target descriptions, standard features
48371 A target description must contain either no registers or all the
48372 target's registers. If the description contains no registers, then
48373 @value{GDBN} will assume a default register layout, selected based on
48374 the architecture. If the description contains any registers, the
48375 default layout will not be used; the standard registers must be
48376 described in the target description, in such a way that @value{GDBN}
48377 can recognize them.
48379 This is accomplished by giving specific names to feature elements
48380 which contain standard registers. @value{GDBN} will look for features
48381 with those names and verify that they contain the expected registers;
48382 if any known feature is missing required registers, or if any required
48383 feature is missing, @value{GDBN} will reject the target
48384 description. You can add additional registers to any of the
48385 standard features --- @value{GDBN} will display them just as if
48386 they were added to an unrecognized feature.
48388 This section lists the known features and their expected contents.
48389 Sample XML documents for these features are included in the
48390 @value{GDBN} source tree, in the directory @file{gdb/features}.
48392 Names recognized by @value{GDBN} should include the name of the
48393 company or organization which selected the name, and the overall
48394 architecture to which the feature applies; so e.g.@: the feature
48395 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
48397 The names of registers are not case sensitive for the purpose
48398 of recognizing standard features, but @value{GDBN} will only display
48399 registers using the capitalization used in the description.
48402 * AArch64 Features::
48406 * LoongArch Features::
48407 * MicroBlaze Features::
48411 * Nios II Features::
48412 * OpenRISC 1000 Features::
48413 * PowerPC Features::
48414 * RISC-V Features::
48416 * S/390 and System z Features::
48422 @node AArch64 Features
48423 @subsection AArch64 Features
48424 @cindex target descriptions, AArch64 features
48426 @subsubsection AArch64 core registers feature
48428 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
48429 targets. It must contain the following:
48433 @samp{x0} through @samp{x30}, the general purpose registers, with size of
48434 64 bits. Register @samp{x30} is also known as the @dfn{link register},
48437 @samp{sp}, the stack pointer register or @samp{x31}. It is 64 bits in size and
48438 has a type of @samp{data_ptr}.
48440 @samp{pc}, the program counter register. It is 64 bits in size and has a type
48441 of @samp{code_ptr}.
48443 @samp{cpsr}, the current program status register. It is 32 bits in size
48444 and has a custom flags type.
48447 The semantics of the individual flags and fields in @samp{cpsr} can change as
48448 new architectural features are added. The current layout can be found in the
48449 aarch64-core.xml file.
48451 Extra registers are allowed in this feature, but they will not affect
48454 @subsubsection AArch64 floating-point registers feature
48456 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
48457 it must contain the following registers:
48461 @samp{v0} through @samp{v31}, the vector registers with size of 128 bits. The
48462 type is a custom vector type.
48464 @samp{fpsr}, the floating-point status register. It is 32 bits in size and has
48465 a custom flags type.
48467 @samp{fpcr}, the floating-point control register. It is 32 bits in size and has
48468 a custom flags type.
48471 The semantics of the individual flags and fields in @samp{fpsr} and @samp{fpcr}
48472 can change as new architectural features are added.
48474 The types for the vector registers, @samp{fpsr} and @samp{fpcr} registers can
48475 be found in the aarch64-fpu.xml file.
48477 Extra registers are allowed in this feature, but they will not affect
48480 @subsubsection AArch64 SVE registers feature
48482 The @samp{org.gnu.gdb.aarch64.sve} feature is optional. If present,
48483 it means the target supports the Scalable Vector Extension and must contain
48484 the following registers:
48488 @samp{z0} through @samp{z31}, the scalable vector registers. Their sizes are
48489 variable and a multiple of 128 bits up to a maximum of 2048 bit. Their type is
48490 a custom union type that helps visualize different sizes of sub-vectors.
48492 @samp{fpsr}, the floating-point status register. It is 32 bits in size and has
48493 a custom flags type.
48495 @samp{fpcr}, the floating-point control register. It is 32 bits in size and has
48496 a custom flags type.
48498 @samp{p0} through @samp{p15}, the predicate registers. Their sizes are
48499 variable, based on the current vector length, and a multiple of
48500 16 bits. Their types are a custom union to help visualize sub-elements.
48502 @samp{ffr}, the First Fault register. It has a variable size based on the
48503 current vector length and is a multiple of 16 bits. The type is the same as
48504 the predicate registers.
48506 @samp{vg}, the vector granule. It represents the number of 64 bits chunks in
48507 a @samp{z} register. It is closely associated with the current vector
48508 length. It has a type of @samp{int}.
48511 When @value{GDBN} sees the SVE feature, it will assume the Scalable Vector
48512 Extension is supported, and will adjust the sizes of the @samp{z}, @samp{p}
48513 and @samp{ffr} registers accordingly, based on the value of @samp{vg}.
48515 @value{GDBN} will also create pseudo-registers equivalent to the @samp{v}
48516 vector registers from the @samp{org.gnu.gdb.aarch64.fpu} feature.
48518 The first 128 bits of the @samp{z} registers overlap the 128 bits of the
48519 @samp{v} registers, so changing one will trigger a change to the other.
48521 For the types of the @samp{z}, @samp{p} and @samp{ffr} registers, please
48522 check the aarch64-sve.c file. No XML file is available for this feature
48523 because it is dynamically generated based on the current vector length.
48525 The semantics of the individual flags and fields in @samp{fpsr} and @samp{fpcr}
48526 can change as new architectural features are added.
48528 The types for the @samp{fpsr} and @samp{fpcr} registers can be found in the
48529 aarch64-sve.c file, and should match what is described in aarch64-fpu.xml.
48531 Extra registers are allowed in this feature, but they will not affect
48534 @subsubsection AArch64 Pointer Authentication registers feature
48536 The @samp{org.gnu.gdb.aarch64.pauth} optional feature was introduced so
48537 @value{GDBN} could detect support for the Pointer Authentication
48538 extension. If present, it must contain one of two possible register sets.
48540 Pointer Authentication masks for user-mode:
48544 @samp{pauth_dmask}, the user-mode pointer authentication mask for data
48545 pointers. It is 64 bits in size.
48547 @samp{pauth_cmask}, the user-mode pointer authentication mask for code
48548 pointers. It is 64 bits in size.
48551 Pointer Authentication masks for user-mode and kernel-mode:
48555 @samp{pauth_dmask}, the user-mode pointer authentication mask for data
48556 pointers. It is 64 bits in size.
48558 @samp{pauth_cmask}, the user-mode pointer authentication mask for code
48559 pointers. It is 64 bits in size.
48561 @samp{pauth_dmask_high}, the kernel-mode pointer authentication mask for
48562 data pointers. It is 64 bits in size.
48564 @samp{pauth_cmask_high}, the kernel-mode pointer authentication mask for
48565 code pointers. It is 64 bits in size.
48568 If @value{GDBN} sees any of the two sets of registers in this feature, it will
48569 assume the target is capable of signing pointers. If so, @value{GDBN} will
48570 decorate backtraces with a @samp{[PAC]} marker alongside a function that
48571 has a signed link register value that needs to be unmasked/decoded.
48573 @value{GDBN} will also use the masks to remove non-address bits from pointers.
48575 Extra registers are allowed in this feature, but they will not affect
48578 Please note the @samp{org.gnu.gdb.aarch64.pauth} feature string is deprecated
48579 and must only be used for backwards compatibility with older releases of
48580 @value{GDBN} and @command{gdbserver}. Targets that support Pointer
48581 Authentication must advertise such capability by using the
48582 @samp{org.gnu.gdb.aarch64.pauth_v2} feature string instead.
48584 The @samp{org.gnu.gdb.aarch64.pauth_v2} feature has the exact same contents
48585 as feature @samp{org.gnu.gdb.aarch64.pauth}.
48587 The reason for having feature @samp{org.gnu.gdb.aarch64.pauth_v2} is a bug in
48588 previous versions of @value{GDBN} (versions 9, 10, 11 and 12). This bug
48589 caused @value{GDBN} to crash whenever the target reported support for Pointer
48590 Authentication (using feature string @samp{org.gnu.gdb.aarch64.pauth}) and also
48591 reported additional system registers that were not accounted for by
48592 @value{GDBN}. This is more common when using emulators and on bare-metal
48593 debugging scenarios.
48595 It can also happen if a newer gdbserver is used with an old @value{GDBN} that
48596 has the bug. In such a case, the newer gdbserver might report Pointer
48597 Authentication support via the @samp{org.gnu.gdb.aarch64.pauth} feature string
48598 and also report additional registers the older @value{GDBN} does not know
48599 about, potentially leading to a crash.
48601 @subsubsection AArch64 TLS registers feature
48603 The @samp{org.gnu.gdb.aarch64.tls} optional feature was introduced to expose
48604 the TLS registers to @value{GDBN}. If present, it must contain either one
48605 of the following register sets.
48611 @samp{tpidr}, the software thread id register. It is 64 bits in size and has a
48612 type of @samp{data_ptr}.
48615 Both @samp{tpidr} and @samp{tpidr2}.
48619 @samp{tpidr}, the software thread id register. It is 64 bits in size and has a
48620 type of @samp{data_ptr}.
48622 @samp{tpidr2}, the second software thread id register. It is 64 bits in size
48623 and has a type of @samp{data_ptr}. It may be used in the future alongside
48624 the Scalable Matrix Extension for a lazy restore scheme.
48627 If @value{GDBN} sees this feature, it will attempt to find one of the
48628 variations of the register set. If @samp{tpidr2} is available,
48629 @value{GDBN} may act on it to display additional data in the future.
48631 There is no XML for this feature as the presence of @samp{tpidr2} is
48632 determined dynamically at runtime.
48634 Extra registers are allowed in this feature, but they will not affect
48637 @subsubsection AArch64 MTE registers feature
48639 The @samp{org.gnu.gdb.aarch64.mte} optional feature was introduced so
48640 @value{GDBN} could detect support for the Memory Tagging Extension and
48641 control memory tagging settings. If present, this feature must have the
48642 following register:
48646 @samp{tag_ctl}, the tag control register. It is 64 bits in size and has a type
48650 Memory Tagging detection is done via a runtime check though, so the presence
48651 of this feature and register is not enough to enable memory tagging support.
48653 This restriction may be lifted in the future.
48655 Extra registers are allowed in this feature, but they will not affect
48658 @subsubsection AArch64 SME registers feature
48660 The @samp{org.gnu.gdb.aarch64.sme} feature is optional. If present,
48661 it should contain registers @code{ZA}, @code{SVG} and @code{SVCR}.
48662 @xref{AArch64 SME}.
48667 @code{ZA} is a register represented by a vector of @var{svl}x@var{svl}
48671 @code{SVG} is a 64-bit register containing the value of @var{svg}. @xref{svg}.
48674 @code{SVCR} is a 64-bit status pseudo-register with two valid bits. Bit 0
48675 (@sc{sm}) shows whether the streaming @acronym{SVE} mode is enabled or disabled.
48676 Bit 1 (@sc{ZA}) shows whether the @code{ZA} register state is active (in use) or
48678 @xref{aarch64 sme svcr}.
48680 The rest of the unused bits of the @code{SVCR} pseudo-register is undefined
48681 and reserved. Such bits should not be used and may be defined by future
48682 extensions of the architecture.
48686 Extra registers are allowed in this feature, but they will not affect
48689 The @samp{org.gnu.gdb.aarch64.sme} feature is required when the target also
48690 reports support for the @samp{org.gnu.gdb.aarch64.sme2} feature.
48692 @subsubsection AArch64 SME2 registers feature
48694 The @samp{org.gnu.gdb.aarch64.sme2} feature is optional. If present,
48695 then the @samp{org.gnu.gdb.aarch64.sme} feature must also be present. The
48696 @samp{org.gnu.gdb.aarch64.sme2} feature should contain the following:
48697 @xref{AArch64 SME2}.
48702 @code{ZT0} is a register of 512 bits (64 bytes). It is defined as a vector
48707 Extra registers are allowed in this feature, but they will not affect
48711 @subsection ARC Features
48712 @cindex target descriptions, ARC Features
48714 ARC processors are so configurable that even core registers and their numbers
48715 are not predetermined completely. Moreover, @emph{flags} and @emph{PC}
48716 registers, which are important to @value{GDBN}, are not ``core'' registers in
48717 ARC. Therefore, there are two features that their presence is mandatory:
48718 @samp{org.gnu.gdb.arc.core} and @samp{org.gnu.gdb.arc.aux}.
48720 The @samp{org.gnu.gdb.arc.core} feature is required for all targets. It must
48725 @samp{r0} through @samp{r25} for normal register file targets.
48727 @samp{r0} through @samp{r3}, and @samp{r10} through @samp{r15} for reduced
48728 register file targets.
48730 @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}@footnote{Not necessary for ARCv1.},
48731 @samp{blink}, @samp{lp_count}, @samp{pcl}.
48734 In case of an ARCompact target (ARCv1 ISA), the @samp{org.gnu.gdb.arc.core}
48735 feature may contain registers @samp{ilink1} and @samp{ilink2}. While in case
48736 of ARC EM and ARC HS targets (ARCv2 ISA), register @samp{ilink} may be present.
48737 The difference between ARCv1 and ARCv2 is the naming of registers @emph{29th}
48738 and @emph{30th}. They are called @samp{ilink1} and @samp{ilink2} for ARCv1 and
48739 are optional. For ARCv2, they are called @samp{ilink} and @samp{r30} and only
48740 @samp{ilink} is optional. The optionality of @samp{ilink*} registers is
48741 because of their inaccessibility during user space debugging sessions.
48743 Extension core registers @samp{r32} through @samp{r59} are optional and their
48744 existence depends on the configuration. When debugging GNU/Linux applications,
48745 i.e.@: user space debugging, these core registers are not available.
48747 The @samp{org.gnu.gdb.arc.aux} feature is required for all ARC targets. Here
48748 is the list of registers pertinent to this feature:
48752 mandatory: @samp{pc} and @samp{status32}.
48754 optional: @samp{lp_start}, @samp{lp_end}, and @samp{bta}.
48758 @subsection ARM Features
48759 @cindex target descriptions, ARM features
48761 @subsubsection Core register set for non-M-profile
48763 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
48764 ARM targets. It must contain the following registers:
48768 @samp{r0} through @samp{r12}. The general purpose registers. They are 32 bits
48769 in size and have a type of @samp{uint32}.
48771 @samp{sp}, the stack pointer register, also known as @samp{r13}. It is 32 bits
48772 in size and has a type of @samp{data_ptr}.
48774 @samp{lr}, the link register. It is 32 bits in size.
48776 @samp{pc}, the program counter register. It is 32 bit in size and of type
48779 @samp{cpsr}, the current program status register containing all the status
48780 bits. It is 32 bits in size. Historically this register was hardwired to
48781 number 25, but debugging stubs that report XML do not need to use this number
48785 Extra registers are allowed in this feature, but they will not affect
48788 @subsubsection Core register set for M-profile
48790 For M-profile targets (e.g.@: Cortex-M3), the @samp{org.gnu.gdb.arm.core}
48791 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}, and it is a required
48792 feature. It must contain the following registers:
48796 @samp{r0} through @samp{r12}, the general purpose registers. They have a size
48797 of 32 bits and a type of @samp{uint32}.
48799 @samp{sp}, the stack pointer register, also known as @samp{r13}. It has a size
48800 of 32 bits and a type of @samp{data_ptr}.
48802 @samp{lr}, the link register. It has a size of 32 bits.
48804 @samp{pc}, the program counter register. It has a size of 32 bits and a type
48805 of @samp{code_ptr}.
48807 @samp{xpsr}, the program status register containing all the status
48808 bits. It has a size of 32 bits. Historically this register was hardwired to
48809 number 25, but debugging stubs that report XML do not need to use this number
48813 Upon seeing this feature, @value{GDBN} will acknowledge that it is dealing
48814 with an M-profile target. This means @value{GDBN} will use hooks and
48815 configurations that are meaningful to M-profiles.
48817 Extra registers are allowed in this feature, but they will not affect
48820 @subsubsection FPA registers feature (obsolete)
48822 The @samp{org.gnu.gdb.arm.fpa} feature is obsolete and should not be
48823 advertised by debugging stubs anymore. It used to advertise registers for
48824 the old FPA architecture that has long been discontinued in toolchains.
48826 It is kept in @value{GDBN} for backward compatibility purposes so older
48827 debugging stubs that don't support XML target descriptions still work
48828 correctly. One such example is the KGDB debugging stub from
48829 Linux or BSD kernels.
48831 The description below is for historical purposes only. This feature
48832 used to contain the following registers:
48836 @samp{f0} through @samp{f8}. The floating point registers. They are 96 bits
48837 in size and of type @samp{arm_fpa_ext}. @samp{f0} is pinned to register
48840 @samp{fps}, the status register. It has a size of 32 bits.
48843 @subsubsection M-profile Vector Extension (MVE)
48845 Also known as Helium, the M-profile Vector Extension is advertised via the
48846 optional @samp{org.gnu.gdb.arm.m-profile-mve} feature.
48848 It must contain the following:
48852 @samp{vpr}, the vector predication status and control register. It is 32 bits
48853 in size and has a custom flags type. The flags type is laid out in a way that
48854 exposes the @samp{P0} field from bits 0 to 15, the @samp{MASK01} field from
48855 bits 16 to 19 and the @samp{MASK23} field from bits 20 to 23.
48857 Bits 24 through 31 are reserved.
48860 When this feature is available, @value{GDBN} will synthesize the @samp{p0}
48861 pseudo-register from @samp{vpr} contents.
48863 This feature must only be advertised if the target is M-profile. Advertising
48864 this feature for targets that are not M-profile may cause @value{GDBN} to
48865 assume the target is M-profile when it isn't.
48867 If the @samp{org.gnu.gdb.arm.vfp} feature is available alongside the
48868 @samp{org.gnu.gdb.arm.m-profile-mve} feature, @value{GDBN} will
48869 synthesize the @samp{q} pseudo-registers from @samp{d} register
48872 Extra registers are allowed in this feature, but they will not affect
48875 @subsubsection XScale iwMMXt feature
48877 The XScale @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
48878 it must contain the following:
48882 @samp{wR0} through @samp{wR15}, registers with size 64 bits and a custom type
48883 @samp{iwmmxt_vec64i}. @samp{iwmmxt_vec64i} is a union of four other
48884 types: @samp{uint64}, a 2-element vector of @samp{uint32}, a 4-element
48885 vector of @samp{uint16} and a 8-element vector of @samp{uint8}.
48887 @samp{wCGR0} through @samp{wCGR3}, registers with size 32 bits and
48891 The following registers are optional:
48895 @samp{wCID}, register with size of 32 bits and type @samp{int}.
48897 @samp{wCon}, register with size 32 bits and type @samp{int}.
48899 @samp{wCSSF}, register with size 32 bits and type @samp{int}.
48901 @samp{wCASF}, register with size 32 bit and type @samp{int}.
48904 This feature should only be reported if the target is XScale.
48906 Extra registers are allowed in this feature, but they will not affect
48909 @subsubsection Vector Floating-Point (VFP) feature
48911 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
48912 should contain one of two possible sets of values depending on whether
48913 VFP version 2 or VFP version 3 is in use.
48919 @samp{d0} through @samp{d15}. The double-precision registers. They are
48920 64 bits in size and have type @samp{ieee_double}.
48922 @samp{fpscr}, the floating-point status and control register. It has a size
48923 of 32 bits and a type of @samp{int}.
48930 @samp{d0} through @samp{d31}. The double-precision registers. They are
48931 64 bits in size and have type @samp{ieee_double}.
48933 @samp{fpscr}, the floating-point status and control register. It has a size
48934 of 32 bits and a type of @samp{int}.
48937 If this feature is available, @value{GDBN} will synthesize the
48938 single-precision floating-point registers from halves of the double-precision
48939 registers as pseudo-registers.
48941 Extra registers are allowed in this feature, but they will not affect
48944 @subsubsection NEON architecture feature
48946 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
48947 need to contain registers; it instructs @value{GDBN} to display the
48948 VFP double-precision registers as vectors and to synthesize the
48949 quad-precision registers from pairs of double-precision registers.
48950 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
48951 be present and include 32 double-precision registers.
48953 Extra registers are allowed in this feature, but they will not affect
48956 @subsubsection M-profile Pointer Authentication and Branch Target Identification feature
48958 The @samp{org.gnu.gdb.arm.m-profile-pacbti} feature is optional, and
48959 acknowledges support for the ARMv8.1-m PACBTI extensions.
48961 This feature doesn't contain any required registers, and it only serves as a
48962 hint to @value{GDBN} that the debugging stub supports the ARMv8.1-m PACBTI
48965 When @value{GDBN} sees this feature, it will track return address signing
48966 states and will decorate backtraces using the [PAC] marker, similar to
48967 AArch64's PAC extension.
48968 @xref{AArch64 PAC}.
48970 Extra registers are allowed in this feature, but they will not affect
48973 @subsubsection M-profile system registers feature
48975 The @samp{org.gnu.gdb.arm.m-system} optional feature was introduced as a way to
48976 inform @value{GDBN} about additional system registers.
48978 At the moment this feature must contain the following:
48982 @samp{msp}, the main stack pointer register. It is 32 bits in size with
48983 type @samp{data_ptr}.
48985 @samp{psp}, the process stack pointer register. It is 32 bits in size with
48986 type @samp{data_ptr}.
48989 This feature must only be advertised for M-profile targets. When @value{GDBN}
48990 sees this feature, it will attempt to track the values of @samp{msp} and
48991 @samp{psp} across frames.
48993 Extra registers are allowed in this feature, but they will not affect
48996 @subsubsection M-profile Security Extensions feature
48998 The @samp{org.gnu.gdb.arm.secext} optional feature was introduced so
48999 @value{GDBN} could better support the switching of stack pointers and
49000 secure states in the Security Extensions.
49002 At the moment this feature must contain the following:
49006 @samp{msp_ns}, the main stack pointer register (non-secure state). It is
49007 32 bits in size with type @samp{data_ptr}.
49009 @samp{psp_ns}, the process stack pointer register (non-secure state). It is
49010 32 bits in size with type @samp{data_ptr}.
49012 @samp{msp_s}, the main stack pointer register (secure state). It is 32 bits
49013 in size with type @samp{data_ptr}.
49015 @samp{psp_s}, the process stack pointer register (secure state). It is 32 bits
49016 in size with type @samp{data_ptr}.
49019 When @value{GDBN} sees this feature, it will attempt to track the values of
49020 all 4 stack pointers across secure state transitions, potentially improving
49021 unwinding when applications switch between security states.
49023 Extra registers are allowed in this feature, but they will not affect
49026 @subsubsection TLS registers feature
49028 The optional @samp{org.gnu.gdb.arm.tls} feature contains TLS registers.
49030 Currently it contains the following:
49034 @samp{tpidruro}, the user read-only thread id register. It is 32 bits in size
49035 and has type @samp{data_ptr}.
49038 At the moment @value{GDBN} looks for this feature, but doesn't do anything
49039 with it other than displaying it.
49041 Extra registers are allowed in this feature, but they will not affect
49044 @node i386 Features
49045 @subsection i386 Features
49046 @cindex target descriptions, i386 features
49048 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
49049 targets. It should describe the following registers:
49053 @samp{eax} through @samp{edi} plus @samp{eip} for i386
49055 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
49057 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
49058 @samp{fs}, @samp{gs}
49060 @samp{st0} through @samp{st7}
49062 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
49063 @samp{foseg}, @samp{fooff} and @samp{fop}
49066 The register sets may be different, depending on the target.
49068 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
49069 describe registers:
49073 @samp{xmm0} through @samp{xmm7} for i386
49075 @samp{xmm0} through @samp{xmm15} for amd64
49080 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
49081 @samp{org.gnu.gdb.i386.sse} feature. It should
49082 describe the upper 128 bits of @sc{ymm} registers:
49086 @samp{ymm0h} through @samp{ymm7h} for i386
49088 @samp{ymm0h} through @samp{ymm15h} for amd64
49091 The @samp{org.gnu.gdb.i386.mpx} is an optional feature representing Intel
49092 Memory Protection Extension (MPX). It should describe the following registers:
49096 @samp{bnd0raw} through @samp{bnd3raw} for i386 and amd64.
49098 @samp{bndcfgu} and @samp{bndstatus} for i386 and amd64.
49101 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
49102 describe a single register, @samp{orig_eax}.
49104 The @samp{org.gnu.gdb.i386.segments} feature is optional. It should
49105 describe two system registers: @samp{fs_base} and @samp{gs_base}.
49107 The @samp{org.gnu.gdb.i386.avx512} feature is optional and requires the
49108 @samp{org.gnu.gdb.i386.avx} feature. It should
49109 describe additional @sc{xmm} registers:
49113 @samp{xmm16h} through @samp{xmm31h}, only valid for amd64.
49116 It should describe the upper 128 bits of additional @sc{ymm} registers:
49120 @samp{ymm16h} through @samp{ymm31h}, only valid for amd64.
49124 describe the upper 256 bits of @sc{zmm} registers:
49128 @samp{zmm0h} through @samp{zmm7h} for i386.
49130 @samp{zmm0h} through @samp{zmm15h} for amd64.
49134 describe the additional @sc{zmm} registers:
49138 @samp{zmm16h} through @samp{zmm31h}, only valid for amd64.
49141 The @samp{org.gnu.gdb.i386.pkeys} feature is optional. It should
49142 describe a single register, @samp{pkru}. It is a 32-bit register
49143 valid for i386 and amd64.
49145 @node LoongArch Features
49146 @subsection LoongArch Features
49147 @cindex target descriptions, LoongArch Features
49149 The @samp{org.gnu.gdb.loongarch.base} feature is required for LoongArch
49150 targets. It should contain the registers @samp{r0} through @samp{r31},
49151 @samp{pc}, and @samp{badv}. Either the architectural names (@samp{r0},
49152 @samp{r1}, etc) can be used, or the ABI names (@samp{zero}, @samp{ra}, etc).
49154 The @samp{org.gnu.gdb.loongarch.fpu} feature is optional. If present,
49155 it should contain registers @samp{f0} through @samp{f31}, @samp{fcc},
49158 @node MicroBlaze Features
49159 @subsection MicroBlaze Features
49160 @cindex target descriptions, MicroBlaze features
49162 The @samp{org.gnu.gdb.microblaze.core} feature is required for MicroBlaze
49163 targets. It should contain registers @samp{r0} through @samp{r31},
49164 @samp{rpc}, @samp{rmsr}, @samp{rear}, @samp{resr}, @samp{rfsr}, @samp{rbtr},
49165 @samp{rpvr}, @samp{rpvr1} through @samp{rpvr11}, @samp{redr}, @samp{rpid},
49166 @samp{rzpr}, @samp{rtlbx}, @samp{rtlbsx}, @samp{rtlblo}, and @samp{rtlbhi}.
49168 The @samp{org.gnu.gdb.microblaze.stack-protect} feature is optional.
49169 If present, it should contain registers @samp{rshr} and @samp{rslr}
49171 @node MIPS Features
49172 @subsection @acronym{MIPS} Features
49173 @cindex target descriptions, @acronym{MIPS} features
49175 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
49176 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
49177 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
49180 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
49181 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
49182 registers. They may be 32-bit or 64-bit depending on the target.
49184 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
49185 it may be optional in a future version of @value{GDBN}. It should
49186 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
49187 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
49189 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
49190 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
49191 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
49192 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
49194 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
49195 contain a single register, @samp{restart}, which is used by the
49196 Linux kernel to control restartable syscalls.
49198 @node M68K Features
49199 @subsection M68K Features
49200 @cindex target descriptions, M68K features
49203 @item @samp{org.gnu.gdb.m68k.core}
49204 @itemx @samp{org.gnu.gdb.coldfire.core}
49205 @itemx @samp{org.gnu.gdb.fido.core}
49206 One of those features must be always present.
49207 The feature that is present determines which flavor of m68k is
49208 used. The feature that is present should contain registers
49209 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
49210 @samp{sp}, @samp{ps} and @samp{pc}.
49212 @item @samp{org.gnu.gdb.coldfire.fp}
49213 This feature is optional. If present, it should contain registers
49214 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
49217 Note that, despite the fact that this feature's name says
49218 @samp{coldfire}, it is used to describe any floating point registers.
49219 The size of the registers must match the main m68k flavor; so, for
49220 example, if the primary feature is reported as @samp{coldfire}, then
49221 64-bit floating point registers are required.
49224 @node NDS32 Features
49225 @subsection NDS32 Features
49226 @cindex target descriptions, NDS32 features
49228 The @samp{org.gnu.gdb.nds32.core} feature is required for NDS32
49229 targets. It should contain at least registers @samp{r0} through
49230 @samp{r10}, @samp{r15}, @samp{fp}, @samp{gp}, @samp{lp}, @samp{sp},
49233 The @samp{org.gnu.gdb.nds32.fpu} feature is optional. If present,
49234 it should contain 64-bit double-precision floating-point registers
49235 @samp{fd0} through @emph{fdN}, which should be @samp{fd3}, @samp{fd7},
49236 @samp{fd15}, or @samp{fd31} based on the FPU configuration implemented.
49238 @emph{Note:} The first sixteen 64-bit double-precision floating-point
49239 registers are overlapped with the thirty-two 32-bit single-precision
49240 floating-point registers. The 32-bit single-precision registers, if
49241 not being listed explicitly, will be synthesized from halves of the
49242 overlapping 64-bit double-precision registers. Listing 32-bit
49243 single-precision registers explicitly is deprecated, and the
49244 support to it could be totally removed some day.
49246 @node Nios II Features
49247 @subsection Nios II Features
49248 @cindex target descriptions, Nios II features
49250 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
49251 targets. It should contain the 32 core registers (@samp{zero},
49252 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
49253 @samp{pc}, and the 16 control registers (@samp{status} through
49256 @node OpenRISC 1000 Features
49257 @subsection Openrisc 1000 Features
49258 @cindex target descriptions, OpenRISC 1000 features
49260 The @samp{org.gnu.gdb.or1k.group0} feature is required for OpenRISC 1000
49261 targets. It should contain the 32 general purpose registers (@samp{r0}
49262 through @samp{r31}), @samp{ppc}, @samp{npc} and @samp{sr}.
49264 @node PowerPC Features
49265 @subsection PowerPC Features
49266 @cindex target descriptions, PowerPC features
49268 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
49269 targets. It should contain registers @samp{r0} through @samp{r31},
49270 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
49271 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
49273 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
49274 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
49276 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
49277 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr}, and
49278 @samp{vrsave}. @value{GDBN} will define pseudo-registers @samp{v0}
49279 through @samp{v31} as aliases for the corresponding @samp{vrX}
49282 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
49283 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN} will
49284 combine these registers with the floating point registers (@samp{f0}
49285 through @samp{f31}) and the altivec registers (@samp{vr0} through
49286 @samp{vr31}) to present the 128-bit wide registers @samp{vs0} through
49287 @samp{vs63}, the set of vector-scalar registers for POWER7.
49288 Therefore, this feature requires both @samp{org.gnu.gdb.power.fpu} and
49289 @samp{org.gnu.gdb.power.altivec}.
49291 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
49292 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
49293 @samp{spefscr}. SPE targets should provide 32-bit registers in
49294 @samp{org.gnu.gdb.power.core} and provide the upper halves in
49295 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
49296 these to present registers @samp{ev0} through @samp{ev31} to the
49299 The @samp{org.gnu.gdb.power.ppr} feature is optional. It should
49300 contain the 64-bit register @samp{ppr}.
49302 The @samp{org.gnu.gdb.power.dscr} feature is optional. It should
49303 contain the 64-bit register @samp{dscr}.
49305 The @samp{org.gnu.gdb.power.tar} feature is optional. It should
49306 contain the 64-bit register @samp{tar}.
49308 The @samp{org.gnu.gdb.power.ebb} feature is optional. It should
49309 contain registers @samp{bescr}, @samp{ebbhr} and @samp{ebbrr}, all
49312 The @samp{org.gnu.gdb.power.linux.pmu} feature is optional. It should
49313 contain registers @samp{mmcr0}, @samp{mmcr2}, @samp{siar}, @samp{sdar}
49314 and @samp{sier}, all 64-bit wide. This is the subset of the isa 2.07
49315 server PMU registers provided by @sc{gnu}/Linux.
49317 The @samp{org.gnu.gdb.power.htm.spr} feature is optional. It should
49318 contain registers @samp{tfhar}, @samp{texasr} and @samp{tfiar}, all
49321 The @samp{org.gnu.gdb.power.htm.core} feature is optional. It should
49322 contain the checkpointed general-purpose registers @samp{cr0} through
49323 @samp{cr31}, as well as the checkpointed registers @samp{clr} and
49324 @samp{cctr}. These registers may all be either 32-bit or 64-bit
49325 depending on the target. It should also contain the checkpointed
49326 registers @samp{ccr} and @samp{cxer}, which should both be 32-bit
49329 The @samp{org.gnu.gdb.power.htm.fpu} feature is optional. It should
49330 contain the checkpointed 64-bit floating-point registers @samp{cf0}
49331 through @samp{cf31}, as well as the checkpointed 64-bit register
49334 The @samp{org.gnu.gdb.power.htm.altivec} feature is optional. It
49335 should contain the checkpointed altivec registers @samp{cvr0} through
49336 @samp{cvr31}, all 128-bit wide. It should also contain the
49337 checkpointed registers @samp{cvscr} and @samp{cvrsave}, both 32-bit
49340 The @samp{org.gnu.gdb.power.htm.vsx} feature is optional. It should
49341 contain registers @samp{cvs0h} through @samp{cvs31h}. @value{GDBN}
49342 will combine these registers with the checkpointed floating point
49343 registers (@samp{cf0} through @samp{cf31}) and the checkpointed
49344 altivec registers (@samp{cvr0} through @samp{cvr31}) to present the
49345 128-bit wide checkpointed vector-scalar registers @samp{cvs0} through
49346 @samp{cvs63}. Therefore, this feature requires both
49347 @samp{org.gnu.gdb.power.htm.altivec} and
49348 @samp{org.gnu.gdb.power.htm.fpu}.
49350 The @samp{org.gnu.gdb.power.htm.ppr} feature is optional. It should
49351 contain the 64-bit checkpointed register @samp{cppr}.
49353 The @samp{org.gnu.gdb.power.htm.dscr} feature is optional. It should
49354 contain the 64-bit checkpointed register @samp{cdscr}.
49356 The @samp{org.gnu.gdb.power.htm.tar} feature is optional. It should
49357 contain the 64-bit checkpointed register @samp{ctar}.
49360 @node RISC-V Features
49361 @subsection RISC-V Features
49362 @cindex target descriptions, RISC-V Features
49364 The @samp{org.gnu.gdb.riscv.cpu} feature is required for RISC-V
49365 targets. It should contain the registers @samp{x0} through
49366 @samp{x31}, and @samp{pc}. Either the architectural names (@samp{x0},
49367 @samp{x1}, etc) can be used, or the ABI names (@samp{zero}, @samp{ra},
49370 The @samp{org.gnu.gdb.riscv.fpu} feature is optional. If present, it
49371 should contain registers @samp{f0} through @samp{f31}, @samp{fflags},
49372 @samp{frm}, and @samp{fcsr}. As with the cpu feature, either the
49373 architectural register names, or the ABI names can be used.
49375 The @samp{org.gnu.gdb.riscv.virtual} feature is optional. If present,
49376 it should contain registers that are not backed by real registers on
49377 the target, but are instead virtual, where the register value is
49378 derived from other target state. In many ways these are like
49379 @value{GDBN}s pseudo-registers, except implemented by the target.
49380 Currently the only register expected in this set is the one byte
49381 @samp{priv} register that contains the target's privilege level in the
49382 least significant two bits.
49384 The @samp{org.gnu.gdb.riscv.csr} feature is optional. If present, it
49385 should contain all of the target's standard CSRs. Standard CSRs are
49386 those defined in the RISC-V specification documents. There is some
49387 overlap between this feature and the fpu feature; the @samp{fflags},
49388 @samp{frm}, and @samp{fcsr} registers could be in either feature. The
49389 expectation is that these registers will be in the fpu feature if the
49390 target has floating point hardware, but can be moved into the csr
49391 feature if the target has the floating point control registers, but no
49392 other floating point hardware.
49394 The @samp{org.gnu.gdb.riscv.vector} feature is optional. If present,
49395 it should contain registers @samp{v0} through @samp{v31}, all of which
49396 must be the same size.
49399 @subsection RX Features
49400 @cindex target descriptions, RX Features
49402 The @samp{org.gnu.gdb.rx.core} feature is required for RX
49403 targets. It should contain the registers @samp{r0} through
49404 @samp{r15}, @samp{usp}, @samp{isp}, @samp{psw}, @samp{pc}, @samp{intb},
49405 @samp{bpsw}, @samp{bpc}, @samp{fintv}, @samp{fpsw}, and @samp{acc}.
49407 @node S/390 and System z Features
49408 @subsection S/390 and System z Features
49409 @cindex target descriptions, S/390 features
49410 @cindex target descriptions, System z features
49412 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
49413 System z targets. It should contain the PSW and the 16 general
49414 registers. In particular, System z targets should provide the 64-bit
49415 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
49416 S/390 targets should provide the 32-bit versions of these registers.
49417 A System z target that runs in 31-bit addressing mode should provide
49418 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
49419 register's upper halves @samp{r0h} through @samp{r15h}, and their
49420 lower halves @samp{r0l} through @samp{r15l}.
49422 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
49423 contain the 64-bit registers @samp{f0} through @samp{f15}, and
49426 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
49427 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
49429 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
49430 contain the register @samp{orig_r2}, which is 64-bit wide on System z
49431 targets and 32-bit otherwise. In addition, the feature may contain
49432 the @samp{last_break} register, whose width depends on the addressing
49433 mode, as well as the @samp{system_call} register, which is always
49436 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
49437 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
49438 @samp{atia}, and @samp{tr0} through @samp{tr15}.
49440 The @samp{org.gnu.gdb.s390.vx} feature is optional. It should contain
49441 64-bit wide registers @samp{v0l} through @samp{v15l}, which will be
49442 combined by @value{GDBN} with the floating point registers @samp{f0}
49443 through @samp{f15} to present the 128-bit wide vector registers
49444 @samp{v0} through @samp{v15}. In addition, this feature should
49445 contain the 128-bit wide vector registers @samp{v16} through
49448 The @samp{org.gnu.gdb.s390.gs} feature is optional. It should contain
49449 the 64-bit wide guarded-storage-control registers @samp{gsd},
49450 @samp{gssm}, and @samp{gsepla}.
49452 The @samp{org.gnu.gdb.s390.gsbc} feature is optional. It should contain
49453 the 64-bit wide guarded-storage broadcast control registers
49454 @samp{bc_gsd}, @samp{bc_gssm}, and @samp{bc_gsepla}.
49456 @node Sparc Features
49457 @subsection Sparc Features
49458 @cindex target descriptions, sparc32 features
49459 @cindex target descriptions, sparc64 features
49460 The @samp{org.gnu.gdb.sparc.cpu} feature is required for sparc32/sparc64
49461 targets. It should describe the following registers:
49465 @samp{g0} through @samp{g7}
49467 @samp{o0} through @samp{o7}
49469 @samp{l0} through @samp{l7}
49471 @samp{i0} through @samp{i7}
49474 They may be 32-bit or 64-bit depending on the target.
49476 Also the @samp{org.gnu.gdb.sparc.fpu} feature is required for sparc32/sparc64
49477 targets. It should describe the following registers:
49481 @samp{f0} through @samp{f31}
49483 @samp{f32} through @samp{f62} for sparc64
49486 The @samp{org.gnu.gdb.sparc.cp0} feature is required for sparc32/sparc64
49487 targets. It should describe the following registers:
49491 @samp{y}, @samp{psr}, @samp{wim}, @samp{tbr}, @samp{pc}, @samp{npc},
49492 @samp{fsr}, and @samp{csr} for sparc32
49494 @samp{pc}, @samp{npc}, @samp{state}, @samp{fsr}, @samp{fprs}, and @samp{y}
49498 @node TIC6x Features
49499 @subsection TMS320C6x Features
49500 @cindex target descriptions, TIC6x features
49501 @cindex target descriptions, TMS320C6x features
49502 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
49503 targets. It should contain registers @samp{A0} through @samp{A15},
49504 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
49506 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
49507 contain registers @samp{A16} through @samp{A31} and @samp{B16}
49508 through @samp{B31}.
49510 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
49511 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
49513 @node Operating System Information
49514 @appendix Operating System Information
49515 @cindex operating system information
49517 Users of @value{GDBN} often wish to obtain information about the state of
49518 the operating system running on the target---for example the list of
49519 processes, or the list of open files. This section describes the
49520 mechanism that makes it possible. This mechanism is similar to the
49521 target features mechanism (@pxref{Target Descriptions}), but focuses
49522 on a different aspect of target.
49524 Operating system information is retrieved from the target via the
49525 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
49526 read}). The object name in the request should be @samp{osdata}, and
49527 the @var{annex} identifies the data to be fetched.
49534 @appendixsection Process list
49535 @cindex operating system information, process list
49537 When requesting the process list, the @var{annex} field in the
49538 @samp{qXfer} request should be @samp{processes}. The returned data is
49539 an XML document. The formal syntax of this document is defined in
49540 @file{gdb/features/osdata.dtd}.
49542 An example document is:
49545 <?xml version="1.0"?>
49546 <!DOCTYPE target SYSTEM "osdata.dtd">
49547 <osdata type="processes">
49549 <column name="pid">1</column>
49550 <column name="user">root</column>
49551 <column name="command">/sbin/init</column>
49552 <column name="cores">1,2,3</column>
49557 Each item should include a column whose name is @samp{pid}. The value
49558 of that column should identify the process on the target. The
49559 @samp{user} and @samp{command} columns are optional, and will be
49560 displayed by @value{GDBN}. The @samp{cores} column, if present,
49561 should contain a comma-separated list of cores that this process
49562 is running on. Target may provide additional columns,
49563 which @value{GDBN} currently ignores.
49565 @node Trace File Format
49566 @appendix Trace File Format
49567 @cindex trace file format
49569 The trace file comes in three parts: a header, a textual description
49570 section, and a trace frame section with binary data.
49572 The header has the form @code{\x7fTRACE0\n}. The first byte is
49573 @code{0x7f} so as to indicate that the file contains binary data,
49574 while the @code{0} is a version number that may have different values
49577 The description section consists of multiple lines of @sc{ascii} text
49578 separated by newline characters (@code{0xa}). The lines may include a
49579 variety of optional descriptive or context-setting information, such
49580 as tracepoint definitions or register set size. @value{GDBN} will
49581 ignore any line that it does not recognize. An empty line marks the end
49586 Specifies the size of a register block in bytes. This is equal to the
49587 size of a @code{g} packet payload in the remote protocol. @var{size}
49588 is an ascii decimal number. There should be only one such line in
49589 a single trace file.
49591 @item status @var{status}
49592 Trace status. @var{status} has the same format as a @code{qTStatus}
49593 remote packet reply. There should be only one such line in a single trace
49596 @item tp @var{payload}
49597 Tracepoint definition. The @var{payload} has the same format as
49598 @code{qTfP}/@code{qTsP} remote packet reply payload. A single tracepoint
49599 may take multiple lines of definition, corresponding to the multiple
49602 @item tsv @var{payload}
49603 Trace state variable definition. The @var{payload} has the same format as
49604 @code{qTfV}/@code{qTsV} remote packet reply payload. A single variable
49605 may take multiple lines of definition, corresponding to the multiple
49608 @item tdesc @var{payload}
49609 Target description in XML format. The @var{payload} is a single line of
49610 the XML file. All such lines should be concatenated together to get
49611 the original XML file. This file is in the same format as @code{qXfer}
49612 @code{features} payload, and corresponds to the main @code{target.xml}
49613 file. Includes are not allowed.
49617 The trace frame section consists of a number of consecutive frames.
49618 Each frame begins with a two-byte tracepoint number, followed by a
49619 four-byte size giving the amount of data in the frame. The data in
49620 the frame consists of a number of blocks, each introduced by a
49621 character indicating its type (at least register, memory, and trace
49622 state variable). The data in this section is raw binary, not a
49623 hexadecimal or other encoding; its endianness matches the target's
49626 @c FIXME bi-arch may require endianness/arch info in description section
49629 @item R @var{bytes}
49630 Register block. The number and ordering of bytes matches that of a
49631 @code{g} packet in the remote protocol. Note that these are the
49632 actual bytes, in target order, not a hexadecimal encoding.
49634 @item M @var{address} @var{length} @var{bytes}...
49635 Memory block. This is a contiguous block of memory, at the 8-byte
49636 address @var{address}, with a 2-byte length @var{length}, followed by
49637 @var{length} bytes.
49639 @item V @var{number} @var{value}
49640 Trace state variable block. This records the 8-byte signed value
49641 @var{value} of trace state variable numbered @var{number}.
49645 Future enhancements of the trace file format may include additional types
49648 @node Index Section Format
49649 @appendix @code{.gdb_index} section format
49650 @cindex .gdb_index section format
49651 @cindex index section format
49653 This section documents the index section that is created by @code{save
49654 gdb-index} (@pxref{Index Files}). The index section is
49655 DWARF-specific; some knowledge of DWARF is assumed in this
49658 The mapped index file format is designed to be directly
49659 @code{mmap}able on any architecture. In most cases, a datum is
49660 represented using a little-endian 32-bit integer value, called an
49661 @code{offset_type}. Big endian machines must byte-swap the values
49662 before using them. Exceptions to this rule are noted. The data is
49663 laid out such that alignment is always respected.
49665 A mapped index consists of several areas, laid out in order.
49669 The file header. This is a sequence of values, of @code{offset_type}
49670 unless otherwise noted:
49674 The version number, currently 9. Versions 1, 2 and 3 are obsolete.
49675 Version 4 uses a different hashing function from versions 5 and 6.
49676 Version 6 includes symbols for inlined functions, whereas versions 4
49677 and 5 do not. Version 7 adds attributes to the CU indices in the
49678 symbol table. Version 8 specifies that symbols from DWARF type units
49679 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
49680 compilation unit (@samp{DW_TAG_comp_unit}) using the type. Version 9 adds
49681 the name and the language of the main function to the index.
49683 @value{GDBN} will only read version 4, 5, or 6 indices
49684 by specifying @code{set use-deprecated-index-sections on}.
49685 GDB has a workaround for potentially broken version 7 indices so it is
49686 currently not flagged as deprecated.
49689 The offset, from the start of the file, of the CU list.
49692 The offset, from the start of the file, of the types CU list. Note
49693 that this area can be empty, in which case this offset will be equal
49694 to the next offset.
49697 The offset, from the start of the file, of the address area.
49700 The offset, from the start of the file, of the symbol table.
49703 The offset, from the start of the file, of the shortcut table.
49706 The offset, from the start of the file, of the constant pool.
49710 The CU list. This is a sequence of pairs of 64-bit little-endian
49711 values, sorted by the CU offset. The first element in each pair is
49712 the offset of a CU in the @code{.debug_info} section. The second
49713 element in each pair is the length of that CU. References to a CU
49714 elsewhere in the map are done using a CU index, which is just the
49715 0-based index into this table. Note that if there are type CUs, then
49716 conceptually CUs and type CUs form a single list for the purposes of
49720 The types CU list. This is a sequence of triplets of 64-bit
49721 little-endian values. In a triplet, the first value is the CU offset,
49722 the second value is the type offset in the CU, and the third value is
49723 the type signature. The types CU list is not sorted.
49726 The address area. The address area consists of a sequence of address
49727 entries. Each address entry has three elements:
49731 The low address. This is a 64-bit little-endian value.
49734 The high address. This is a 64-bit little-endian value. Like
49735 @code{DW_AT_high_pc}, the value is one byte beyond the end.
49738 The CU index. This is an @code{offset_type} value.
49742 The symbol table. This is an open-addressed hash table. The size of
49743 the hash table is always a power of 2.
49745 Each slot in the hash table consists of a pair of @code{offset_type}
49746 values. The first value is the offset of the symbol's name in the
49747 constant pool. The second value is the offset of the CU vector in the
49750 If both values are 0, then this slot in the hash table is empty. This
49751 is ok because while 0 is a valid constant pool index, it cannot be a
49752 valid index for both a string and a CU vector.
49754 The hash value for a table entry is computed by applying an
49755 iterative hash function to the symbol's name. Starting with an
49756 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
49757 the string is incorporated into the hash using the formula depending on the
49762 The formula is @code{r = r * 67 + c - 113}.
49764 @item Versions 5 to 7
49765 The formula is @code{r = r * 67 + tolower (c) - 113}.
49768 The terminating @samp{\0} is not incorporated into the hash.
49770 The step size used in the hash table is computed via
49771 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
49772 value, and @samp{size} is the size of the hash table. The step size
49773 is used to find the next candidate slot when handling a hash
49776 The names of C@t{++} symbols in the hash table are canonicalized. We
49777 don't currently have a simple description of the canonicalization
49778 algorithm; if you intend to create new index sections, you must read
49781 @item The shortcut table
49782 This is a data structure with the following fields:
49785 @item Language of main
49786 An @code{offset_type} value indicating the language of the main function as a
49787 @code{DW_LANG_} constant. This value will be zero if main function information
49791 An @code{offset_type} value indicating the offset of the main function's name
49792 in the constant pool. This value must be ignored if the value for the language
49797 The constant pool. This is simply a bunch of bytes. It is organized
49798 so that alignment is correct: CU vectors are stored first, followed by
49801 A CU vector in the constant pool is a sequence of @code{offset_type}
49802 values. The first value is the number of CU indices in the vector.
49803 Each subsequent value is the index and symbol attributes of a CU in
49804 the CU list. This element in the hash table is used to indicate which
49805 CUs define the symbol and how the symbol is used.
49806 See below for the format of each CU index+attributes entry.
49808 A string in the constant pool is zero-terminated.
49811 Attributes were added to CU index values in @code{.gdb_index} version 7.
49812 If a symbol has multiple uses within a CU then there is one
49813 CU index+attributes value for each use.
49815 The format of each CU index+attributes entry is as follows
49821 This is the index of the CU in the CU list.
49823 These bits are reserved for future purposes and must be zero.
49825 The kind of the symbol in the CU.
49829 This value is reserved and should not be used.
49830 By reserving zero the full @code{offset_type} value is backwards compatible
49831 with previous versions of the index.
49833 The symbol is a type.
49835 The symbol is a variable or an enum value.
49837 The symbol is a function.
49839 Any other kind of symbol.
49841 These values are reserved.
49845 This bit is zero if the value is global and one if it is static.
49847 The determination of whether a symbol is global or static is complicated.
49848 The authorative reference is the file @file{dwarf2read.c} in
49849 @value{GDBN} sources.
49853 This pseudo-code describes the computation of a symbol's kind and
49854 global/static attributes in the index.
49857 is_external = get_attribute (die, DW_AT_external);
49858 language = get_attribute (cu_die, DW_AT_language);
49861 case DW_TAG_typedef:
49862 case DW_TAG_base_type:
49863 case DW_TAG_subrange_type:
49867 case DW_TAG_enumerator:
49869 is_static = language != CPLUS;
49871 case DW_TAG_subprogram:
49873 is_static = ! (is_external || language == ADA);
49875 case DW_TAG_constant:
49877 is_static = ! is_external;
49879 case DW_TAG_variable:
49881 is_static = ! is_external;
49883 case DW_TAG_namespace:
49887 case DW_TAG_class_type:
49888 case DW_TAG_interface_type:
49889 case DW_TAG_structure_type:
49890 case DW_TAG_union_type:
49891 case DW_TAG_enumeration_type:
49893 is_static = language != CPLUS;
49901 @appendix Download debugging resources with Debuginfod
49904 @code{debuginfod} is an HTTP server for distributing ELF, DWARF and source
49907 With the @code{debuginfod} client library, @file{libdebuginfod}, @value{GDBN}
49908 can query servers using the build IDs associated with missing debug info,
49909 executables and source files in order to download them on demand.
49911 For instructions on building @value{GDBN} with @file{libdebuginfod},
49912 @pxref{Configure Options,,--with-debuginfod}. @code{debuginfod} is packaged
49913 with @code{elfutils}, starting with version 0.178. See
49914 @uref{https://sourceware.org/elfutils/Debuginfod.html} for more information
49915 regarding @code{debuginfod}.
49918 * Debuginfod Settings:: Configuring debuginfod with @value{GDBN}
49921 @node Debuginfod Settings
49922 @section Debuginfod Settings
49924 @value{GDBN} provides the following commands for configuring @code{debuginfod}.
49927 @kindex set debuginfod enabled
49928 @anchor{set debuginfod enabled}
49929 @item set debuginfod enabled
49930 @itemx set debuginfod enabled on
49931 @cindex enable debuginfod
49932 @value{GDBN} may query @code{debuginfod} servers for missing debug info and
49933 source files. @value{GDBN} may also download individual ELF/DWARF sections
49934 such as @code{.gdb_index} to help reduce the total amount of data downloaded
49935 from @code{debuginfod} servers; this can be controlled by @w{@code{maint
49936 set debuginfod download-sections}} (@pxref{Maintenance Commands, maint set
49937 debuginfod download-sections}).
49939 @item set debuginfod enabled off
49940 @value{GDBN} will not attempt to query @code{debuginfod} servers when missing
49941 debug info or source files. By default, @code{debuginfod enabled} is set to
49942 @code{off} for non-interactive sessions.
49944 @item set debuginfod enabled ask
49945 @value{GDBN} will prompt the user to enable or disable @code{debuginfod} before
49946 attempting to perform the next query. By default, @code{debuginfod enabled}
49947 is set to @code{ask} for interactive sessions.
49949 @kindex show debuginfod enabled
49950 @item show debuginfod enabled
49951 Display whether @code{debuginfod enabled} is set to @code{on}, @code{off} or
49954 @kindex set debuginfod urls
49955 @cindex configure debuginfod URLs
49956 @item set debuginfod urls
49957 @itemx set debuginfod urls @var{urls}
49958 Set the space-separated list of URLs that @code{debuginfod} will attempt to
49959 query. Only @code{http://}, @code{https://} and @code{file://} protocols
49960 should be used. The default value of @code{debuginfod urls} is copied from
49961 the @var{DEBUGINFOD_URLS} environment variable.
49963 @kindex show debuginfod urls
49964 @item show debuginfod urls
49965 Display the list of URLs that @code{debuginfod} will attempt to query.
49967 @kindex set debuginfod verbose
49968 @cindex debuginfod verbosity
49969 @item set debuginfod verbose
49970 @itemx set debuginfod verbose @var{n}
49971 Enable or disable @code{debuginfod}-related output. Use a non-zero value
49972 to enable and @code{0} to disable. @code{debuginfod} output is shown by
49975 @kindex show debuginfod verbose
49976 @item show debuginfod verbose
49977 Show the current verbosity setting.
49982 @appendix Manual pages
49986 * gdb man:: The GNU Debugger man page
49987 * gdbserver man:: Remote Server for the GNU Debugger man page
49988 * gcore man:: Generate a core file of a running program
49989 * gdbinit man:: gdbinit scripts
49990 * gdb-add-index man:: Add index files to speed up GDB
49996 @c man title gdb The GNU Debugger
49998 @c man begin SYNOPSIS gdb
49999 gdb [OPTIONS] [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
50002 @c man begin DESCRIPTION gdb
50003 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
50004 going on ``inside'' another program while it executes -- or what another
50005 program was doing at the moment it crashed.
50007 @value{GDBN} can do four main kinds of things (plus other things in support of
50008 these) to help you catch bugs in the act:
50012 Start your program, specifying anything that might affect its behavior.
50015 Make your program stop on specified conditions.
50018 Examine what has happened, when your program has stopped.
50021 Change things in your program, so you can experiment with correcting the
50022 effects of one bug and go on to learn about another.
50025 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
50028 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
50029 commands from the terminal until you tell it to exit with the @value{GDBN}
50030 command @code{quit} or @code{exit}. You can get online help from @value{GDBN} itself
50031 by using the command @code{help}.
50033 You can run @code{gdb} with no arguments or options; but the most
50034 usual way to start @value{GDBN} is with one argument or two, specifying an
50035 executable program as the argument:
50041 You can also start with both an executable program and a core file specified:
50047 You can, instead, specify a process ID as a second argument or use option
50048 @code{-p}, if you want to debug a running process:
50056 would attach @value{GDBN} to process @code{1234}. With option @option{-p} you
50057 can omit the @var{program} filename.
50059 Here are some of the most frequently needed @value{GDBN} commands:
50061 @c pod2man highlights the right hand side of the @item lines.
50063 @item break [@var{file}:][@var{function}|@var{line}]
50064 Set a breakpoint at @var{function} or @var{line} (in @var{file}).
50066 @item run [@var{arglist}]
50067 Start your program (with @var{arglist}, if specified).
50070 Backtrace: display the program stack.
50072 @item print @var{expr}
50073 Display the value of an expression.
50076 Continue running your program (after stopping, e.g.@: at a breakpoint).
50079 Execute next program line (after stopping); step @emph{over} any
50080 function calls in the line.
50082 @item edit [@var{file}:]@var{function}
50083 look at the program line where it is presently stopped.
50085 @item list [@var{file}:]@var{function}
50086 type the text of the program in the vicinity of where it is presently stopped.
50089 Execute next program line (after stopping); step @emph{into} any
50090 function calls in the line.
50092 @item help [@var{name}]
50093 Show information about @value{GDBN} command @var{name}, or general information
50094 about using @value{GDBN}.
50098 Exit from @value{GDBN}.
50102 For full details on @value{GDBN},
50103 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
50104 by Richard M. Stallman and Roland H. Pesch. The same text is available online
50105 as the @code{gdb} entry in the @code{info} program.
50109 @c man begin OPTIONS gdb
50110 Any arguments other than options specify an executable
50111 file and core file (or process ID); that is, the first argument
50112 encountered with no
50113 associated option flag is equivalent to a @option{--se} option, and the second,
50114 if any, is equivalent to a @option{-c} option if it's the name of a file.
50116 both long and abbreviated forms; both are shown here. The long forms are also
50117 recognized if you truncate them, so long as enough of the option is
50118 present to be unambiguous.
50120 The abbreviated forms are shown here with @samp{-} and long forms are shown
50121 with @samp{--} to reflect how they are shown in @option{--help}. However,
50122 @value{GDBN} recognizes all of the following conventions for most options:
50125 @item --option=@var{value}
50126 @item --option @var{value}
50127 @item -option=@var{value}
50128 @item -option @var{value}
50129 @item --o=@var{value}
50130 @item --o @var{value}
50131 @item -o=@var{value}
50132 @item -o @var{value}
50135 All the options and command line arguments you give are processed
50136 in sequential order. The order makes a difference when the @option{-x}
50142 List all options, with brief explanations.
50144 @item --symbols=@var{file}
50145 @itemx -s @var{file}
50146 Read symbol table from @var{file}.
50149 Enable writing into executable and core files.
50151 @item --exec=@var{file}
50152 @itemx -e @var{file}
50153 Use @var{file} as the executable file to execute when
50154 appropriate, and for examining pure data in conjunction with a core
50157 @item --se=@var{file}
50158 Read symbol table from @var{file} and use it as the executable
50161 @item --core=@var{file}
50162 @itemx -c @var{file}
50163 Use @var{file} as a core dump to examine.
50165 @item --command=@var{file}
50166 @itemx -x @var{file}
50167 Execute @value{GDBN} commands from @var{file}.
50169 @item --eval-command=@var{command}
50170 @item -ex @var{command}
50171 Execute given @value{GDBN} @var{command}.
50173 @item --init-eval-command=@var{command}
50175 Execute @value{GDBN} @var{command} before loading the inferior.
50177 @item --directory=@var{directory}
50178 @itemx -d @var{directory}
50179 Add @var{directory} to the path to search for source files.
50182 Do not execute commands from @file{~/.config/gdb/gdbinit},
50183 @file{~/.gdbinit}, @file{~/.config/gdb/gdbearlyinit}, or
50184 @file{~/.gdbearlyinit}
50188 Do not execute commands from any @file{.gdbinit} or
50189 @file{.gdbearlyinit} initialization files.
50194 ``Quiet''. Do not print the introductory and copyright messages. These
50195 messages are also suppressed in batch mode.
50198 Run in batch mode. Exit with status @code{0} after processing all the command
50199 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
50200 Exit with nonzero status if an error occurs in executing the @value{GDBN}
50201 commands in the command files.
50203 Batch mode may be useful for running @value{GDBN} as a filter, for example to
50204 download and run a program on another computer; in order to make this
50205 more useful, the message
50208 Program exited normally.
50212 (which is ordinarily issued whenever a program running under @value{GDBN} control
50213 terminates) is not issued when running in batch mode.
50215 @item --batch-silent
50216 Run in batch mode, just like @option{--batch}, but totally silent. All @value{GDBN}
50217 output is suppressed (stderr is unaffected). This is much quieter than
50218 @option{--silent} and would be useless for an interactive session.
50220 This is particularly useful when using targets that give @samp{Loading section}
50221 messages, for example.
50223 Note that targets that give their output via @value{GDBN}, as opposed to writing
50224 directly to @code{stdout}, will also be made silent.
50226 @item --args @var{prog} [@var{arglist}]
50227 Change interpretation of command line so that arguments following this
50228 option are passed as arguments to the inferior. As an example, take
50229 the following command:
50236 It would start @value{GDBN} with @option{-q}, not printing the introductory message. On
50237 the other hand, using:
50240 gdb --args ./a.out -q
50244 starts @value{GDBN} with the introductory message, and passes the option to the inferior.
50246 @item --pid=@var{pid}
50247 Attach @value{GDBN} to an already running program, with the PID @var{pid}.
50250 Open the terminal user interface.
50253 Read all symbols from the given symfile on the first access.
50256 Do not read symbol files.
50258 @item --return-child-result
50259 @value{GDBN}'s exit code will be the same as the child's exit code.
50261 @item --configuration
50262 Print details about GDB configuration and then exit.
50265 Print version information and then exit.
50267 @item --cd=@var{directory}
50268 Run @value{GDBN} using @var{directory} as its working directory,
50269 instead of the current directory.
50271 @item --data-directory=@var{directory}
50273 Run @value{GDBN} using @var{directory} as its data directory. The data
50274 directory is where @value{GDBN} searches for its auxiliary files.
50278 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
50279 @value{GDBN} to output the full file name and line number in a standard,
50280 recognizable fashion each time a stack frame is displayed (which
50281 includes each time the program stops). This recognizable format looks
50282 like two @samp{\032} characters, followed by the file name, line number
50283 and character position separated by colons, and a newline. The
50284 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
50285 characters as a signal to display the source code for the frame.
50287 @item -b @var{baudrate}
50288 Set the line speed (baud rate or bits per second) of any serial
50289 interface used by @value{GDBN} for remote debugging.
50291 @item -l @var{timeout}
50292 Set timeout, in seconds, for remote debugging.
50294 @item --tty=@var{device}
50295 Run using @var{device} for your program's standard input and output.
50299 @c man begin SEEALSO gdb
50301 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
50302 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
50303 documentation are properly installed at your site, the command
50310 should give you access to the complete manual.
50312 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
50313 Richard M. Stallman and Roland H. Pesch, July 1991.
50317 @node gdbserver man
50318 @heading gdbserver man
50320 @c man title gdbserver Remote Server for the GNU Debugger
50322 @c man begin SYNOPSIS gdbserver
50323 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
50325 gdbserver --attach @var{comm} @var{pid}
50327 gdbserver --multi @var{comm}
50331 @c man begin DESCRIPTION gdbserver
50332 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
50333 than the one which is running the program being debugged.
50336 @subheading Usage (server (target) side)
50339 Usage (server (target) side):
50342 First, you need to have a copy of the program you want to debug put onto
50343 the target system. The program can be stripped to save space if needed, as
50344 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
50345 the @value{GDBN} running on the host system.
50347 To use the server, you log on to the target system, and run the @command{gdbserver}
50348 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
50349 your program, and (c) its arguments. The general syntax is:
50352 target> gdbserver @var{comm} @var{program} [@var{args} ...]
50355 For example, using a serial port, you might say:
50359 @c @file would wrap it as F</dev/com1>.
50360 target> gdbserver /dev/com1 emacs foo.txt
50363 target> gdbserver @file{/dev/com1} emacs foo.txt
50367 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
50368 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
50369 waits patiently for the host @value{GDBN} to communicate with it.
50371 To use a TCP connection, you could say:
50374 target> gdbserver host:2345 emacs foo.txt
50377 This says pretty much the same thing as the last example, except that we are
50378 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
50379 that we are expecting to see a TCP connection from @code{host} to local TCP port
50380 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
50381 want for the port number as long as it does not conflict with any existing TCP
50382 ports on the target system. This same port number must be used in the host
50383 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
50384 you chose a port number that conflicts with another service, @command{gdbserver} will
50385 print an error message and exit.
50387 @command{gdbserver} can also attach to running programs.
50388 This is accomplished via the @option{--attach} argument. The syntax is:
50391 target> gdbserver --attach @var{comm} @var{pid}
50394 @var{pid} is the process ID of a currently running process. It isn't
50395 necessary to point @command{gdbserver} at a binary for the running process.
50397 To start @code{gdbserver} without supplying an initial command to run
50398 or process ID to attach, use the @option{--multi} command line option.
50399 In such case you should connect using @kbd{target extended-remote} to start
50400 the program you want to debug.
50403 target> gdbserver --multi @var{comm}
50407 @subheading Usage (host side)
50413 You need an unstripped copy of the target program on your host system, since
50414 @value{GDBN} needs to examine its symbol tables and such. Start up @value{GDBN} as you normally
50415 would, with the target program as the first argument. (You may need to use the
50416 @option{--baud} option if the serial line is running at anything except 9600 baud.)
50417 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
50418 new command you need to know about is @code{target remote}
50419 (or @code{target extended-remote}). Its argument is either
50420 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
50421 descriptor. For example:
50425 @c @file would wrap it as F</dev/ttyb>.
50426 (@value{GDBP}) target remote /dev/ttyb
50429 (@value{GDBP}) target remote @file{/dev/ttyb}
50434 communicates with the server via serial line @file{/dev/ttyb}, and:
50437 (@value{GDBP}) target remote the-target:2345
50441 communicates via a TCP connection to port 2345 on host `the-target', where
50442 you previously started up @command{gdbserver} with the same port number. Note that for
50443 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
50444 command, otherwise you may get an error that looks something like
50445 `Connection refused'.
50447 @command{gdbserver} can also debug multiple inferiors at once,
50450 the @value{GDBN} manual in node @code{Inferiors Connections and Programs}
50451 -- shell command @code{info -f gdb -n 'Inferiors Connections and Programs'}.
50454 @ref{Inferiors Connections and Programs}.
50456 In such case use the @code{extended-remote} @value{GDBN} command variant:
50459 (@value{GDBP}) target extended-remote the-target:2345
50462 The @command{gdbserver} option @option{--multi} may or may not be used in such
50466 @c man begin OPTIONS gdbserver
50467 There are three different modes for invoking @command{gdbserver}:
50472 Debug a specific program specified by its program name:
50475 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
50478 The @var{comm} parameter specifies how should the server communicate
50479 with @value{GDBN}; it is either a device name (to use a serial line),
50480 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
50481 stdin/stdout of @code{gdbserver}. Specify the name of the program to
50482 debug in @var{prog}. Any remaining arguments will be passed to the
50483 program verbatim. When the program exits, @value{GDBN} will close the
50484 connection, and @code{gdbserver} will exit.
50487 Debug a specific program by specifying the process ID of a running
50491 gdbserver --attach @var{comm} @var{pid}
50494 The @var{comm} parameter is as described above. Supply the process ID
50495 of a running program in @var{pid}; @value{GDBN} will do everything
50496 else. Like with the previous mode, when the process @var{pid} exits,
50497 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
50500 Multi-process mode -- debug more than one program/process:
50503 gdbserver --multi @var{comm}
50506 In this mode, @value{GDBN} can instruct @command{gdbserver} which
50507 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
50508 close the connection when a process being debugged exits, so you can
50509 debug several processes in the same session.
50512 In each of the modes you may specify these options:
50517 List all options, with brief explanations.
50520 This option causes @command{gdbserver} to print its version number and exit.
50523 @command{gdbserver} will attach to a running program. The syntax is:
50526 target> gdbserver --attach @var{comm} @var{pid}
50529 @var{pid} is the process ID of a currently running process. It isn't
50530 necessary to point @command{gdbserver} at a binary for the running process.
50533 To start @code{gdbserver} without supplying an initial command to run
50534 or process ID to attach, use this command line option.
50535 Then you can connect using @kbd{target extended-remote} and start
50536 the program you want to debug. The syntax is:
50539 target> gdbserver --multi @var{comm}
50543 Instruct @code{gdbserver} to display extra status information about the debugging
50545 This option is intended for @code{gdbserver} development and for bug reports to
50548 @item --remote-debug
50549 Instruct @code{gdbserver} to display remote protocol debug output.
50550 This option is intended for @code{gdbserver} development and for bug reports to
50553 @item --debug-file=@var{filename}
50554 Instruct @code{gdbserver} to send any debug output to the given @var{filename}.
50555 This option is intended for @code{gdbserver} development and for bug reports to
50558 @item --debug-format=option1@r{[},option2,...@r{]}
50559 Instruct @code{gdbserver} to include extra information in each line
50560 of debugging output.
50561 @xref{Other Command-Line Arguments for gdbserver}.
50564 Specify a wrapper to launch programs
50565 for debugging. The option should be followed by the name of the
50566 wrapper, then any command-line arguments to pass to the wrapper, then
50567 @kbd{--} indicating the end of the wrapper arguments.
50570 By default, @command{gdbserver} keeps the listening TCP port open, so that
50571 additional connections are possible. However, if you start @code{gdbserver}
50572 with the @option{--once} option, it will stop listening for any further
50573 connection attempts after connecting to the first @value{GDBN} session.
50575 @c --disable-packet is not documented for users.
50577 @c --disable-randomization and --no-disable-randomization are superseded by
50578 @c QDisableRandomization.
50583 @c man begin SEEALSO gdbserver
50585 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
50586 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
50587 documentation are properly installed at your site, the command
50593 should give you access to the complete manual.
50595 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
50596 Richard M. Stallman and Roland H. Pesch, July 1991.
50603 @c man title gcore Generate a core file of a running program
50606 @c man begin SYNOPSIS gcore
50607 gcore [-a] [-o @var{prefix}] @var{pid1} [@var{pid2}...@var{pidN}]
50611 @c man begin DESCRIPTION gcore
50612 Generate core dumps of one or more running programs with process IDs
50613 @var{pid1}, @var{pid2}, etc. A core file produced by @command{gcore}
50614 is equivalent to one produced by the kernel when the process crashes
50615 (and when @kbd{ulimit -c} was used to set up an appropriate core dump
50616 limit). However, unlike after a crash, after @command{gcore} finishes
50617 its job the program remains running without any change.
50620 @c man begin OPTIONS gcore
50623 Dump all memory mappings. The actual effect of this option depends on
50624 the Operating System. On @sc{gnu}/Linux, it will disable
50625 @code{use-coredump-filter} (@pxref{set use-coredump-filter}) and
50626 enable @code{dump-excluded-mappings} (@pxref{set
50627 dump-excluded-mappings}).
50629 @item -o @var{prefix}
50630 The optional argument @var{prefix} specifies the prefix to be used
50631 when composing the file names of the core dumps. The file name is
50632 composed as @file{@var{prefix}.@var{pid}}, where @var{pid} is the
50633 process ID of the running program being analyzed by @command{gcore}.
50634 If not specified, @var{prefix} defaults to @var{gcore}.
50638 @c man begin SEEALSO gcore
50640 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
50641 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
50642 documentation are properly installed at your site, the command
50649 should give you access to the complete manual.
50651 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
50652 Richard M. Stallman and Roland H. Pesch, July 1991.
50659 @c man title gdbinit GDB initialization scripts
50662 @c man begin SYNOPSIS gdbinit
50663 @ifset SYSTEM_GDBINIT
50664 @value{SYSTEM_GDBINIT}
50667 @ifset SYSTEM_GDBINIT_DIR
50668 @value{SYSTEM_GDBINIT_DIR}/*
50671 ~/.config/gdb/gdbinit
50679 @c man begin DESCRIPTION gdbinit
50680 These files contain @value{GDBN} commands to automatically execute during
50681 @value{GDBN} startup. The lines of contents are canned sequences of commands,
50684 the @value{GDBN} manual in node @code{Sequences}
50685 -- shell command @code{info -f gdb -n Sequences}.
50691 Please read more in
50693 the @value{GDBN} manual in node @code{Startup}
50694 -- shell command @code{info -f gdb -n Startup}.
50701 @ifset SYSTEM_GDBINIT
50702 @item @value{SYSTEM_GDBINIT}
50704 @ifclear SYSTEM_GDBINIT
50705 @item (not enabled with @code{--with-system-gdbinit} during compilation)
50707 System-wide initialization file. It is executed unless user specified
50708 @value{GDBN} option @code{-nx} or @code{-n}.
50711 the @value{GDBN} manual in node @code{System-wide configuration}
50712 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
50714 @ifset SYSTEM_GDBINIT_DIR
50715 @item @value{SYSTEM_GDBINIT_DIR}
50717 @ifclear SYSTEM_GDBINIT_DIR
50718 @item (not enabled with @code{--with-system-gdbinit-dir} during compilation)
50720 System-wide initialization directory. All files in this directory are
50721 executed on startup unless user specified @value{GDBN} option @code{-nx} or
50722 @code{-n}, as long as they have a recognized file extension.
50725 the @value{GDBN} manual in node @code{System-wide configuration}
50726 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
50729 @ref{System-wide configuration}.
50732 @item @file{~/.config/gdb/gdbinit} or @file{~/.gdbinit}
50733 User initialization file. It is executed unless user specified
50734 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
50736 @item @file{.gdbinit}
50737 Initialization file for current directory. It may need to be enabled with
50738 @value{GDBN} security command @code{set auto-load local-gdbinit}.
50741 the @value{GDBN} manual in node @code{Init File in the Current Directory}
50742 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
50745 @ref{Init File in the Current Directory}.
50750 @c man begin SEEALSO gdbinit
50752 gdb(1), @code{info -f gdb -n Startup}
50754 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
50755 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
50756 documentation are properly installed at your site, the command
50762 should give you access to the complete manual.
50764 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
50765 Richard M. Stallman and Roland H. Pesch, July 1991.
50769 @node gdb-add-index man
50770 @heading gdb-add-index
50771 @pindex gdb-add-index
50772 @anchor{gdb-add-index}
50774 @c man title gdb-add-index Add index files to speed up GDB
50776 @c man begin SYNOPSIS gdb-add-index
50777 gdb-add-index @var{filename}
50780 @c man begin DESCRIPTION gdb-add-index
50781 When @value{GDBN} finds a symbol file, it scans the symbols in the
50782 file in order to construct an internal symbol table. This lets most
50783 @value{GDBN} operations work quickly--at the cost of a delay early on.
50784 For large programs, this delay can be quite lengthy, so @value{GDBN}
50785 provides a way to build an index, which speeds up startup.
50787 To determine whether a file contains such an index, use the command
50788 @kbd{readelf -S filename}: the index is stored in a section named
50789 @code{.gdb_index}. The index file can only be produced on systems
50790 which use ELF binaries and DWARF debug information (i.e., sections
50791 named @code{.debug_*}).
50793 @command{gdb-add-index} uses @value{GDBN} and @command{objdump} found
50794 in the @env{PATH} environment variable. If you want to use different
50795 versions of these programs, you can specify them through the
50796 @env{GDB} and @env{OBJDUMP} environment variables.
50800 the @value{GDBN} manual in node @code{Index Files}
50801 -- shell command @kbd{info -f gdb -n "Index Files"}.
50808 @c man begin SEEALSO gdb-add-index
50810 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
50811 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
50812 documentation are properly installed at your site, the command
50818 should give you access to the complete manual.
50820 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
50821 Richard M. Stallman and Roland H. Pesch, July 1991.
50827 @node GNU Free Documentation License
50828 @appendix GNU Free Documentation License
50831 @node Concept Index
50832 @unnumbered Concept Index
50836 @node Command and Variable Index
50837 @unnumbered Command, Variable, and Function Index
50842 % I think something like @@colophon should be in texinfo. In the
50844 \long\def\colophon{\hbox to0pt{}\vfill
50845 \centerline{The body of this manual is set in}
50846 \centerline{\fontname\tenrm,}
50847 \centerline{with headings in {\bf\fontname\tenbf}}
50848 \centerline{and examples in {\tt\fontname\tentt}.}
50849 \centerline{{\it\fontname\tenit\/},}
50850 \centerline{{\bf\fontname\tenbf}, and}
50851 \centerline{{\sl\fontname\tensl\/}}
50852 \centerline{are used for emphasis.}\vfill}
50854 % Blame: doc@@cygnus.com, 1991.