1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988-2020 Free Software Foundation, Inc.
3 @c Copyright (C) 2020 Advanced Micro Devices, Inc. All rights reserved.
6 @c makeinfo ignores cmds prev to setfilename, so its arg cannot make use
7 @c of @set vars. However, you can override filename with makeinfo -o.
14 @settitle Debugging with @value{GDBN}
15 @setchapternewpage odd
24 @c To avoid file-name clashes between index.html and Index.html, when
25 @c the manual is produced on a Posix host and then moved to a
26 @c case-insensitive filesystem (e.g., MS-Windows), we separate the
27 @c indices into two: Concept Index and all the rest.
31 @c readline appendices use @vindex, @findex and @ftable,
32 @c annotate.texi and gdbmi use @findex.
35 @c !!set GDB manual's edition---not the same as GDB version!
36 @c This is updated by GNU Press.
39 @c !!set GDB edit command default editor
42 @c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER.
44 @c This is a dir.info fragment to support semi-automated addition of
45 @c manuals to an info tree.
46 @dircategory Software development
48 * ROCgdb: (gdb). The ROCm GNU debugger.
49 @c * gdbserver: (gdb) Server. The GNU debugging server.
53 @c man begin COPYRIGHT
55 Copyright @copyright{} 1988-2020 Free Software Foundation, Inc.
57 Copyright @copyright{} 2020 Advanced Micro Devices, Inc. All rights reserved.
59 Permission is granted to copy, distribute and/or modify this document
60 under the terms of the GNU Free Documentation License, Version 1.3 or
61 any later version published by the Free Software Foundation; with the
62 Invariant Sections being ``Free Software'' and ``Free Software Needs
63 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
64 and with the Back-Cover Texts as in (a) below.
66 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
67 this GNU Manual. Buying copies from GNU Press supports the FSF in
68 developing GNU and promoting software freedom.''
73 This file documents the @sc{gnu} debugger @value{GDBN}.
75 This is the @value{EDITION} Edition, of @cite{Debugging with
76 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
77 @ifset VERSION_PACKAGE
78 @value{VERSION_PACKAGE}
80 Version @value{GDBVN}.
86 @title Debugging with @value{GDBN}
87 @subtitle The @sc{gnu} Source-Level Debugger
89 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
90 @ifset VERSION_PACKAGE
92 @subtitle @value{VERSION_PACKAGE}
94 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
98 \hfill (Send bugs and comments on @value{GDBN} to @value{BUGURL}.)\par
99 \hfill {\it Debugging with @value{GDBN}}\par
100 \hfill \TeX{}info \texinfoversion\par
104 @c Comment out publisher until upstreamed:
105 @c @vskip 0pt plus 1filll
106 @c Published by the Free Software Foundation @*
107 @c 51 Franklin Street, Fifth Floor,
108 @c Boston, MA 02110-1301, USA@*
109 @c ISBN 978-0-9831592-3-0 @*
116 @node Top, Summary, (dir), (dir)
118 @top Debugging with @value{GDBN}
120 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
122 This is the @value{EDITION} Edition, for @value{GDBN}
123 @ifset VERSION_PACKAGE
124 @value{VERSION_PACKAGE}
126 Version @value{GDBVN}.
128 Copyright (C) 1988-2020 Free Software Foundation, Inc.
130 This edition of the GDB manual is dedicated to the memory of Fred
131 Fish. Fred was a long-standing contributor to GDB and to Free
132 software in general. We will miss him.
135 * Summary:: Summary of @value{GDBN}
136 * Sample Session:: A sample @value{GDBN} session
138 * Invocation:: Getting in and out of @value{GDBN}
139 * Commands:: @value{GDBN} commands
140 * Running:: Running programs under @value{GDBN}
141 * Stopping:: Stopping and continuing
142 * Reverse Execution:: Running programs backward
143 * Process Record and Replay:: Recording inferior's execution and replaying it
144 * Stack:: Examining the stack
145 * Source:: Examining source files
146 * Data:: Examining data
147 * Optimized Code:: Debugging optimized code
148 * Macros:: Preprocessor Macros
149 * Tracepoints:: Debugging remote targets non-intrusively
150 * Overlays:: Debugging programs that use overlays
152 * Languages:: Using @value{GDBN} with different languages
154 * Symbols:: Examining the symbol table
155 * Altering:: Altering execution
156 * GDB Files:: @value{GDBN} files
157 * Targets:: Specifying a debugging target
158 * Heterogeneous Debugging:: Debugging Heterogeneous Programs
159 * Remote Debugging:: Debugging remote programs
160 * Configurations:: Configuration-specific information
161 * Controlling GDB:: Controlling @value{GDBN}
162 * Extending GDB:: Extending @value{GDBN}
163 * Interpreters:: Command Interpreters
164 * TUI:: @value{GDBN} Text User Interface
165 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
166 * GDB/MI:: @value{GDBN}'s Machine Interface.
167 * Annotations:: @value{GDBN}'s annotation interface.
168 * JIT Interface:: Using the JIT debugging interface.
169 * In-Process Agent:: In-Process Agent
171 * GDB Bugs:: Reporting bugs in @value{GDBN}
173 @ifset SYSTEM_READLINE
174 * Command Line Editing: (rluserman). Command Line Editing
175 * Using History Interactively: (history). Using History Interactively
177 @ifclear SYSTEM_READLINE
178 * Command Line Editing:: Command Line Editing
179 * Using History Interactively:: Using History Interactively
181 * In Memoriam:: In Memoriam
182 * Formatting Documentation:: How to format and print @value{GDBN} documentation
183 * Installing GDB:: Installing @value{GDBN}
184 * Maintenance Commands:: Maintenance Commands
185 * Remote Protocol:: GDB Remote Serial Protocol
186 * Agent Expressions:: The @value{GDBN} Agent Expression Mechanism
187 * Target Descriptions:: How targets can describe themselves to
189 * Operating System Information:: Getting additional information from
191 * Trace File Format:: @value{GDBN} trace file format
192 * Index Section Format:: .gdb_index section format
193 * Man Pages:: Manual pages
194 * Copying:: GNU General Public License says
195 how you can copy and share @value{GDBN}
196 * GNU Free Documentation License:: The license for this documentation
197 * Concept Index:: Index of @value{GDBN} concepts
198 * Command and Variable Index:: Index of @value{GDBN} commands, variables,
199 functions, and Python data types
207 @unnumbered Summary of @value{GDBN}
209 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
210 going on ``inside'' another program while it executes---or what another
211 program was doing at the moment it crashed.
213 @value{GDBN} can do four main kinds of things (plus other things in support of
214 these) to help you catch bugs in the act:
218 Start your program, specifying anything that might affect its behavior.
221 Make your program stop on specified conditions.
224 Examine what has happened, when your program has stopped.
227 Change things in your program, so you can experiment with correcting the
228 effects of one bug and go on to learn about another.
231 You can use @value{GDBN} to debug programs written in C and C@t{++}.
232 For more information, see @ref{Supported Languages,,Supported Languages}.
233 For more information, see @ref{C,,C and C++}.
235 Support for D is partial. For information on D, see
239 Support for Modula-2 is partial. For information on Modula-2, see
240 @ref{Modula-2,,Modula-2}.
242 Support for OpenCL C is partial. For information on OpenCL C, see
243 @ref{OpenCL C,,OpenCL C}.
246 Debugging Pascal programs which use sets, subranges, file variables, or
247 nested functions does not currently work. @value{GDBN} does not support
248 entering expressions, printing values, or similar features using Pascal
252 @value{GDBN} can be used to debug programs written in Fortran, although
253 it may be necessary to refer to some variables with a trailing
256 @value{GDBN} can be used to debug programs written in Objective-C,
257 using either the Apple/NeXT or the GNU Objective-C runtime.
260 * Free Software:: Freely redistributable software
261 * Free Documentation:: Free Software Needs Free Documentation
262 * Contributors:: Contributors to GDB
266 @unnumberedsec Free Software
268 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
269 General Public License
270 (GPL). The GPL gives you the freedom to copy or adapt a licensed
271 program---but every person getting a copy also gets with it the
272 freedom to modify that copy (which means that they must get access to
273 the source code), and the freedom to distribute further copies.
274 Typical software companies use copyrights to limit your freedoms; the
275 Free Software Foundation uses the GPL to preserve these freedoms.
277 Fundamentally, the General Public License is a license which says that
278 you have these freedoms and that you cannot take these freedoms away
281 @node Free Documentation
282 @unnumberedsec Free Software Needs Free Documentation
284 The biggest deficiency in the free software community today is not in
285 the software---it is the lack of good free documentation that we can
286 include with the free software. Many of our most important
287 programs do not come with free reference manuals and free introductory
288 texts. Documentation is an essential part of any software package;
289 when an important free software package does not come with a free
290 manual and a free tutorial, that is a major gap. We have many such
293 Consider Perl, for instance. The tutorial manuals that people
294 normally use are non-free. How did this come about? Because the
295 authors of those manuals published them with restrictive terms---no
296 copying, no modification, source files not available---which exclude
297 them from the free software world.
299 That wasn't the first time this sort of thing happened, and it was far
300 from the last. Many times we have heard a GNU user eagerly describe a
301 manual that he is writing, his intended contribution to the community,
302 only to learn that he had ruined everything by signing a publication
303 contract to make it non-free.
305 Free documentation, like free software, is a matter of freedom, not
306 price. The problem with the non-free manual is not that publishers
307 charge a price for printed copies---that in itself is fine. (The Free
308 Software Foundation sells printed copies of manuals, too.) The
309 problem is the restrictions on the use of the manual. Free manuals
310 are available in source code form, and give you permission to copy and
311 modify. Non-free manuals do not allow this.
313 The criteria of freedom for a free manual are roughly the same as for
314 free software. Redistribution (including the normal kinds of
315 commercial redistribution) must be permitted, so that the manual can
316 accompany every copy of the program, both on-line and on paper.
318 Permission for modification of the technical content is crucial too.
319 When people modify the software, adding or changing features, if they
320 are conscientious they will change the manual too---so they can
321 provide accurate and clear documentation for the modified program. A
322 manual that leaves you no choice but to write a new manual to document
323 a changed version of the program is not really available to our
326 Some kinds of limits on the way modification is handled are
327 acceptable. For example, requirements to preserve the original
328 author's copyright notice, the distribution terms, or the list of
329 authors, are ok. It is also no problem to require modified versions
330 to include notice that they were modified. Even entire sections that
331 may not be deleted or changed are acceptable, as long as they deal
332 with nontechnical topics (like this one). These kinds of restrictions
333 are acceptable because they don't obstruct the community's normal use
336 However, it must be possible to modify all the @emph{technical}
337 content of the manual, and then distribute the result in all the usual
338 media, through all the usual channels. Otherwise, the restrictions
339 obstruct the use of the manual, it is not free, and we need another
340 manual to replace it.
342 Please spread the word about this issue. Our community continues to
343 lose manuals to proprietary publishing. If we spread the word that
344 free software needs free reference manuals and free tutorials, perhaps
345 the next person who wants to contribute by writing documentation will
346 realize, before it is too late, that only free manuals contribute to
347 the free software community.
349 If you are writing documentation, please insist on publishing it under
350 the GNU Free Documentation License or another free documentation
351 license. Remember that this decision requires your approval---you
352 don't have to let the publisher decide. Some commercial publishers
353 will use a free license if you insist, but they will not propose the
354 option; it is up to you to raise the issue and say firmly that this is
355 what you want. If the publisher you are dealing with refuses, please
356 try other publishers. If you're not sure whether a proposed license
357 is free, write to @email{licensing@@gnu.org}.
359 You can encourage commercial publishers to sell more free, copylefted
360 manuals and tutorials by buying them, and particularly by buying
361 copies from the publishers that paid for their writing or for major
362 improvements. Meanwhile, try to avoid buying non-free documentation
363 at all. Check the distribution terms of a manual before you buy it,
364 and insist that whoever seeks your business must respect your freedom.
365 Check the history of the book, and try to reward the publishers that
366 have paid or pay the authors to work on it.
368 The Free Software Foundation maintains a list of free documentation
369 published by other publishers, at
370 @url{http://www.fsf.org/doc/other-free-books.html}.
373 @unnumberedsec Contributors to @value{GDBN}
375 Richard Stallman was the original author of @value{GDBN}, and of many
376 other @sc{gnu} programs. Many others have contributed to its
377 development. This section attempts to credit major contributors. One
378 of the virtues of free software is that everyone is free to contribute
379 to it; with regret, we cannot actually acknowledge everyone here. The
380 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
381 blow-by-blow account.
383 Changes much prior to version 2.0 are lost in the mists of time.
386 @emph{Plea:} Additions to this section are particularly welcome. If you
387 or your friends (or enemies, to be evenhanded) have been unfairly
388 omitted from this list, we would like to add your names!
391 So that they may not regard their many labors as thankless, we
392 particularly thank those who shepherded @value{GDBN} through major
394 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
395 Jim Blandy (release 4.18);
396 Jason Molenda (release 4.17);
397 Stan Shebs (release 4.14);
398 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
399 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
400 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
401 Jim Kingdon (releases 3.5, 3.4, and 3.3);
402 and Randy Smith (releases 3.2, 3.1, and 3.0).
404 Richard Stallman, assisted at various times by Peter TerMaat, Chris
405 Hanson, and Richard Mlynarik, handled releases through 2.8.
407 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
408 in @value{GDBN}, with significant additional contributions from Per
409 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
410 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
411 much general update work leading to release 3.0).
413 @value{GDBN} uses the BFD subroutine library to examine multiple
414 object-file formats; BFD was a joint project of David V.
415 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
417 David Johnson wrote the original COFF support; Pace Willison did
418 the original support for encapsulated COFF.
420 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
422 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
423 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
425 Jean-Daniel Fekete contributed Sun 386i support.
426 Chris Hanson improved the HP9000 support.
427 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
428 David Johnson contributed Encore Umax support.
429 Jyrki Kuoppala contributed Altos 3068 support.
430 Jeff Law contributed HP PA and SOM support.
431 Keith Packard contributed NS32K support.
432 Doug Rabson contributed Acorn Risc Machine support.
433 Bob Rusk contributed Harris Nighthawk CX-UX support.
434 Chris Smith contributed Convex support (and Fortran debugging).
435 Jonathan Stone contributed Pyramid support.
436 Michael Tiemann contributed SPARC support.
437 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
438 Pace Willison contributed Intel 386 support.
439 Jay Vosburgh contributed Symmetry support.
440 Marko Mlinar contributed OpenRISC 1000 support.
442 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
444 Rich Schaefer and Peter Schauer helped with support of SunOS shared
447 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
448 about several machine instruction sets.
450 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
451 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
452 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
453 and RDI targets, respectively.
455 Brian Fox is the author of the readline libraries providing
456 command-line editing and command history.
458 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
459 Modula-2 support, and contributed the Languages chapter of this manual.
461 Fred Fish wrote most of the support for Unix System Vr4.
462 He also enhanced the command-completion support to cover C@t{++} overloaded
465 Hitachi America (now Renesas America), Ltd. sponsored the support for
466 H8/300, H8/500, and Super-H processors.
468 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
470 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
473 Toshiba sponsored the support for the TX39 Mips processor.
475 Matsushita sponsored the support for the MN10200 and MN10300 processors.
477 Fujitsu sponsored the support for SPARClite and FR30 processors.
479 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
482 Michael Snyder added support for tracepoints.
484 Stu Grossman wrote gdbserver.
486 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
487 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
489 The following people at the Hewlett-Packard Company contributed
490 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
491 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
492 compiler, and the Text User Interface (nee Terminal User Interface):
493 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
494 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
495 provided HP-specific information in this manual.
497 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
498 Robert Hoehne made significant contributions to the DJGPP port.
500 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
501 development since 1991. Cygnus engineers who have worked on @value{GDBN}
502 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
503 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
504 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
505 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
506 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
507 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
508 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
509 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
510 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
511 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
512 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
513 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
514 Zuhn have made contributions both large and small.
516 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
517 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
519 Jim Blandy added support for preprocessor macros, while working for Red
522 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
523 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
524 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
525 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
526 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
527 with the migration of old architectures to this new framework.
529 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
530 unwinder framework, this consisting of a fresh new design featuring
531 frame IDs, independent frame sniffers, and the sentinel frame. Mark
532 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
533 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
534 trad unwinders. The architecture-specific changes, each involving a
535 complete rewrite of the architecture's frame code, were carried out by
536 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
537 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
538 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
539 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
542 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
543 Tensilica, Inc.@: contributed support for Xtensa processors. Others
544 who have worked on the Xtensa port of @value{GDBN} in the past include
545 Steve Tjiang, John Newlin, and Scott Foehner.
547 Michael Eager and staff of Xilinx, Inc., contributed support for the
548 Xilinx MicroBlaze architecture.
550 Initial support for the FreeBSD/mips target and native configuration
551 was developed by SRI International and the University of Cambridge
552 Computer Laboratory under DARPA/AFRL contract FA8750-10-C-0237
553 ("CTSRD"), as part of the DARPA CRASH research programme.
555 Initial support for the FreeBSD/riscv target and native configuration
556 was developed by SRI International and the University of Cambridge
557 Computer Laboratory (Department of Computer Science and Technology)
558 under DARPA contract HR0011-18-C-0016 ("ECATS"), as part of the DARPA
559 SSITH research programme.
561 The original port to the OpenRISC 1000 is believed to be due to
562 Alessandro Forin and Per Bothner. More recent ports have been the work
563 of Jeremy Bennett, Franck Jullien, Stefan Wallentowitz and
566 Initial support for heterogeneous program debugging and the
567 @acronym{AMD GPU} targets was developed by the following people at the
568 Advanced Micro Devices company: Scott Linder, Laurent Morichetti,
569 Qingchuan Shi, Tony Tye, and Zoran Zaric.
572 @chapter A Sample @value{GDBN} Session
574 You can use this manual at your leisure to read all about @value{GDBN}.
575 However, a handful of commands are enough to get started using the
576 debugger. This chapter illustrates those commands.
579 In this sample session, we emphasize user input like this: @b{input},
580 to make it easier to pick out from the surrounding output.
583 @c FIXME: this example may not be appropriate for some configs, where
584 @c FIXME...primary interest is in remote use.
586 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
587 processor) exhibits the following bug: sometimes, when we change its
588 quote strings from the default, the commands used to capture one macro
589 definition within another stop working. In the following short @code{m4}
590 session, we define a macro @code{foo} which expands to @code{0000}; we
591 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
592 same thing. However, when we change the open quote string to
593 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
594 procedure fails to define a new synonym @code{baz}:
603 @b{define(bar,defn(`foo'))}
607 @b{changequote(<QUOTE>,<UNQUOTE>)}
609 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
612 m4: End of input: 0: fatal error: EOF in string
616 Let us use @value{GDBN} to try to see what is going on.
619 $ @b{@value{GDBP} m4}
620 @c FIXME: this falsifies the exact text played out, to permit smallbook
621 @c FIXME... format to come out better.
622 @value{GDBN} is free software and you are welcome to distribute copies
623 of it under certain conditions; type "show copying" to see
625 There is absolutely no warranty for @value{GDBN}; type "show warranty"
628 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
633 @value{GDBN} reads only enough symbol data to know where to find the
634 rest when needed; as a result, the first prompt comes up very quickly.
635 We now tell @value{GDBN} to use a narrower display width than usual, so
636 that examples fit in this manual.
639 (@value{GDBP}) @b{set width 70}
643 We need to see how the @code{m4} built-in @code{changequote} works.
644 Having looked at the source, we know the relevant subroutine is
645 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
646 @code{break} command.
649 (@value{GDBP}) @b{break m4_changequote}
650 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
654 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
655 control; as long as control does not reach the @code{m4_changequote}
656 subroutine, the program runs as usual:
659 (@value{GDBP}) @b{run}
660 Starting program: /work/Editorial/gdb/gnu/m4/m4
668 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
669 suspends execution of @code{m4}, displaying information about the
670 context where it stops.
673 @b{changequote(<QUOTE>,<UNQUOTE>)}
675 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
677 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
681 Now we use the command @code{n} (@code{next}) to advance execution to
682 the next line of the current function.
686 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
691 @code{set_quotes} looks like a promising subroutine. We can go into it
692 by using the command @code{s} (@code{step}) instead of @code{next}.
693 @code{step} goes to the next line to be executed in @emph{any}
694 subroutine, so it steps into @code{set_quotes}.
698 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
700 530 if (lquote != def_lquote)
704 The display that shows the subroutine where @code{m4} is now
705 suspended (and its arguments) is called a stack frame display. It
706 shows a summary of the stack. We can use the @code{backtrace}
707 command (which can also be spelled @code{bt}), to see where we are
708 in the stack as a whole: the @code{backtrace} command displays a
709 stack frame for each active subroutine.
712 (@value{GDBP}) @b{bt}
713 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
715 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
717 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
718 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
720 #4 0x79dc in expand_input () at macro.c:40
721 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
725 We step through a few more lines to see what happens. The first two
726 times, we can use @samp{s}; the next two times we use @code{n} to avoid
727 falling into the @code{xstrdup} subroutine.
731 0x3b5c 532 if (rquote != def_rquote)
733 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
734 def_lquote : xstrdup(lq);
736 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
739 538 len_lquote = strlen(rquote);
743 The last line displayed looks a little odd; we can examine the variables
744 @code{lquote} and @code{rquote} to see if they are in fact the new left
745 and right quotes we specified. We use the command @code{p}
746 (@code{print}) to see their values.
749 (@value{GDBP}) @b{p lquote}
750 $1 = 0x35d40 "<QUOTE>"
751 (@value{GDBP}) @b{p rquote}
752 $2 = 0x35d50 "<UNQUOTE>"
756 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
757 To look at some context, we can display ten lines of source
758 surrounding the current line with the @code{l} (@code{list}) command.
764 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
766 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
769 538 len_lquote = strlen(rquote);
770 539 len_rquote = strlen(lquote);
777 Let us step past the two lines that set @code{len_lquote} and
778 @code{len_rquote}, and then examine the values of those variables.
782 539 len_rquote = strlen(lquote);
785 (@value{GDBP}) @b{p len_lquote}
787 (@value{GDBP}) @b{p len_rquote}
792 That certainly looks wrong, assuming @code{len_lquote} and
793 @code{len_rquote} are meant to be the lengths of @code{lquote} and
794 @code{rquote} respectively. We can set them to better values using
795 the @code{p} command, since it can print the value of
796 any expression---and that expression can include subroutine calls and
800 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
802 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
807 Is that enough to fix the problem of using the new quotes with the
808 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
809 executing with the @code{c} (@code{continue}) command, and then try the
810 example that caused trouble initially:
816 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
823 Success! The new quotes now work just as well as the default ones. The
824 problem seems to have been just the two typos defining the wrong
825 lengths. We allow @code{m4} exit by giving it an EOF as input:
829 Program exited normally.
833 The message @samp{Program exited normally.} is from @value{GDBN}; it
834 indicates @code{m4} has finished executing. We can end our @value{GDBN}
835 session with the @value{GDBN} @code{quit} command.
838 (@value{GDBP}) @b{quit}
842 @chapter Getting In and Out of @value{GDBN}
844 This chapter discusses how to start @value{GDBN}, and how to get out of it.
848 type @samp{@value{GDBP}} to start @value{GDBN}.
850 type @kbd{quit} or @kbd{Ctrl-d} to exit.
854 * Invoking GDB:: How to start @value{GDBN}
855 * Quitting GDB:: How to quit @value{GDBN}
856 * Shell Commands:: How to use shell commands inside @value{GDBN}
857 * Logging Output:: How to log @value{GDBN}'s output to a file
861 @section Invoking @value{GDBN}
863 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
864 @value{GDBN} reads commands from the terminal until you tell it to exit.
866 You can also run @code{@value{GDBP}} with a variety of arguments and options,
867 to specify more of your debugging environment at the outset.
869 The command-line options described here are designed
870 to cover a variety of situations; in some environments, some of these
871 options may effectively be unavailable.
873 The most usual way to start @value{GDBN} is with one argument,
874 specifying an executable program:
877 @value{GDBP} @var{program}
881 You can also start with both an executable program and a core file
885 @value{GDBP} @var{program} @var{core}
888 You can, instead, specify a process ID as a second argument or use option
889 @code{-p}, if you want to debug a running process:
892 @value{GDBP} @var{program} 1234
897 would attach @value{GDBN} to process @code{1234}. With option @option{-p} you
898 can omit the @var{program} filename.
900 Taking advantage of the second command-line argument requires a fairly
901 complete operating system; when you use @value{GDBN} as a remote
902 debugger attached to a bare board, there may not be any notion of
903 ``process'', and there is often no way to get a core dump. @value{GDBN}
904 will warn you if it is unable to attach or to read core dumps.
906 You can optionally have @code{@value{GDBP}} pass any arguments after the
907 executable file to the inferior using @code{--args}. This option stops
910 @value{GDBP} --args gcc -O2 -c foo.c
912 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
913 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
915 You can run @code{@value{GDBP}} without printing the front material, which describes
916 @value{GDBN}'s non-warranty, by specifying @code{--silent}
917 (or @code{-q}/@code{--quiet}):
920 @value{GDBP} --silent
924 You can further control how @value{GDBN} starts up by using command-line
925 options. @value{GDBN} itself can remind you of the options available.
935 to display all available options and briefly describe their use
936 (@samp{@value{GDBP} -h} is a shorter equivalent).
938 All options and command line arguments you give are processed
939 in sequential order. The order makes a difference when the
940 @samp{-x} option is used.
944 * File Options:: Choosing files
945 * Mode Options:: Choosing modes
946 * Startup:: What @value{GDBN} does during startup
950 @subsection Choosing Files
952 When @value{GDBN} starts, it reads any arguments other than options as
953 specifying an executable file and core file (or process ID). This is
954 the same as if the arguments were specified by the @samp{-se} and
955 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
956 first argument that does not have an associated option flag as
957 equivalent to the @samp{-se} option followed by that argument; and the
958 second argument that does not have an associated option flag, if any, as
959 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
960 If the second argument begins with a decimal digit, @value{GDBN} will
961 first attempt to attach to it as a process, and if that fails, attempt
962 to open it as a corefile. If you have a corefile whose name begins with
963 a digit, you can prevent @value{GDBN} from treating it as a pid by
964 prefixing it with @file{./}, e.g.@: @file{./12345}.
966 If @value{GDBN} has not been configured to included core file support,
967 such as for most embedded targets, then it will complain about a second
968 argument and ignore it.
970 Many options have both long and short forms; both are shown in the
971 following list. @value{GDBN} also recognizes the long forms if you truncate
972 them, so long as enough of the option is present to be unambiguous.
973 (If you prefer, you can flag option arguments with @samp{--} rather
974 than @samp{-}, though we illustrate the more usual convention.)
976 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
977 @c way, both those who look for -foo and --foo in the index, will find
981 @item -symbols @var{file}
983 @cindex @code{--symbols}
985 Read symbol table from file @var{file}.
987 @item -exec @var{file}
989 @cindex @code{--exec}
991 Use file @var{file} as the executable file to execute when appropriate,
992 and for examining pure data in conjunction with a core dump.
996 Read symbol table from file @var{file} and use it as the executable
999 @item -core @var{file}
1000 @itemx -c @var{file}
1001 @cindex @code{--core}
1003 Use file @var{file} as a core dump to examine.
1005 @item -pid @var{number}
1006 @itemx -p @var{number}
1007 @cindex @code{--pid}
1009 Connect to process ID @var{number}, as with the @code{attach} command.
1011 @item -command @var{file}
1012 @itemx -x @var{file}
1013 @cindex @code{--command}
1015 Execute commands from file @var{file}. The contents of this file is
1016 evaluated exactly as the @code{source} command would.
1017 @xref{Command Files,, Command files}.
1019 @item -eval-command @var{command}
1020 @itemx -ex @var{command}
1021 @cindex @code{--eval-command}
1023 Execute a single @value{GDBN} command.
1025 This option may be used multiple times to call multiple commands. It may
1026 also be interleaved with @samp{-command} as required.
1029 @value{GDBP} -ex 'target sim' -ex 'load' \
1030 -x setbreakpoints -ex 'run' a.out
1033 @item -init-command @var{file}
1034 @itemx -ix @var{file}
1035 @cindex @code{--init-command}
1037 Execute commands from file @var{file} before loading the inferior (but
1038 after loading gdbinit files).
1041 @item -init-eval-command @var{command}
1042 @itemx -iex @var{command}
1043 @cindex @code{--init-eval-command}
1045 Execute a single @value{GDBN} command before loading the inferior (but
1046 after loading gdbinit files).
1049 @item -directory @var{directory}
1050 @itemx -d @var{directory}
1051 @cindex @code{--directory}
1053 Add @var{directory} to the path to search for source and script files.
1057 @cindex @code{--readnow}
1059 Read each symbol file's entire symbol table immediately, rather than
1060 the default, which is to read it incrementally as it is needed.
1061 This makes startup slower, but makes future operations faster.
1064 @anchor{--readnever}
1065 @cindex @code{--readnever}, command-line option
1066 Do not read each symbol file's symbolic debug information. This makes
1067 startup faster but at the expense of not being able to perform
1068 symbolic debugging. DWARF unwind information is also not read,
1069 meaning backtraces may become incomplete or inaccurate. One use of
1070 this is when a user simply wants to do the following sequence: attach,
1071 dump core, detach. Loading the debugging information in this case is
1072 an unnecessary cause of delay.
1076 @subsection Choosing Modes
1078 You can run @value{GDBN} in various alternative modes---for example, in
1079 batch mode or quiet mode.
1087 Do not execute commands found in any initialization file.
1088 There are three init files, loaded in the following order:
1091 @item @file{system.gdbinit}
1092 This is the system-wide init file.
1093 Its location is specified with the @code{--with-system-gdbinit}
1094 configure option (@pxref{System-wide configuration}).
1095 It is loaded first when @value{GDBN} starts, before command line options
1096 have been processed.
1097 @item @file{system.gdbinit.d}
1098 This is the system-wide init directory.
1099 Its location is specified with the @code{--with-system-gdbinit-dir}
1100 configure option (@pxref{System-wide configuration}).
1101 Files in this directory are loaded in alphabetical order immediately after
1102 system.gdbinit (if enabled) when @value{GDBN} starts, before command line
1103 options have been processed. Files need to have a recognized scripting
1104 language extension (@file{.py}/@file{.scm}) or be named with a @file{.gdb}
1105 extension to be interpreted as regular @value{GDBN} commands. @value{GDBN}
1106 will not recurse into any subdirectories of this directory.
1107 @item @file{~/.gdbinit}
1108 This is the init file in your home directory.
1109 It is loaded next, after @file{system.gdbinit}, and before
1110 command options have been processed.
1111 @item @file{./.gdbinit}
1112 This is the init file in the current directory.
1113 It is loaded last, after command line options other than @code{-x} and
1114 @code{-ex} have been processed. Command line options @code{-x} and
1115 @code{-ex} are processed last, after @file{./.gdbinit} has been loaded.
1118 For further documentation on startup processing, @xref{Startup}.
1119 For documentation on how to write command files,
1120 @xref{Command Files,,Command Files}.
1125 Do not execute commands found in @file{~/.gdbinit}, the init file
1126 in your home directory.
1132 @cindex @code{--quiet}
1133 @cindex @code{--silent}
1135 ``Quiet''. Do not print the introductory and copyright messages. These
1136 messages are also suppressed in batch mode.
1139 @cindex @code{--batch}
1140 Run in batch mode. Exit with status @code{0} after processing all the
1141 command files specified with @samp{-x} (and all commands from
1142 initialization files, if not inhibited with @samp{-n}). Exit with
1143 nonzero status if an error occurs in executing the @value{GDBN} commands
1144 in the command files. Batch mode also disables pagination, sets unlimited
1145 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1146 off} were in effect (@pxref{Messages/Warnings}).
1148 Batch mode may be useful for running @value{GDBN} as a filter, for
1149 example to download and run a program on another computer; in order to
1150 make this more useful, the message
1153 Program exited normally.
1157 (which is ordinarily issued whenever a program running under
1158 @value{GDBN} control terminates) is not issued when running in batch
1162 @cindex @code{--batch-silent}
1163 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1164 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1165 unaffected). This is much quieter than @samp{-silent} and would be useless
1166 for an interactive session.
1168 This is particularly useful when using targets that give @samp{Loading section}
1169 messages, for example.
1171 Note that targets that give their output via @value{GDBN}, as opposed to
1172 writing directly to @code{stdout}, will also be made silent.
1174 @item -return-child-result
1175 @cindex @code{--return-child-result}
1176 The return code from @value{GDBN} will be the return code from the child
1177 process (the process being debugged), with the following exceptions:
1181 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1182 internal error. In this case the exit code is the same as it would have been
1183 without @samp{-return-child-result}.
1185 The user quits with an explicit value. E.g., @samp{quit 1}.
1187 The child process never runs, or is not allowed to terminate, in which case
1188 the exit code will be -1.
1191 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1192 when @value{GDBN} is being used as a remote program loader or simulator
1197 @cindex @code{--nowindows}
1199 ``No windows''. If @value{GDBN} comes with a graphical user interface
1200 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1201 interface. If no GUI is available, this option has no effect.
1205 @cindex @code{--windows}
1207 If @value{GDBN} includes a GUI, then this option requires it to be
1210 @item -cd @var{directory}
1212 Run @value{GDBN} using @var{directory} as its working directory,
1213 instead of the current directory.
1215 @item -data-directory @var{directory}
1216 @itemx -D @var{directory}
1217 @cindex @code{--data-directory}
1219 Run @value{GDBN} using @var{directory} as its data directory.
1220 The data directory is where @value{GDBN} searches for its
1221 auxiliary files. @xref{Data Files}.
1225 @cindex @code{--fullname}
1227 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1228 subprocess. It tells @value{GDBN} to output the full file name and line
1229 number in a standard, recognizable fashion each time a stack frame is
1230 displayed (which includes each time your program stops). This
1231 recognizable format looks like two @samp{\032} characters, followed by
1232 the file name, line number and character position separated by colons,
1233 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1234 @samp{\032} characters as a signal to display the source code for the
1237 @item -annotate @var{level}
1238 @cindex @code{--annotate}
1239 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1240 effect is identical to using @samp{set annotate @var{level}}
1241 (@pxref{Annotations}). The annotation @var{level} controls how much
1242 information @value{GDBN} prints together with its prompt, values of
1243 expressions, source lines, and other types of output. Level 0 is the
1244 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1245 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1246 that control @value{GDBN}, and level 2 has been deprecated.
1248 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1252 @cindex @code{--args}
1253 Change interpretation of command line so that arguments following the
1254 executable file are passed as command line arguments to the inferior.
1255 This option stops option processing.
1257 @item -baud @var{bps}
1259 @cindex @code{--baud}
1261 Set the line speed (baud rate or bits per second) of any serial
1262 interface used by @value{GDBN} for remote debugging.
1264 @item -l @var{timeout}
1266 Set the timeout (in seconds) of any communication used by @value{GDBN}
1267 for remote debugging.
1269 @item -tty @var{device}
1270 @itemx -t @var{device}
1271 @cindex @code{--tty}
1273 Run using @var{device} for your program's standard input and output.
1274 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1276 @c resolve the situation of these eventually
1278 @cindex @code{--tui}
1279 Activate the @dfn{Text User Interface} when starting. The Text User
1280 Interface manages several text windows on the terminal, showing
1281 source, assembly, registers and @value{GDBN} command outputs
1282 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Do not use this
1283 option if you run @value{GDBN} from Emacs (@pxref{Emacs, ,
1284 Using @value{GDBN} under @sc{gnu} Emacs}).
1286 @item -interpreter @var{interp}
1287 @cindex @code{--interpreter}
1288 Use the interpreter @var{interp} for interface with the controlling
1289 program or device. This option is meant to be set by programs which
1290 communicate with @value{GDBN} using it as a back end.
1291 @xref{Interpreters, , Command Interpreters}.
1293 @samp{--interpreter=mi} (or @samp{--interpreter=mi3}) causes
1294 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} version 3 (@pxref{GDB/MI, ,
1295 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 9.1. @sc{gdb/mi}
1296 version 2 (@code{mi2}), included in @value{GDBN} 6.0 and version 1 (@code{mi1}),
1297 included in @value{GDBN} 5.3, are also available. Earlier @sc{gdb/mi}
1298 interfaces are no longer supported.
1301 @cindex @code{--write}
1302 Open the executable and core files for both reading and writing. This
1303 is equivalent to the @samp{set write on} command inside @value{GDBN}
1307 @cindex @code{--statistics}
1308 This option causes @value{GDBN} to print statistics about time and
1309 memory usage after it completes each command and returns to the prompt.
1312 @cindex @code{--version}
1313 This option causes @value{GDBN} to print its version number and
1314 no-warranty blurb, and exit.
1316 @item -configuration
1317 @cindex @code{--configuration}
1318 This option causes @value{GDBN} to print details about its build-time
1319 configuration parameters, and then exit. These details can be
1320 important when reporting @value{GDBN} bugs (@pxref{GDB Bugs}).
1325 @subsection What @value{GDBN} Does During Startup
1326 @cindex @value{GDBN} startup
1328 Here's the description of what @value{GDBN} does during session startup:
1332 Sets up the command interpreter as specified by the command line
1333 (@pxref{Mode Options, interpreter}).
1337 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1338 used when building @value{GDBN}; @pxref{System-wide configuration,
1339 ,System-wide configuration and settings}) and the files in the system-wide
1340 gdbinit directory (if @option{--with-system-gdbinit-dir} was used) and executes
1341 all the commands in those files. The files need to be named with a @file{.gdb}
1342 extension to be interpreted as @value{GDBN} commands, or they can be written
1343 in a supported scripting language with an appropriate file extension.
1345 @anchor{Home Directory Init File}
1347 Reads the init file (if any) in your home directory@footnote{On
1348 DOS/Windows systems, the home directory is the one pointed to by the
1349 @code{HOME} environment variable.} and executes all the commands in
1352 @anchor{Option -init-eval-command}
1354 Executes commands and command files specified by the @samp{-iex} and
1355 @samp{-ix} options in their specified order. Usually you should use the
1356 @samp{-ex} and @samp{-x} options instead, but this way you can apply
1357 settings before @value{GDBN} init files get executed and before inferior
1361 Processes command line options and operands.
1363 @anchor{Init File in the Current Directory during Startup}
1365 Reads and executes the commands from init file (if any) in the current
1366 working directory as long as @samp{set auto-load local-gdbinit} is set to
1367 @samp{on} (@pxref{Init File in the Current Directory}).
1368 This is only done if the current directory is
1369 different from your home directory. Thus, you can have more than one
1370 init file, one generic in your home directory, and another, specific
1371 to the program you are debugging, in the directory where you invoke
1375 If the command line specified a program to debug, or a process to
1376 attach to, or a core file, @value{GDBN} loads any auto-loaded
1377 scripts provided for the program or for its loaded shared libraries.
1378 @xref{Auto-loading}.
1380 If you wish to disable the auto-loading during startup,
1381 you must do something like the following:
1384 $ gdb -iex "set auto-load python-scripts off" myprogram
1387 Option @samp{-ex} does not work because the auto-loading is then turned
1391 Executes commands and command files specified by the @samp{-ex} and
1392 @samp{-x} options in their specified order. @xref{Command Files}, for
1393 more details about @value{GDBN} command files.
1396 Reads the command history recorded in the @dfn{history file}.
1397 @xref{Command History}, for more details about the command history and the
1398 files where @value{GDBN} records it.
1401 Init files use the same syntax as @dfn{command files} (@pxref{Command
1402 Files}) and are processed by @value{GDBN} in the same way. The init
1403 file in your home directory can set options (such as @samp{set
1404 complaints}) that affect subsequent processing of command line options
1405 and operands. Init files are not executed if you use the @samp{-nx}
1406 option (@pxref{Mode Options, ,Choosing Modes}).
1408 To display the list of init files loaded by gdb at startup, you
1409 can use @kbd{gdb --help}.
1411 @cindex init file name
1412 @cindex @file{.gdbinit}
1413 @cindex @file{gdb.ini}
1414 The @value{GDBN} init files are normally called @file{.gdbinit}.
1415 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1416 the limitations of file names imposed by DOS filesystems. The Windows
1417 port of @value{GDBN} uses the standard name, but if it finds a
1418 @file{gdb.ini} file in your home directory, it warns you about that
1419 and suggests to rename the file to the standard name.
1423 @section Quitting @value{GDBN}
1424 @cindex exiting @value{GDBN}
1425 @cindex leaving @value{GDBN}
1428 @kindex quit @r{[}@var{expression}@r{]}
1429 @kindex q @r{(@code{quit})}
1430 @item quit @r{[}@var{expression}@r{]}
1432 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1433 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1434 do not supply @var{expression}, @value{GDBN} will terminate normally;
1435 otherwise it will terminate using the result of @var{expression} as the
1440 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1441 terminates the action of any @value{GDBN} command that is in progress and
1442 returns to @value{GDBN} command level. It is safe to type the interrupt
1443 character at any time because @value{GDBN} does not allow it to take effect
1444 until a time when it is safe.
1446 If you have been using @value{GDBN} to control an attached process or
1447 device, you can release it with the @code{detach} command
1448 (@pxref{Attach, ,Debugging an Already-running Process}).
1450 @node Shell Commands
1451 @section Shell Commands
1453 If you need to execute occasional shell commands during your
1454 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1455 just use the @code{shell} command.
1460 @cindex shell escape
1461 @item shell @var{command-string}
1462 @itemx !@var{command-string}
1463 Invoke a standard shell to execute @var{command-string}.
1464 Note that no space is needed between @code{!} and @var{command-string}.
1465 If it exists, the environment variable @code{SHELL} determines which
1466 shell to run. Otherwise @value{GDBN} uses the default shell
1467 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1470 The utility @code{make} is often needed in development environments.
1471 You do not have to use the @code{shell} command for this purpose in
1476 @cindex calling make
1477 @item make @var{make-args}
1478 Execute the @code{make} program with the specified
1479 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1485 @cindex send the output of a gdb command to a shell command
1487 @item pipe [@var{command}] | @var{shell_command}
1488 @itemx | [@var{command}] | @var{shell_command}
1489 @itemx pipe -d @var{delim} @var{command} @var{delim} @var{shell_command}
1490 @itemx | -d @var{delim} @var{command} @var{delim} @var{shell_command}
1491 Executes @var{command} and sends its output to @var{shell_command}.
1492 Note that no space is needed around @code{|}.
1493 If no @var{command} is provided, the last command executed is repeated.
1495 In case the @var{command} contains a @code{|}, the option @code{-d @var{delim}}
1496 can be used to specify an alternate delimiter string @var{delim} that separates
1497 the @var{command} from the @var{shell_command}.
1502 (@value{GDBP}) p var
1512 (@value{GDBP}) pipe p var|wc
1514 (@value{GDBP}) |p var|wc -l
1518 (@value{GDBP}) p /x var
1526 (@value{GDBP}) ||grep red
1530 (@value{GDBP}) | -d ! echo this contains a | char\n ! sed -e 's/|/PIPE/'
1531 this contains a PIPE char
1532 (@value{GDBP}) | -d xxx echo this contains a | char!\n xxx sed -e 's/|/PIPE/'
1533 this contains a PIPE char!
1539 The convenience variables @code{$_shell_exitcode} and @code{$_shell_exitsignal}
1540 can be used to examine the exit status of the last shell command launched
1541 by @code{shell}, @code{make}, @code{pipe} and @code{|}.
1542 @xref{Convenience Vars,, Convenience Variables}.
1544 @node Logging Output
1545 @section Logging Output
1546 @cindex logging @value{GDBN} output
1547 @cindex save @value{GDBN} output to a file
1549 You may want to save the output of @value{GDBN} commands to a file.
1550 There are several commands to control @value{GDBN}'s logging.
1554 @item set logging on
1556 @item set logging off
1558 @cindex logging file name
1559 @item set logging file @var{file}
1560 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1561 @item set logging overwrite [on|off]
1562 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1563 you want @code{set logging on} to overwrite the logfile instead.
1564 @item set logging redirect [on|off]
1565 By default, @value{GDBN} output will go to both the terminal and the logfile.
1566 Set @code{redirect} if you want output to go only to the log file.
1567 @item set logging debugredirect [on|off]
1568 By default, @value{GDBN} debug output will go to both the terminal and the logfile.
1569 Set @code{debugredirect} if you want debug output to go only to the log file.
1570 @kindex show logging
1572 Show the current values of the logging settings.
1575 You can also redirect the output of a @value{GDBN} command to a
1576 shell command. @xref{pipe}.
1578 @chapter @value{GDBN} Commands
1580 You can abbreviate a @value{GDBN} command to the first few letters of the command
1581 name, if that abbreviation is unambiguous; and you can repeat certain
1582 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1583 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1584 show you the alternatives available, if there is more than one possibility).
1587 * Command Syntax:: How to give commands to @value{GDBN}
1588 * Command Settings:: How to change default behavior of commands
1589 * Completion:: Command completion
1590 * Command Options:: Command options
1591 * Help:: How to ask @value{GDBN} for help
1594 @node Command Syntax
1595 @section Command Syntax
1597 A @value{GDBN} command is a single line of input. There is no limit on
1598 how long it can be. It starts with a command name, which is followed by
1599 arguments whose meaning depends on the command name. For example, the
1600 command @code{step} accepts an argument which is the number of times to
1601 step, as in @samp{step 5}. You can also use the @code{step} command
1602 with no arguments. Some commands do not allow any arguments.
1604 @cindex abbreviation
1605 @value{GDBN} command names may always be truncated if that abbreviation is
1606 unambiguous. Other possible command abbreviations are listed in the
1607 documentation for individual commands. In some cases, even ambiguous
1608 abbreviations are allowed; for example, @code{s} is specially defined as
1609 equivalent to @code{step} even though there are other commands whose
1610 names start with @code{s}. You can test abbreviations by using them as
1611 arguments to the @code{help} command.
1613 @cindex repeating commands
1614 @kindex RET @r{(repeat last command)}
1615 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1616 repeat the previous command. Certain commands (for example, @code{run})
1617 will not repeat this way; these are commands whose unintentional
1618 repetition might cause trouble and which you are unlikely to want to
1619 repeat. User-defined commands can disable this feature; see
1620 @ref{Define, dont-repeat}.
1622 The @code{list} and @code{x} commands, when you repeat them with
1623 @key{RET}, construct new arguments rather than repeating
1624 exactly as typed. This permits easy scanning of source or memory.
1626 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1627 output, in a way similar to the common utility @code{more}
1628 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1629 @key{RET} too many in this situation, @value{GDBN} disables command
1630 repetition after any command that generates this sort of display.
1632 @kindex # @r{(a comment)}
1634 Any text from a @kbd{#} to the end of the line is a comment; it does
1635 nothing. This is useful mainly in command files (@pxref{Command
1636 Files,,Command Files}).
1638 @cindex repeating command sequences
1639 @kindex Ctrl-o @r{(operate-and-get-next)}
1640 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1641 commands. This command accepts the current line, like @key{RET}, and
1642 then fetches the next line relative to the current line from the history
1646 @node Command Settings
1647 @section Command Settings
1648 @cindex default behavior of commands, changing
1649 @cindex default settings, changing
1651 Many commands change their behavior according to command-specific
1652 variables or settings. These settings can be changed with the
1653 @code{set} subcommands. For example, the @code{print} command
1654 (@pxref{Data, ,Examining Data}) prints arrays differently depending on
1655 settings changeable with the commands @code{set print elements
1656 NUMBER-OF-ELEMENTS} and @code{set print array-indexes}, among others.
1658 You can change these settings to your preference in the gdbinit files
1659 loaded at @value{GDBN} startup. @xref{Startup}.
1661 The settings can also be changed interactively during the debugging
1662 session. For example, to change the limit of array elements to print,
1663 you can do the following:
1665 (@value{GDBP}) set print elements 10
1666 (@value{GDBP}) print some_array
1667 $1 = @{0, 10, 20, 30, 40, 50, 60, 70, 80, 90...@}
1670 The above @code{set print elements 10} command changes the number of
1671 elements to print from the default of 200 to 10. If you only intend
1672 this limit of 10 to be used for printing @code{some_array}, then you
1673 must restore the limit back to 200, with @code{set print elements
1676 Some commands allow overriding settings with command options. For
1677 example, the @code{print} command supports a number of options that
1678 allow overriding relevant global print settings as set by @code{set
1679 print} subcommands. @xref{print options}. The example above could be
1682 (@value{GDBP}) print -elements 10 -- some_array
1683 $1 = @{0, 10, 20, 30, 40, 50, 60, 70, 80, 90...@}
1686 Alternatively, you can use the @code{with} command to change a setting
1687 temporarily, for the duration of a command invocation.
1690 @kindex with command
1691 @kindex w @r{(@code{with})}
1693 @cindex temporarily change settings
1694 @item with @var{setting} [@var{value}] [-- @var{command}]
1695 @itemx w @var{setting} [@var{value}] [-- @var{command}]
1696 Temporarily set @var{setting} to @var{value} for the duration of
1699 @var{setting} is any setting you can change with the @code{set}
1700 subcommands. @var{value} is the value to assign to @code{setting}
1701 while running @code{command}.
1703 If no @var{command} is provided, the last command executed is
1706 If a @var{command} is provided, it must be preceded by a double dash
1707 (@code{--}) separator. This is required because some settings accept
1708 free-form arguments, such as expressions or filenames.
1710 For example, the command
1712 (@value{GDBP}) with print array on -- print some_array
1715 is equivalent to the following 3 commands:
1717 (@value{GDBP}) set print array on
1718 (@value{GDBP}) print some_array
1719 (@value{GDBP}) set print array off
1722 The @code{with} command is particularly useful when you want to
1723 override a setting while running user-defined commands, or commands
1724 defined in Python or Guile. @xref{Extending GDB,, Extending GDB}.
1727 (@value{GDBP}) with print pretty on -- my_complex_command
1730 To change several settings for the same command, you can nest
1731 @code{with} commands. For example, @code{with language ada -- with
1732 print elements 10} temporarily changes the language to Ada and sets a
1733 limit of 10 elements to print for arrays and strings.
1738 @section Command Completion
1741 @cindex word completion
1742 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1743 only one possibility; it can also show you what the valid possibilities
1744 are for the next word in a command, at any time. This works for @value{GDBN}
1745 commands, @value{GDBN} subcommands, command options, and the names of symbols
1748 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1749 of a word. If there is only one possibility, @value{GDBN} fills in the
1750 word, and waits for you to finish the command (or press @key{RET} to
1751 enter it). For example, if you type
1753 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1754 @c complete accuracy in these examples; space introduced for clarity.
1755 @c If texinfo enhancements make it unnecessary, it would be nice to
1756 @c replace " @key" by "@key" in the following...
1758 (@value{GDBP}) info bre @key{TAB}
1762 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1763 the only @code{info} subcommand beginning with @samp{bre}:
1766 (@value{GDBP}) info breakpoints
1770 You can either press @key{RET} at this point, to run the @code{info
1771 breakpoints} command, or backspace and enter something else, if
1772 @samp{breakpoints} does not look like the command you expected. (If you
1773 were sure you wanted @code{info breakpoints} in the first place, you
1774 might as well just type @key{RET} immediately after @samp{info bre},
1775 to exploit command abbreviations rather than command completion).
1777 If there is more than one possibility for the next word when you press
1778 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1779 characters and try again, or just press @key{TAB} a second time;
1780 @value{GDBN} displays all the possible completions for that word. For
1781 example, you might want to set a breakpoint on a subroutine whose name
1782 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1783 just sounds the bell. Typing @key{TAB} again displays all the
1784 function names in your program that begin with those characters, for
1788 (@value{GDBP}) b make_ @key{TAB}
1789 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1790 make_a_section_from_file make_environ
1791 make_abs_section make_function_type
1792 make_blockvector make_pointer_type
1793 make_cleanup make_reference_type
1794 make_command make_symbol_completion_list
1795 (@value{GDBP}) b make_
1799 After displaying the available possibilities, @value{GDBN} copies your
1800 partial input (@samp{b make_} in the example) so you can finish the
1803 If you just want to see the list of alternatives in the first place, you
1804 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1805 means @kbd{@key{META} ?}. You can type this either by holding down a
1806 key designated as the @key{META} shift on your keyboard (if there is
1807 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1809 If the number of possible completions is large, @value{GDBN} will
1810 print as much of the list as it has collected, as well as a message
1811 indicating that the list may be truncated.
1814 (@value{GDBP}) b m@key{TAB}@key{TAB}
1816 <... the rest of the possible completions ...>
1817 *** List may be truncated, max-completions reached. ***
1822 This behavior can be controlled with the following commands:
1825 @kindex set max-completions
1826 @item set max-completions @var{limit}
1827 @itemx set max-completions unlimited
1828 Set the maximum number of completion candidates. @value{GDBN} will
1829 stop looking for more completions once it collects this many candidates.
1830 This is useful when completing on things like function names as collecting
1831 all the possible candidates can be time consuming.
1832 The default value is 200. A value of zero disables tab-completion.
1833 Note that setting either no limit or a very large limit can make
1835 @kindex show max-completions
1836 @item show max-completions
1837 Show the maximum number of candidates that @value{GDBN} will collect and show
1841 @cindex quotes in commands
1842 @cindex completion of quoted strings
1843 Sometimes the string you need, while logically a ``word'', may contain
1844 parentheses or other characters that @value{GDBN} normally excludes from
1845 its notion of a word. To permit word completion to work in this
1846 situation, you may enclose words in @code{'} (single quote marks) in
1847 @value{GDBN} commands.
1849 A likely situation where you might need this is in typing an
1850 expression that involves a C@t{++} symbol name with template
1851 parameters. This is because when completing expressions, GDB treats
1852 the @samp{<} character as word delimiter, assuming that it's the
1853 less-than comparison operator (@pxref{C Operators, , C and C@t{++}
1856 For example, when you want to call a C@t{++} template function
1857 interactively using the @code{print} or @code{call} commands, you may
1858 need to distinguish whether you mean the version of @code{name} that
1859 was specialized for @code{int}, @code{name<int>()}, or the version
1860 that was specialized for @code{float}, @code{name<float>()}. To use
1861 the word-completion facilities in this situation, type a single quote
1862 @code{'} at the beginning of the function name. This alerts
1863 @value{GDBN} that it may need to consider more information than usual
1864 when you press @key{TAB} or @kbd{M-?} to request word completion:
1867 (@value{GDBP}) p 'func< @kbd{M-?}
1868 func<int>() func<float>()
1869 (@value{GDBP}) p 'func<
1872 When setting breakpoints however (@pxref{Specify Location}), you don't
1873 usually need to type a quote before the function name, because
1874 @value{GDBN} understands that you want to set a breakpoint on a
1878 (@value{GDBP}) b func< @kbd{M-?}
1879 func<int>() func<float>()
1880 (@value{GDBP}) b func<
1883 This is true even in the case of typing the name of C@t{++} overloaded
1884 functions (multiple definitions of the same function, distinguished by
1885 argument type). For example, when you want to set a breakpoint you
1886 don't need to distinguish whether you mean the version of @code{name}
1887 that takes an @code{int} parameter, @code{name(int)}, or the version
1888 that takes a @code{float} parameter, @code{name(float)}.
1891 (@value{GDBP}) b bubble( @kbd{M-?}
1892 bubble(int) bubble(double)
1893 (@value{GDBP}) b bubble(dou @kbd{M-?}
1897 See @ref{quoting names} for a description of other scenarios that
1900 For more information about overloaded functions, see @ref{C Plus Plus
1901 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1902 overload-resolution off} to disable overload resolution;
1903 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1905 @cindex completion of structure field names
1906 @cindex structure field name completion
1907 @cindex completion of union field names
1908 @cindex union field name completion
1909 When completing in an expression which looks up a field in a
1910 structure, @value{GDBN} also tries@footnote{The completer can be
1911 confused by certain kinds of invalid expressions. Also, it only
1912 examines the static type of the expression, not the dynamic type.} to
1913 limit completions to the field names available in the type of the
1917 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1918 magic to_fputs to_rewind
1919 to_data to_isatty to_write
1920 to_delete to_put to_write_async_safe
1925 This is because the @code{gdb_stdout} is a variable of the type
1926 @code{struct ui_file} that is defined in @value{GDBN} sources as
1933 ui_file_flush_ftype *to_flush;
1934 ui_file_write_ftype *to_write;
1935 ui_file_write_async_safe_ftype *to_write_async_safe;
1936 ui_file_fputs_ftype *to_fputs;
1937 ui_file_read_ftype *to_read;
1938 ui_file_delete_ftype *to_delete;
1939 ui_file_isatty_ftype *to_isatty;
1940 ui_file_rewind_ftype *to_rewind;
1941 ui_file_put_ftype *to_put;
1946 @node Command Options
1947 @section Command options
1949 @cindex command options
1950 Some commands accept options starting with a leading dash. For
1951 example, @code{print -pretty}. Similarly to command names, you can
1952 abbreviate a @value{GDBN} option to the first few letters of the
1953 option name, if that abbreviation is unambiguous, and you can also use
1954 the @key{TAB} key to get @value{GDBN} to fill out the rest of a word
1955 in an option (or to show you the alternatives available, if there is
1956 more than one possibility).
1958 @cindex command options, raw input
1959 Some commands take raw input as argument. For example, the print
1960 command processes arbitrary expressions in any of the languages
1961 supported by @value{GDBN}. With such commands, because raw input may
1962 start with a leading dash that would be confused with an option or any
1963 of its abbreviations, e.g.@: @code{print -p} (short for @code{print
1964 -pretty} or printing negative @code{p}?), if you specify any command
1965 option, then you must use a double-dash (@code{--}) delimiter to
1966 indicate the end of options.
1968 @cindex command options, boolean
1970 Some options are described as accepting an argument which can be
1971 either @code{on} or @code{off}. These are known as @dfn{boolean
1972 options}. Similarly to boolean settings commands---@code{on} and
1973 @code{off} are the typical values, but any of @code{1}, @code{yes} and
1974 @code{enable} can also be used as ``true'' value, and any of @code{0},
1975 @code{no} and @code{disable} can also be used as ``false'' value. You
1976 can also omit a ``true'' value, as it is implied by default.
1978 For example, these are equivalent:
1981 (@value{GDBP}) print -object on -pretty off -element unlimited -- *myptr
1982 (@value{GDBP}) p -o -p 0 -e u -- *myptr
1985 You can discover the set of options some command accepts by completing
1986 on @code{-} after the command name. For example:
1989 (@value{GDBP}) print -@key{TAB}@key{TAB}
1990 -address -max-depth -raw-values -union
1991 -array -null-stop -repeats -vtbl
1992 -array-indexes -object -static-members
1993 -elements -pretty -symbol
1996 Completion will in some cases guide you with a suggestion of what kind
1997 of argument an option expects. For example:
2000 (@value{GDBP}) print -elements @key{TAB}@key{TAB}
2004 Here, the option expects a number (e.g., @code{100}), not literal
2005 @code{NUMBER}. Such metasyntactical arguments are always presented in
2008 (For more on using the @code{print} command, see @ref{Data, ,Examining
2012 @section Getting Help
2013 @cindex online documentation
2016 You can always ask @value{GDBN} itself for information on its commands,
2017 using the command @code{help}.
2020 @kindex h @r{(@code{help})}
2023 You can use @code{help} (abbreviated @code{h}) with no arguments to
2024 display a short list of named classes of commands:
2028 List of classes of commands:
2030 aliases -- Aliases of other commands
2031 breakpoints -- Making program stop at certain points
2032 data -- Examining data
2033 files -- Specifying and examining files
2034 internals -- Maintenance commands
2035 obscure -- Obscure features
2036 running -- Running the program
2037 stack -- Examining the stack
2038 status -- Status inquiries
2039 support -- Support facilities
2040 tracepoints -- Tracing of program execution without
2041 stopping the program
2042 user-defined -- User-defined commands
2044 Type "help" followed by a class name for a list of
2045 commands in that class.
2046 Type "help" followed by command name for full
2048 Command name abbreviations are allowed if unambiguous.
2051 @c the above line break eliminates huge line overfull...
2053 @item help @var{class}
2054 Using one of the general help classes as an argument, you can get a
2055 list of the individual commands in that class. For example, here is the
2056 help display for the class @code{status}:
2059 (@value{GDBP}) help status
2064 @c Line break in "show" line falsifies real output, but needed
2065 @c to fit in smallbook page size.
2066 info -- Generic command for showing things
2067 about the program being debugged
2068 show -- Generic command for showing things
2071 Type "help" followed by command name for full
2073 Command name abbreviations are allowed if unambiguous.
2077 @item help @var{command}
2078 With a command name as @code{help} argument, @value{GDBN} displays a
2079 short paragraph on how to use that command.
2082 @item apropos [-v] @var{regexp}
2083 The @code{apropos} command searches through all of the @value{GDBN}
2084 commands, and their documentation, for the regular expression specified in
2085 @var{args}. It prints out all matches found. The optional flag @samp{-v},
2086 which stands for @samp{verbose}, indicates to output the full documentation
2087 of the matching commands and highlight the parts of the documentation
2088 matching @var{regexp}. For example:
2099 alias -- Define a new command that is an alias of an existing command
2100 aliases -- Aliases of other commands
2101 d -- Delete some breakpoints or auto-display expressions
2102 del -- Delete some breakpoints or auto-display expressions
2103 delete -- Delete some breakpoints or auto-display expressions
2111 apropos -v cut.*thread apply
2115 results in the below output, where @samp{cut for 'thread apply}
2116 is highlighted if styling is enabled.
2120 taas -- Apply a command to all threads (ignoring errors
2123 shortcut for 'thread apply all -s COMMAND'
2125 tfaas -- Apply a command to all frames of all threads
2126 (ignoring errors and empty output).
2127 Usage: tfaas COMMAND
2128 shortcut for 'thread apply all -s frame apply all -s COMMAND'
2133 @item complete @var{args}
2134 The @code{complete @var{args}} command lists all the possible completions
2135 for the beginning of a command. Use @var{args} to specify the beginning of the
2136 command you want completed. For example:
2142 @noindent results in:
2153 @noindent This is intended for use by @sc{gnu} Emacs.
2156 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
2157 and @code{show} to inquire about the state of your program, or the state
2158 of @value{GDBN} itself. Each command supports many topics of inquiry; this
2159 manual introduces each of them in the appropriate context. The listings
2160 under @code{info} and under @code{show} in the Command, Variable, and
2161 Function Index point to all the sub-commands. @xref{Command and Variable
2167 @kindex i @r{(@code{info})}
2169 This command (abbreviated @code{i}) is for describing the state of your
2170 program. For example, you can show the arguments passed to a function
2171 with @code{info args}, list the registers currently in use with @code{info
2172 registers}, or list the breakpoints you have set with @code{info breakpoints}.
2173 You can get a complete list of the @code{info} sub-commands with
2174 @w{@code{help info}}.
2178 You can assign the result of an expression to an environment variable with
2179 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
2180 @code{set prompt $}.
2184 In contrast to @code{info}, @code{show} is for describing the state of
2185 @value{GDBN} itself.
2186 You can change most of the things you can @code{show}, by using the
2187 related command @code{set}; for example, you can control what number
2188 system is used for displays with @code{set radix}, or simply inquire
2189 which is currently in use with @code{show radix}.
2192 To display all the settable parameters and their current
2193 values, you can use @code{show} with no arguments; you may also use
2194 @code{info set}. Both commands produce the same display.
2195 @c FIXME: "info set" violates the rule that "info" is for state of
2196 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
2197 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
2201 Here are several miscellaneous @code{show} subcommands, all of which are
2202 exceptional in lacking corresponding @code{set} commands:
2205 @kindex show version
2206 @cindex @value{GDBN} version number
2208 Show what version of @value{GDBN} is running. You should include this
2209 information in @value{GDBN} bug-reports. If multiple versions of
2210 @value{GDBN} are in use at your site, you may need to determine which
2211 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
2212 commands are introduced, and old ones may wither away. Also, many
2213 system vendors ship variant versions of @value{GDBN}, and there are
2214 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
2215 The version number is the same as the one announced when you start
2218 @kindex show copying
2219 @kindex info copying
2220 @cindex display @value{GDBN} copyright
2223 Display information about permission for copying @value{GDBN}.
2225 @kindex show warranty
2226 @kindex info warranty
2228 @itemx info warranty
2229 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
2230 if your version of @value{GDBN} comes with one.
2232 @kindex show configuration
2233 @item show configuration
2234 Display detailed information about the way @value{GDBN} was configured
2235 when it was built. This displays the optional arguments passed to the
2236 @file{configure} script and also configuration parameters detected
2237 automatically by @command{configure}. When reporting a @value{GDBN}
2238 bug (@pxref{GDB Bugs}), it is important to include this information in
2244 @chapter Running Programs Under @value{GDBN}
2246 When you run a program under @value{GDBN}, you must first generate
2247 debugging information when you compile it.
2249 You may start @value{GDBN} with its arguments, if any, in an environment
2250 of your choice. If you are doing native debugging, you may redirect
2251 your program's input and output, debug an already running process, or
2252 kill a child process.
2255 * Compilation:: Compiling for debugging
2256 * Starting:: Starting your program
2257 * Arguments:: Your program's arguments
2258 * Environment:: Your program's environment
2260 * Working Directory:: Your program's working directory
2261 * Input/Output:: Your program's input and output
2262 * Attach:: Debugging an already-running process
2263 * Kill Process:: Killing the child process
2265 * Inferiors and Programs:: Debugging multiple inferiors and programs
2266 * Threads:: Debugging programs with multiple threads
2267 * Forks:: Debugging forks
2268 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
2272 @section Compiling for Debugging
2274 In order to debug a program effectively, you need to generate
2275 debugging information when you compile it. This debugging information
2276 is stored in the object file; it describes the data type of each
2277 variable or function and the correspondence between source line numbers
2278 and addresses in the executable code.
2280 To request debugging information, specify the @option{-g} option when
2281 you run the compiler. However, to use the most expressive format
2282 available, including @value{GDBN} extensions if at all possible,
2283 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports
2284 @w{@option{-ggdb}} which produces debugging information for use by
2285 @value{GDBN}. We recommend that you @emph{always} use
2286 @w{@option{-ggdb}} instead of plain @option{-g} if it is supported by
2287 the compiler you are using.
2289 Programs that are to be shipped to your customers are compiled with
2290 optimizations, using the @option{-O} compiler option. However, some
2291 compilers are unable to handle the @option{-g} and @option{-O} options
2292 together. Using those compilers, you cannot generate optimized
2293 executables containing debugging information.
2295 @value{NGCC} supports @option{-g} with or without @option{-O}, making
2296 it possible to debug optimized code. We recommend that you
2297 @emph{always} use @option{-g} whenever you compile a program. You may
2298 think your program is correct, but there is no sense in pushing your
2299 luck. For more information, see @ref{Optimized Code}.
2301 Older versions of the @sc{gnu} C compiler permitted a variant option
2302 @w{@option{-gg}} for debugging information. @value{GDBN} no longer
2303 supports this format; if your @sc{gnu} C compiler has this option, do
2306 @value{GDBN} knows about preprocessor macros and can show you their
2307 expansion (@pxref{Macros}). Most compilers do not include information
2308 about preprocessor macros in the debugging information if you specify
2309 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
2310 the @sc{gnu} C compiler, provides macro information if you are using
2311 the DWARF debugging format, and specify the option @w{@option{-g3}}.
2313 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
2314 gcc, Using the @sc{gnu} Compiler Collection (GCC)}, for more
2315 information on @value{NGCC} options affecting debug information.
2317 You will have the best debugging experience if you use the latest
2318 version of the DWARF debugging format that your compiler supports.
2319 DWARF is currently the most expressive and best supported debugging
2320 format in @value{GDBN}.
2324 @section Starting your Program
2330 @kindex r @r{(@code{run})}
2333 Use the @code{run} command to start your program under @value{GDBN}.
2334 You must first specify the program name with an argument to
2335 @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
2336 @value{GDBN}}), or by using the @code{file} or @code{exec-file}
2337 command (@pxref{Files, ,Commands to Specify Files}).
2341 If you are running your program in an execution environment that
2342 supports processes, @code{run} creates an inferior process and makes
2343 that process run your program. In some environments without processes,
2344 @code{run} jumps to the start of your program. Other targets,
2345 like @samp{remote}, are always running. If you get an error
2346 message like this one:
2349 The "remote" target does not support "run".
2350 Try "help target" or "continue".
2354 then use @code{continue} to run your program. You may need @code{load}
2355 first (@pxref{load}).
2357 The execution of a program is affected by certain information it
2358 receives from its superior. @value{GDBN} provides ways to specify this
2359 information, which you must do @emph{before} starting your program. (You
2360 can change it after starting your program, but such changes only affect
2361 your program the next time you start it.) This information may be
2362 divided into four categories:
2365 @item The @emph{arguments.}
2366 Specify the arguments to give your program as the arguments of the
2367 @code{run} command. If a shell is available on your target, the shell
2368 is used to pass the arguments, so that you may use normal conventions
2369 (such as wildcard expansion or variable substitution) in describing
2371 In Unix systems, you can control which shell is used with the
2372 @code{SHELL} environment variable. If you do not define @code{SHELL},
2373 @value{GDBN} uses the default shell (@file{/bin/sh}). You can disable
2374 use of any shell with the @code{set startup-with-shell} command (see
2377 @item The @emph{environment.}
2378 Your program normally inherits its environment from @value{GDBN}, but you can
2379 use the @value{GDBN} commands @code{set environment} and @code{unset
2380 environment} to change parts of the environment that affect
2381 your program. @xref{Environment, ,Your Program's Environment}.
2383 @item The @emph{working directory.}
2384 You can set your program's working directory with the command
2385 @kbd{set cwd}. If you do not set any working directory with this
2386 command, your program will inherit @value{GDBN}'s working directory if
2387 native debugging, or the remote server's working directory if remote
2388 debugging. @xref{Working Directory, ,Your Program's Working
2391 @item The @emph{standard input and output.}
2392 Your program normally uses the same device for standard input and
2393 standard output as @value{GDBN} is using. You can redirect input and output
2394 in the @code{run} command line, or you can use the @code{tty} command to
2395 set a different device for your program.
2396 @xref{Input/Output, ,Your Program's Input and Output}.
2399 @emph{Warning:} While input and output redirection work, you cannot use
2400 pipes to pass the output of the program you are debugging to another
2401 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2405 When you issue the @code{run} command, your program begins to execute
2406 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2407 of how to arrange for your program to stop. Once your program has
2408 stopped, you may call functions in your program, using the @code{print}
2409 or @code{call} commands. @xref{Data, ,Examining Data}.
2411 If the modification time of your symbol file has changed since the last
2412 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2413 table, and reads it again. When it does this, @value{GDBN} tries to retain
2414 your current breakpoints.
2419 @cindex run to main procedure
2420 The name of the main procedure can vary from language to language.
2421 With C or C@t{++}, the main procedure name is always @code{main}, but
2422 other languages such as Ada do not require a specific name for their
2423 main procedure. The debugger provides a convenient way to start the
2424 execution of the program and to stop at the beginning of the main
2425 procedure, depending on the language used.
2427 The @samp{start} command does the equivalent of setting a temporary
2428 breakpoint at the beginning of the main procedure and then invoking
2429 the @samp{run} command.
2431 @cindex elaboration phase
2432 Some programs contain an @dfn{elaboration} phase where some startup code is
2433 executed before the main procedure is called. This depends on the
2434 languages used to write your program. In C@t{++}, for instance,
2435 constructors for static and global objects are executed before
2436 @code{main} is called. It is therefore possible that the debugger stops
2437 before reaching the main procedure. However, the temporary breakpoint
2438 will remain to halt execution.
2440 Specify the arguments to give to your program as arguments to the
2441 @samp{start} command. These arguments will be given verbatim to the
2442 underlying @samp{run} command. Note that the same arguments will be
2443 reused if no argument is provided during subsequent calls to
2444 @samp{start} or @samp{run}.
2446 It is sometimes necessary to debug the program during elaboration. In
2447 these cases, using the @code{start} command would stop the execution
2448 of your program too late, as the program would have already completed
2449 the elaboration phase. Under these circumstances, either insert
2450 breakpoints in your elaboration code before running your program or
2451 use the @code{starti} command.
2455 @cindex run to first instruction
2456 The @samp{starti} command does the equivalent of setting a temporary
2457 breakpoint at the first instruction of a program's execution and then
2458 invoking the @samp{run} command. For programs containing an
2459 elaboration phase, the @code{starti} command will stop execution at
2460 the start of the elaboration phase.
2462 @anchor{set exec-wrapper}
2463 @kindex set exec-wrapper
2464 @item set exec-wrapper @var{wrapper}
2465 @itemx show exec-wrapper
2466 @itemx unset exec-wrapper
2467 When @samp{exec-wrapper} is set, the specified wrapper is used to
2468 launch programs for debugging. @value{GDBN} starts your program
2469 with a shell command of the form @kbd{exec @var{wrapper}
2470 @var{program}}. Quoting is added to @var{program} and its
2471 arguments, but not to @var{wrapper}, so you should add quotes if
2472 appropriate for your shell. The wrapper runs until it executes
2473 your program, and then @value{GDBN} takes control.
2475 You can use any program that eventually calls @code{execve} with
2476 its arguments as a wrapper. Several standard Unix utilities do
2477 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2478 with @code{exec "$@@"} will also work.
2480 For example, you can use @code{env} to pass an environment variable to
2481 the debugged program, without setting the variable in your shell's
2485 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2489 This command is available when debugging locally on most targets, excluding
2490 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2492 @kindex set startup-with-shell
2493 @anchor{set startup-with-shell}
2494 @item set startup-with-shell
2495 @itemx set startup-with-shell on
2496 @itemx set startup-with-shell off
2497 @itemx show startup-with-shell
2498 On Unix systems, by default, if a shell is available on your target,
2499 @value{GDBN}) uses it to start your program. Arguments of the
2500 @code{run} command are passed to the shell, which does variable
2501 substitution, expands wildcard characters and performs redirection of
2502 I/O. In some circumstances, it may be useful to disable such use of a
2503 shell, for example, when debugging the shell itself or diagnosing
2504 startup failures such as:
2508 Starting program: ./a.out
2509 During startup program terminated with signal SIGSEGV, Segmentation fault.
2513 which indicates the shell or the wrapper specified with
2514 @samp{exec-wrapper} crashed, not your program. Most often, this is
2515 caused by something odd in your shell's non-interactive mode
2516 initialization file---such as @file{.cshrc} for C-shell,
2517 $@file{.zshenv} for the Z shell, or the file specified in the
2518 @samp{BASH_ENV} environment variable for BASH.
2520 @anchor{set auto-connect-native-target}
2521 @kindex set auto-connect-native-target
2522 @item set auto-connect-native-target
2523 @itemx set auto-connect-native-target on
2524 @itemx set auto-connect-native-target off
2525 @itemx show auto-connect-native-target
2527 By default, if not connected to any target yet (e.g., with
2528 @code{target remote}), the @code{run} command starts your program as a
2529 native process under @value{GDBN}, on your local machine. If you're
2530 sure you don't want to debug programs on your local machine, you can
2531 tell @value{GDBN} to not connect to the native target automatically
2532 with the @code{set auto-connect-native-target off} command.
2534 If @code{on}, which is the default, and if @value{GDBN} is not
2535 connected to a target already, the @code{run} command automaticaly
2536 connects to the native target, if one is available.
2538 If @code{off}, and if @value{GDBN} is not connected to a target
2539 already, the @code{run} command fails with an error:
2543 Don't know how to run. Try "help target".
2546 If @value{GDBN} is already connected to a target, @value{GDBN} always
2547 uses it with the @code{run} command.
2549 In any case, you can explicitly connect to the native target with the
2550 @code{target native} command. For example,
2553 (@value{GDBP}) set auto-connect-native-target off
2555 Don't know how to run. Try "help target".
2556 (@value{GDBP}) target native
2558 Starting program: ./a.out
2559 [Inferior 1 (process 10421) exited normally]
2562 In case you connected explicitly to the @code{native} target,
2563 @value{GDBN} remains connected even if all inferiors exit, ready for
2564 the next @code{run} command. Use the @code{disconnect} command to
2567 Examples of other commands that likewise respect the
2568 @code{auto-connect-native-target} setting: @code{attach}, @code{info
2569 proc}, @code{info os}.
2571 @kindex set disable-randomization
2572 @item set disable-randomization
2573 @itemx set disable-randomization on
2574 This option (enabled by default in @value{GDBN}) will turn off the native
2575 randomization of the virtual address space of the started program. This option
2576 is useful for multiple debugging sessions to make the execution better
2577 reproducible and memory addresses reusable across debugging sessions.
2579 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2580 On @sc{gnu}/Linux you can get the same behavior using
2583 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2586 @item set disable-randomization off
2587 Leave the behavior of the started executable unchanged. Some bugs rear their
2588 ugly heads only when the program is loaded at certain addresses. If your bug
2589 disappears when you run the program under @value{GDBN}, that might be because
2590 @value{GDBN} by default disables the address randomization on platforms, such
2591 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2592 disable-randomization off} to try to reproduce such elusive bugs.
2594 On targets where it is available, virtual address space randomization
2595 protects the programs against certain kinds of security attacks. In these
2596 cases the attacker needs to know the exact location of a concrete executable
2597 code. Randomizing its location makes it impossible to inject jumps misusing
2598 a code at its expected addresses.
2600 Prelinking shared libraries provides a startup performance advantage but it
2601 makes addresses in these libraries predictable for privileged processes by
2602 having just unprivileged access at the target system. Reading the shared
2603 library binary gives enough information for assembling the malicious code
2604 misusing it. Still even a prelinked shared library can get loaded at a new
2605 random address just requiring the regular relocation process during the
2606 startup. Shared libraries not already prelinked are always loaded at
2607 a randomly chosen address.
2609 Position independent executables (PIE) contain position independent code
2610 similar to the shared libraries and therefore such executables get loaded at
2611 a randomly chosen address upon startup. PIE executables always load even
2612 already prelinked shared libraries at a random address. You can build such
2613 executable using @command{gcc -fPIE -pie}.
2615 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2616 (as long as the randomization is enabled).
2618 @item show disable-randomization
2619 Show the current setting of the explicit disable of the native randomization of
2620 the virtual address space of the started program.
2625 @section Your Program's Arguments
2627 @cindex arguments (to your program)
2628 The arguments to your program can be specified by the arguments of the
2630 They are passed to a shell, which expands wildcard characters and
2631 performs redirection of I/O, and thence to your program. Your
2632 @code{SHELL} environment variable (if it exists) specifies what shell
2633 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2634 the default shell (@file{/bin/sh} on Unix).
2636 On non-Unix systems, the program is usually invoked directly by
2637 @value{GDBN}, which emulates I/O redirection via the appropriate system
2638 calls, and the wildcard characters are expanded by the startup code of
2639 the program, not by the shell.
2641 @code{run} with no arguments uses the same arguments used by the previous
2642 @code{run}, or those set by the @code{set args} command.
2647 Specify the arguments to be used the next time your program is run. If
2648 @code{set args} has no arguments, @code{run} executes your program
2649 with no arguments. Once you have run your program with arguments,
2650 using @code{set args} before the next @code{run} is the only way to run
2651 it again without arguments.
2655 Show the arguments to give your program when it is started.
2659 @section Your Program's Environment
2661 @cindex environment (of your program)
2662 The @dfn{environment} consists of a set of environment variables and
2663 their values. Environment variables conventionally record such things as
2664 your user name, your home directory, your terminal type, and your search
2665 path for programs to run. Usually you set up environment variables with
2666 the shell and they are inherited by all the other programs you run. When
2667 debugging, it can be useful to try running your program with a modified
2668 environment without having to start @value{GDBN} over again.
2672 @item path @var{directory}
2673 Add @var{directory} to the front of the @code{PATH} environment variable
2674 (the search path for executables) that will be passed to your program.
2675 The value of @code{PATH} used by @value{GDBN} does not change.
2676 You may specify several directory names, separated by whitespace or by a
2677 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2678 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2679 is moved to the front, so it is searched sooner.
2681 You can use the string @samp{$cwd} to refer to whatever is the current
2682 working directory at the time @value{GDBN} searches the path. If you
2683 use @samp{.} instead, it refers to the directory where you executed the
2684 @code{path} command. @value{GDBN} replaces @samp{.} in the
2685 @var{directory} argument (with the current path) before adding
2686 @var{directory} to the search path.
2687 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2688 @c document that, since repeating it would be a no-op.
2692 Display the list of search paths for executables (the @code{PATH}
2693 environment variable).
2695 @kindex show environment
2696 @item show environment @r{[}@var{varname}@r{]}
2697 Print the value of environment variable @var{varname} to be given to
2698 your program when it starts. If you do not supply @var{varname},
2699 print the names and values of all environment variables to be given to
2700 your program. You can abbreviate @code{environment} as @code{env}.
2702 @kindex set environment
2703 @anchor{set environment}
2704 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2705 Set environment variable @var{varname} to @var{value}. The value
2706 changes for your program (and the shell @value{GDBN} uses to launch
2707 it), not for @value{GDBN} itself. The @var{value} may be any string; the
2708 values of environment variables are just strings, and any
2709 interpretation is supplied by your program itself. The @var{value}
2710 parameter is optional; if it is eliminated, the variable is set to a
2712 @c "any string" here does not include leading, trailing
2713 @c blanks. Gnu asks: does anyone care?
2715 For example, this command:
2722 tells the debugged program, when subsequently run, that its user is named
2723 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2724 are not actually required.)
2726 Note that on Unix systems, @value{GDBN} runs your program via a shell,
2727 which also inherits the environment set with @code{set environment}.
2728 If necessary, you can avoid that by using the @samp{env} program as a
2729 wrapper instead of using @code{set environment}. @xref{set
2730 exec-wrapper}, for an example doing just that.
2732 Environment variables that are set by the user are also transmitted to
2733 @command{gdbserver} to be used when starting the remote inferior.
2734 @pxref{QEnvironmentHexEncoded}.
2736 @kindex unset environment
2737 @anchor{unset environment}
2738 @item unset environment @var{varname}
2739 Remove variable @var{varname} from the environment to be passed to your
2740 program. This is different from @samp{set env @var{varname} =};
2741 @code{unset environment} removes the variable from the environment,
2742 rather than assigning it an empty value.
2744 Environment variables that are unset by the user are also unset on
2745 @command{gdbserver} when starting the remote inferior.
2746 @pxref{QEnvironmentUnset}.
2749 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2750 the shell indicated by your @code{SHELL} environment variable if it
2751 exists (or @code{/bin/sh} if not). If your @code{SHELL} variable
2752 names a shell that runs an initialization file when started
2753 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2754 for the Z shell, or the file specified in the @samp{BASH_ENV}
2755 environment variable for BASH---any variables you set in that file
2756 affect your program. You may wish to move setting of environment
2757 variables to files that are only run when you sign on, such as
2758 @file{.login} or @file{.profile}.
2760 @node Working Directory
2761 @section Your Program's Working Directory
2763 @cindex working directory (of your program)
2764 Each time you start your program with @code{run}, the inferior will be
2765 initialized with the current working directory specified by the
2766 @kbd{set cwd} command. If no directory has been specified by this
2767 command, then the inferior will inherit @value{GDBN}'s current working
2768 directory as its working directory if native debugging, or it will
2769 inherit the remote server's current working directory if remote
2774 @cindex change inferior's working directory
2775 @anchor{set cwd command}
2776 @item set cwd @r{[}@var{directory}@r{]}
2777 Set the inferior's working directory to @var{directory}, which will be
2778 @code{glob}-expanded in order to resolve tildes (@file{~}). If no
2779 argument has been specified, the command clears the setting and resets
2780 it to an empty state. This setting has no effect on @value{GDBN}'s
2781 working directory, and it only takes effect the next time you start
2782 the inferior. The @file{~} in @var{directory} is a short for the
2783 @dfn{home directory}, usually pointed to by the @env{HOME} environment
2784 variable. On MS-Windows, if @env{HOME} is not defined, @value{GDBN}
2785 uses the concatenation of @env{HOMEDRIVE} and @env{HOMEPATH} as
2788 You can also change @value{GDBN}'s current working directory by using
2789 the @code{cd} command.
2793 @cindex show inferior's working directory
2795 Show the inferior's working directory. If no directory has been
2796 specified by @kbd{set cwd}, then the default inferior's working
2797 directory is the same as @value{GDBN}'s working directory.
2800 @cindex change @value{GDBN}'s working directory
2802 @item cd @r{[}@var{directory}@r{]}
2803 Set the @value{GDBN} working directory to @var{directory}. If not
2804 given, @var{directory} uses @file{'~'}.
2806 The @value{GDBN} working directory serves as a default for the
2807 commands that specify files for @value{GDBN} to operate on.
2808 @xref{Files, ,Commands to Specify Files}.
2809 @xref{set cwd command}.
2813 Print the @value{GDBN} working directory.
2816 It is generally impossible to find the current working directory of
2817 the process being debugged (since a program can change its directory
2818 during its run). If you work on a system where @value{GDBN} supports
2819 the @code{info proc} command (@pxref{Process Information}), you can
2820 use the @code{info proc} command to find out the
2821 current working directory of the debuggee.
2824 @section Your Program's Input and Output
2829 By default, the program you run under @value{GDBN} does input and output to
2830 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2831 to its own terminal modes to interact with you, but it records the terminal
2832 modes your program was using and switches back to them when you continue
2833 running your program.
2836 @kindex info terminal
2838 Displays information recorded by @value{GDBN} about the terminal modes your
2842 You can redirect your program's input and/or output using shell
2843 redirection with the @code{run} command. For example,
2850 starts your program, diverting its output to the file @file{outfile}.
2853 @cindex controlling terminal
2854 Another way to specify where your program should do input and output is
2855 with the @code{tty} command. This command accepts a file name as
2856 argument, and causes this file to be the default for future @code{run}
2857 commands. It also resets the controlling terminal for the child
2858 process, for future @code{run} commands. For example,
2865 directs that processes started with subsequent @code{run} commands
2866 default to do input and output on the terminal @file{/dev/ttyb} and have
2867 that as their controlling terminal.
2869 An explicit redirection in @code{run} overrides the @code{tty} command's
2870 effect on the input/output device, but not its effect on the controlling
2873 When you use the @code{tty} command or redirect input in the @code{run}
2874 command, only the input @emph{for your program} is affected. The input
2875 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2876 for @code{set inferior-tty}.
2878 @cindex inferior tty
2879 @cindex set inferior controlling terminal
2880 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2881 display the name of the terminal that will be used for future runs of your
2885 @item set inferior-tty [ @var{tty} ]
2886 @kindex set inferior-tty
2887 Set the tty for the program being debugged to @var{tty}. Omitting @var{tty}
2888 restores the default behavior, which is to use the same terminal as
2891 @item show inferior-tty
2892 @kindex show inferior-tty
2893 Show the current tty for the program being debugged.
2897 @section Debugging an Already-running Process
2902 @item attach @var{process-id}
2903 This command attaches to a running process---one that was started
2904 outside @value{GDBN}. (@code{info files} shows your active
2905 targets.) The command takes as argument a process ID. The usual way to
2906 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2907 or with the @samp{jobs -l} shell command.
2909 @code{attach} does not repeat if you press @key{RET} a second time after
2910 executing the command.
2913 To use @code{attach}, your program must be running in an environment
2914 which supports processes; for example, @code{attach} does not work for
2915 programs on bare-board targets that lack an operating system. You must
2916 also have permission to send the process a signal.
2918 When you use @code{attach}, the debugger finds the program running in
2919 the process first by looking in the current working directory, then (if
2920 the program is not found) by using the source file search path
2921 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2922 the @code{file} command to load the program. @xref{Files, ,Commands to
2925 The first thing @value{GDBN} does after arranging to debug the specified
2926 process is to stop it. You can examine and modify an attached process
2927 with all the @value{GDBN} commands that are ordinarily available when
2928 you start processes with @code{run}. You can insert breakpoints; you
2929 can step and continue; you can modify storage. If you would rather the
2930 process continue running, you may use the @code{continue} command after
2931 attaching @value{GDBN} to the process.
2936 When you have finished debugging the attached process, you can use the
2937 @code{detach} command to release it from @value{GDBN} control. Detaching
2938 the process continues its execution. After the @code{detach} command,
2939 that process and @value{GDBN} become completely independent once more, and you
2940 are ready to @code{attach} another process or start one with @code{run}.
2941 @code{detach} does not repeat if you press @key{RET} again after
2942 executing the command.
2945 If you exit @value{GDBN} while you have an attached process, you detach
2946 that process. If you use the @code{run} command, you kill that process.
2947 By default, @value{GDBN} asks for confirmation if you try to do either of these
2948 things; you can control whether or not you need to confirm by using the
2949 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2953 @section Killing the Child Process
2958 Kill the child process in which your program is running under @value{GDBN}.
2961 This command is useful if you wish to debug a core dump instead of a
2962 running process. @value{GDBN} ignores any core dump file while your program
2965 On some operating systems, a program cannot be executed outside @value{GDBN}
2966 while you have breakpoints set on it inside @value{GDBN}. You can use the
2967 @code{kill} command in this situation to permit running your program
2968 outside the debugger.
2970 The @code{kill} command is also useful if you wish to recompile and
2971 relink your program, since on many systems it is impossible to modify an
2972 executable file while it is running in a process. In this case, when you
2973 next type @code{run}, @value{GDBN} notices that the file has changed, and
2974 reads the symbol table again (while trying to preserve your current
2975 breakpoint settings).
2977 @node Inferiors and Programs
2978 @section Debugging Multiple Inferiors and Programs
2980 @value{GDBN} lets you run and debug multiple programs in a single
2981 session. In addition, @value{GDBN} on some systems may let you run
2982 several programs simultaneously (otherwise you have to exit from one
2983 before starting another). In the most general case, you can have
2984 multiple threads of execution in each of multiple processes, launched
2985 from multiple executables.
2988 @value{GDBN} represents the state of each program execution with an
2989 object called an @dfn{inferior}. An inferior typically corresponds to
2990 a process, but is more general and applies also to targets that do not
2991 have processes. Inferiors may be created before a process runs, and
2992 may be retained after a process exits. Inferiors have unique
2993 identifiers that are different from process ids. Usually each
2994 inferior will also have its own distinct address space, although some
2995 embedded targets may have several inferiors running in different parts
2996 of a single address space. Each inferior may in turn have multiple
2997 threads running in it.
2999 To find out what inferiors exist at any moment, use @w{@code{info
3003 @kindex info inferiors [ @var{id}@dots{} ]
3004 @item info inferiors
3005 Print a list of all inferiors currently being managed by @value{GDBN}.
3006 By default all inferiors are printed, but the argument @var{id}@dots{}
3007 -- a space separated list of inferior numbers -- can be used to limit
3008 the display to just the requested inferiors.
3010 @value{GDBN} displays for each inferior (in this order):
3014 the inferior number assigned by @value{GDBN}
3017 the target system's inferior identifier
3020 the name of the executable the inferior is running.
3025 An asterisk @samp{*} preceding the @value{GDBN} inferior number
3026 indicates the current inferior.
3030 @c end table here to get a little more width for example
3033 (@value{GDBP}) info inferiors
3034 Num Description Executable
3035 2 process 2307 hello
3036 * 1 process 3401 goodbye
3039 To switch focus between inferiors, use the @code{inferior} command:
3042 @kindex inferior @var{infno}
3043 @item inferior @var{infno}
3044 Make inferior number @var{infno} the current inferior. The argument
3045 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
3046 in the first field of the @samp{info inferiors} display.
3049 @vindex $_inferior@r{, convenience variable}
3050 The debugger convenience variable @samp{$_inferior} contains the
3051 number of the current inferior. You may find this useful in writing
3052 breakpoint conditional expressions, command scripts, and so forth.
3053 @xref{Convenience Vars,, Convenience Variables}, for general
3054 information on convenience variables.
3056 You can get multiple executables into a debugging session via the
3057 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
3058 systems @value{GDBN} can add inferiors to the debug session
3059 automatically by following calls to @code{fork} and @code{exec}. To
3060 remove inferiors from the debugging session use the
3061 @w{@code{remove-inferiors}} command.
3064 @kindex add-inferior
3065 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
3066 Adds @var{n} inferiors to be run using @var{executable} as the
3067 executable; @var{n} defaults to 1. If no executable is specified,
3068 the inferiors begins empty, with no program. You can still assign or
3069 change the program assigned to the inferior at any time by using the
3070 @code{file} command with the executable name as its argument.
3072 @kindex clone-inferior
3073 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
3074 Adds @var{n} inferiors ready to execute the same program as inferior
3075 @var{infno}; @var{n} defaults to 1, and @var{infno} defaults to the
3076 number of the current inferior. This is a convenient command when you
3077 want to run another instance of the inferior you are debugging.
3080 (@value{GDBP}) info inferiors
3081 Num Description Executable
3082 * 1 process 29964 helloworld
3083 (@value{GDBP}) clone-inferior
3086 (@value{GDBP}) info inferiors
3087 Num Description Executable
3089 * 1 process 29964 helloworld
3092 You can now simply switch focus to inferior 2 and run it.
3094 @kindex remove-inferiors
3095 @item remove-inferiors @var{infno}@dots{}
3096 Removes the inferior or inferiors @var{infno}@dots{}. It is not
3097 possible to remove an inferior that is running with this command. For
3098 those, use the @code{kill} or @code{detach} command first.
3102 To quit debugging one of the running inferiors that is not the current
3103 inferior, you can either detach from it by using the @w{@code{detach
3104 inferior}} command (allowing it to run independently), or kill it
3105 using the @w{@code{kill inferiors}} command:
3108 @kindex detach inferiors @var{infno}@dots{}
3109 @item detach inferior @var{infno}@dots{}
3110 Detach from the inferior or inferiors identified by @value{GDBN}
3111 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
3112 still stays on the list of inferiors shown by @code{info inferiors},
3113 but its Description will show @samp{<null>}.
3115 @kindex kill inferiors @var{infno}@dots{}
3116 @item kill inferiors @var{infno}@dots{}
3117 Kill the inferior or inferiors identified by @value{GDBN} inferior
3118 number(s) @var{infno}@dots{}. Note that the inferior's entry still
3119 stays on the list of inferiors shown by @code{info inferiors}, but its
3120 Description will show @samp{<null>}.
3123 After the successful completion of a command such as @code{detach},
3124 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
3125 a normal process exit, the inferior is still valid and listed with
3126 @code{info inferiors}, ready to be restarted.
3129 To be notified when inferiors are started or exit under @value{GDBN}'s
3130 control use @w{@code{set print inferior-events}}:
3133 @kindex set print inferior-events
3134 @cindex print messages on inferior start and exit
3135 @item set print inferior-events
3136 @itemx set print inferior-events on
3137 @itemx set print inferior-events off
3138 The @code{set print inferior-events} command allows you to enable or
3139 disable printing of messages when @value{GDBN} notices that new
3140 inferiors have started or that inferiors have exited or have been
3141 detached. By default, these messages will not be printed.
3143 @kindex show print inferior-events
3144 @item show print inferior-events
3145 Show whether messages will be printed when @value{GDBN} detects that
3146 inferiors have started, exited or have been detached.
3149 Many commands will work the same with multiple programs as with a
3150 single program: e.g., @code{print myglobal} will simply display the
3151 value of @code{myglobal} in the current inferior.
3154 Occasionally, when debugging @value{GDBN} itself, it may be useful to
3155 get more info about the relationship of inferiors, programs, address
3156 spaces in a debug session. You can do that with the @w{@code{maint
3157 info program-spaces}} command.
3160 @kindex maint info program-spaces
3161 @item maint info program-spaces
3162 Print a list of all program spaces currently being managed by
3165 @value{GDBN} displays for each program space (in this order):
3169 the program space number assigned by @value{GDBN}
3172 the name of the executable loaded into the program space, with e.g.,
3173 the @code{file} command.
3178 An asterisk @samp{*} preceding the @value{GDBN} program space number
3179 indicates the current program space.
3181 In addition, below each program space line, @value{GDBN} prints extra
3182 information that isn't suitable to display in tabular form. For
3183 example, the list of inferiors bound to the program space.
3186 (@value{GDBP}) maint info program-spaces
3190 Bound inferiors: ID 1 (process 21561)
3193 Here we can see that no inferior is running the program @code{hello},
3194 while @code{process 21561} is running the program @code{goodbye}. On
3195 some targets, it is possible that multiple inferiors are bound to the
3196 same program space. The most common example is that of debugging both
3197 the parent and child processes of a @code{vfork} call. For example,
3200 (@value{GDBP}) maint info program-spaces
3203 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
3206 Here, both inferior 2 and inferior 1 are running in the same program
3207 space as a result of inferior 1 having executed a @code{vfork} call.
3211 @section Debugging Programs with Multiple Threads
3213 @cindex threads of execution
3214 @cindex multiple threads
3215 @cindex switching threads
3216 In some operating systems, such as GNU/Linux and Solaris, a single program
3217 may have more than one @dfn{thread} of execution. The precise semantics
3218 of threads differ from one operating system to another, but in general
3219 the threads of a single program are akin to multiple processes---except
3220 that they share one address space (that is, they can all examine and
3221 modify the same variables). On the other hand, each thread has its own
3222 registers and execution stack, and perhaps private memory.
3224 @value{GDBN} provides these facilities for debugging multi-thread
3228 @item automatic notification of new threads
3229 @item @samp{thread @var{thread-id}}, a command to switch among threads
3230 @item @samp{info threads}, a command to inquire about existing threads
3231 @item @samp{thread apply @r{[}@var{thread-id-list} @r{|} all@r{]} @var{args}},
3232 a command to apply a command to a list of threads
3233 @item thread-specific breakpoints
3234 @item @samp{set print thread-events}, which controls printing of
3235 messages on thread start and exit.
3236 @item @samp{set libthread-db-search-path @var{path}}, which lets
3237 the user specify which @code{libthread_db} to use if the default choice
3238 isn't compatible with the program.
3241 @cindex focus of debugging
3242 @cindex current thread
3243 The @value{GDBN} thread debugging facility allows you to observe all
3244 threads while your program runs---but whenever @value{GDBN} takes
3245 control, one thread in particular is always the focus of debugging.
3246 This thread is called the @dfn{current thread}. Debugging commands show
3247 program information from the perspective of the current thread.
3249 @cindex @code{New} @var{systag} message
3250 @cindex thread identifier (system)
3251 @anchor{target system thread identifier}
3252 @c FIXME-implementors!! It would be more helpful if the [New...] message
3253 @c included GDB's numeric thread handle, so you could just go to that
3254 @c thread without first checking `info threads'.
3255 Whenever @value{GDBN} detects a new thread in your program, it displays
3256 the target system's identification for the thread with a message in the
3257 form @samp{[New @var{systag}]}, where @var{systag} is a thread identifier
3258 whose form varies depending on the particular system. For example, on
3259 @sc{gnu}/Linux, you might see
3262 [New Thread 0x41e02940 (LWP 25582)]
3266 when @value{GDBN} notices a new thread. In contrast, on other systems,
3267 the @var{systag} is simply something like @samp{process 368}, with no
3270 @c FIXME!! (1) Does the [New...] message appear even for the very first
3271 @c thread of a program, or does it only appear for the
3272 @c second---i.e.@: when it becomes obvious we have a multithread
3274 @c (2) *Is* there necessarily a first thread always? Or do some
3275 @c multithread systems permit starting a program with multiple
3276 @c threads ab initio?
3278 @anchor{thread numbers}
3279 @cindex thread number, per inferior
3280 @cindex thread identifier (GDB)
3281 For debugging purposes, @value{GDBN} associates its own thread number
3282 ---always a single integer---with each thread of an inferior. This
3283 number is unique between all threads of an inferior, but not unique
3284 between threads of different inferiors.
3286 @cindex qualified thread ID
3287 You can refer to a given thread in an inferior using the qualified
3288 @var{inferior-num}.@var{thread-num} syntax, also known as
3289 @dfn{qualified thread ID}, with @var{inferior-num} being the inferior
3290 number and @var{thread-num} being the thread number of the given
3291 inferior. For example, thread @code{2.3} refers to thread number 3 of
3292 inferior 2. If you omit @var{inferior-num} (e.g., @code{thread 3}),
3293 then @value{GDBN} infers you're referring to a thread of the current
3296 Until you create a second inferior, @value{GDBN} does not show the
3297 @var{inferior-num} part of thread IDs, even though you can always use
3298 the full @var{inferior-num}.@var{thread-num} form to refer to threads
3299 of inferior 1, the initial inferior.
3301 @anchor{thread ID list}
3302 @cindex thread ID list
3303 Some commands accept a space-separated @dfn{thread ID list} as
3304 argument. A list element can be:
3308 A thread ID as shown in the first field of the @samp{info threads}
3309 display, with or without an inferior qualifier. E.g., @samp{2.1} or
3313 A range of thread numbers, again with or without an inferior
3314 qualifier, as in @var{inf}.@var{thr1}-@var{thr2} or
3315 @var{thr1}-@var{thr2}. E.g., @samp{1.2-4} or @samp{2-4}.
3318 All threads of an inferior, specified with a star wildcard, with or
3319 without an inferior qualifier, as in @var{inf}.@code{*} (e.g.,
3320 @samp{1.*}) or @code{*}. The former refers to all threads of the
3321 given inferior, and the latter form without an inferior qualifier
3322 refers to all threads of the current inferior.
3326 For example, if the current inferior is 1, and inferior 7 has one
3327 thread with ID 7.1, the thread list @samp{1 2-3 4.5 6.7-9 7.*}
3328 includes threads 1 to 3 of inferior 1, thread 5 of inferior 4, threads
3329 7 to 9 of inferior 6 and all threads of inferior 7. That is, in
3330 expanded qualified form, the same as @samp{1.1 1.2 1.3 4.5 6.7 6.8 6.9
3334 @anchor{global thread number}
3335 @cindex global thread number
3336 @cindex global thread identifier (GDB)
3337 In addition to a @emph{per-inferior} number, each thread is also
3338 assigned a unique @emph{global} number, also known as @dfn{global
3339 thread ID}, a single integer. Unlike the thread number component of
3340 the thread ID, no two threads have the same global ID, even when
3341 you're debugging multiple inferiors.
3343 From @value{GDBN}'s perspective, a process always has at least one
3344 thread. In other words, @value{GDBN} assigns a thread number to the
3345 program's ``main thread'' even if the program is not multi-threaded.
3347 @vindex $_thread@r{, convenience variable}
3348 @vindex $_gthread@r{, convenience variable}
3349 @vindex $_thread_systag@r{, convenience variable}
3350 @vindex $_thread_name@r{, convenience variable}
3351 The debugger convenience variables @samp{$_thread}, @samp{$_gthread},
3352 @samp{$_thread_systag}, and @samp{$_thread_name} contain,
3353 respectively, the per-inferior thread number, the global thread
3354 number, the target system's thread identifier (@var{systag}) string,
3355 and the thread name string of the current thread. You may find this
3356 useful in writing breakpoint conditional expressions, command scripts,
3357 and so forth. @xref{Convenience Vars,, Convenience Variables}, for
3358 general information on convenience variables.
3360 If @value{GDBN} detects the program is multi-threaded, it augments the
3361 usual message about stopping at a breakpoint with the ID and name of
3362 the thread that hit the breakpoint.
3365 Thread 2 "client" hit Breakpoint 1, send_message () at client.c:68
3368 Likewise when the program receives a signal:
3371 Thread 1 "main" received signal SIGINT, Interrupt.
3375 @kindex info threads
3376 @item info threads @r{[}-gid@r{]} @r{[}@var{thread-id-list}@r{]}
3378 Display information about one or more threads. With no arguments
3379 displays information about all threads. You can specify the list of
3380 threads that you want to display using the thread ID list syntax
3381 (@pxref{thread ID list}).
3383 @value{GDBN} displays for each thread (in this order):
3387 the per-inferior thread number assigned by @value{GDBN}
3390 the global thread number assigned by @value{GDBN}, if the
3391 @w{@option{-gid}} option was specified
3394 the target system's thread identifier (@var{systag})
3397 the thread's name, if one is known. A thread can either be named by
3398 the user (see @code{thread name}, below), or, in some cases, by the
3402 the current stack frame summary for that thread
3406 An asterisk @samp{*} to the left of the @value{GDBN} thread number
3407 indicates the current thread.
3411 @c end table here to get a little more width for example
3414 (@value{GDBP}) info threads
3416 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3417 2 process 35 thread 23 0x34e5 in sigpause ()
3418 3 process 35 thread 27 0x34e5 in sigpause ()
3421 If you're debugging multiple inferiors, @value{GDBN} displays thread
3422 IDs using the qualified @var{inferior-num}.@var{thread-num} format.
3423 Otherwise, only @var{thread-num} is shown.
3425 If you specify the @w{@option{-gid}} option, @value{GDBN} displays a
3426 column indicating each thread's global thread ID:
3429 (@value{GDBP}) info threads -gid
3430 Id GId Target Id Frame
3431 1.1 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3432 1.2 3 process 35 thread 23 0x34e5 in sigpause ()
3433 1.3 4 process 35 thread 27 0x34e5 in sigpause ()
3434 * 2.1 2 process 65 thread 1 main (argc=1, argv=0x7ffffff8)
3437 On Solaris, you can display more information about user threads with a
3438 Solaris-specific command:
3441 @item maint info sol-threads
3442 @kindex maint info sol-threads
3443 @cindex thread info (Solaris)
3444 Display info on Solaris user threads.
3448 @kindex thread @r{[}-gid@r{]} @var{thread-id}
3449 @item thread @r{[}-gid@r{]} @var{thread-id}
3450 Make thread ID @var{thread-id} the current thread. The command
3451 argument @var{thread-id} is the @value{GDBN} thread ID: if the
3452 @w{@option{-gid}} option is given it is a global thread identifier, as
3453 shown in the second field of the @samp{info threads -gid} display;
3454 otherwise it is a per process thread identifier, with or without an
3455 inferior qualifier (e.g., @samp{2.1} or @samp{1}), as shown in the
3456 first field of the @samp{info threads} display.
3458 @value{GDBN} responds by displaying the system identifier of the
3459 thread you selected, and its current stack frame summary:
3462 (@value{GDBP}) thread 2
3463 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
3464 #0 some_function (ignore=0x0) at example.c:8
3465 8 printf ("hello\n");
3469 As with the @samp{[New @dots{}]} message, the form of the text after
3470 @samp{Switching to} depends on your system's conventions for identifying
3473 @anchor{thread apply all}
3474 @kindex thread apply
3475 @cindex apply command to several threads
3476 @item thread apply @r{[}@var{thread-id-list} @r{|} all @r{[}-ascending@r{]]} @r{[}@var{flag}@r{]@dots{}} @var{command}
3477 The @code{thread apply} command allows you to apply the named
3478 @var{command} to one or more threads. Specify the threads that you
3479 want affected using the thread ID list syntax (@pxref{thread ID
3480 list}), or specify @code{all} to apply to all threads. To apply a
3481 command to all threads in descending order, type @kbd{thread apply all
3482 @var{command}}. To apply a command to all threads in ascending order,
3483 type @kbd{thread apply all -ascending @var{command}}.
3485 The @var{flag} arguments control what output to produce and how to handle
3486 errors raised when applying @var{command} to a thread. @var{flag}
3487 must start with a @code{-} directly followed by one letter in
3488 @code{qcs}. If several flags are provided, they must be given
3489 individually, such as @code{-c -q}.
3491 By default, @value{GDBN} displays some thread information before the
3492 output produced by @var{command}, and an error raised during the
3493 execution of a @var{command} will abort @code{thread apply}. The
3494 following flags can be used to fine-tune this behavior:
3498 The flag @code{-c}, which stands for @samp{continue}, causes any
3499 errors in @var{command} to be displayed, and the execution of
3500 @code{thread apply} then continues.
3502 The flag @code{-s}, which stands for @samp{silent}, causes any errors
3503 or empty output produced by a @var{command} to be silently ignored.
3504 That is, the execution continues, but the thread information and errors
3507 The flag @code{-q} (@samp{quiet}) disables printing the thread
3511 Flags @code{-c} and @code{-s} cannot be used together.
3514 @cindex apply command to all threads (ignoring errors and empty output)
3515 @item taas [@var{option}]@dots{} @var{command}
3516 Shortcut for @code{thread apply all -s @r{[}@var{option}@r{]@dots{}} @var{command}}.
3517 Applies @var{command} on all threads, ignoring errors and empty output.
3519 The @code{taas} command accepts the same options as the @code{thread
3520 apply all} command. @xref{thread apply all}.
3523 @cindex apply a command to all frames of all threads (ignoring errors and empty output)
3524 @item tfaas [@var{option}]@dots{} @var{command}
3525 Shortcut for @code{thread apply all -s -- frame apply all -s @r{[}@var{option}@r{]@dots{}} @var{command}}.
3526 Applies @var{command} on all frames of all threads, ignoring errors
3527 and empty output. Note that the flag @code{-s} is specified twice:
3528 The first @code{-s} ensures that @code{thread apply} only shows the thread
3529 information of the threads for which @code{frame apply} produces
3530 some output. The second @code{-s} is needed to ensure that @code{frame
3531 apply} shows the frame information of a frame only if the
3532 @var{command} successfully produced some output.
3534 It can for example be used to print a local variable or a function
3535 argument without knowing the thread or frame where this variable or argument
3538 (@value{GDBP}) tfaas p some_local_var_i_do_not_remember_where_it_is
3541 The @code{tfaas} command accepts the same options as the @code{frame
3542 apply} command. @xref{frame apply}.
3545 @cindex name a thread
3546 @anchor{thread name}
3547 @item thread name [@var{name}]
3548 This command assigns a name to the current thread. If no argument is
3549 given, any existing user-specified name is removed. The thread name
3550 appears in the @samp{info threads} display.
3552 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
3553 determine the name of the thread as given by the OS. On these
3554 systems, a name specified with @samp{thread name} will override the
3555 system-give name, and removing the user-specified name will cause
3556 @value{GDBN} to once again display the system-specified name.
3559 @cindex search for a thread
3560 @anchor{thread find}
3561 @item thread find [@var{regexp}]
3562 Search for and display thread ids whose name or @var{systag}
3563 matches the supplied regular expression. The syntax of the regular
3564 expression is that specified by Python's regular expression support.
3566 As well as being the complement to the @samp{thread name} command,
3567 this command also allows you to identify a thread by its target
3568 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
3572 (@value{GDBP}) thread find 26688
3573 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
3574 (@value{GDBP}) info thread 4
3576 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
3579 @kindex set print thread-events
3580 @cindex print messages on thread start and exit
3581 @item set print thread-events
3582 @itemx set print thread-events on
3583 @itemx set print thread-events off
3584 The @code{set print thread-events} command allows you to enable or
3585 disable printing of messages when @value{GDBN} notices that new threads have
3586 started or that threads have exited. By default, these messages will
3587 be printed if detection of these events is supported by the target.
3588 Note that these messages cannot be disabled on all targets.
3590 @kindex show print thread-events
3591 @item show print thread-events
3592 Show whether messages will be printed when @value{GDBN} detects that threads
3593 have started and exited.
3596 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
3597 more information about how @value{GDBN} behaves when you stop and start
3598 programs with multiple threads.
3600 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
3601 watchpoints in programs with multiple threads.
3603 @anchor{set libthread-db-search-path}
3605 @kindex set libthread-db-search-path
3606 @cindex search path for @code{libthread_db}
3607 @item set libthread-db-search-path @r{[}@var{path}@r{]}
3608 If this variable is set, @var{path} is a colon-separated list of
3609 directories @value{GDBN} will use to search for @code{libthread_db}.
3610 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
3611 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
3612 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
3615 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
3616 @code{libthread_db} library to obtain information about threads in the
3617 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
3618 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
3619 specific thread debugging library loading is enabled
3620 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
3622 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
3623 refers to the default system directories that are
3624 normally searched for loading shared libraries. The @samp{$sdir} entry
3625 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
3626 (@pxref{libthread_db.so.1 file}).
3628 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
3629 refers to the directory from which @code{libpthread}
3630 was loaded in the inferior process.
3632 For any @code{libthread_db} library @value{GDBN} finds in above directories,
3633 @value{GDBN} attempts to initialize it with the current inferior process.
3634 If this initialization fails (which could happen because of a version
3635 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
3636 will unload @code{libthread_db}, and continue with the next directory.
3637 If none of @code{libthread_db} libraries initialize successfully,
3638 @value{GDBN} will issue a warning and thread debugging will be disabled.
3640 Setting @code{libthread-db-search-path} is currently implemented
3641 only on some platforms.
3643 @kindex show libthread-db-search-path
3644 @item show libthread-db-search-path
3645 Display current libthread_db search path.
3647 @kindex set debug libthread-db
3648 @kindex show debug libthread-db
3649 @cindex debugging @code{libthread_db}
3650 @item set debug libthread-db
3651 @itemx show debug libthread-db
3652 Turns on or off display of @code{libthread_db}-related events.
3653 Use @code{1} to enable, @code{0} to disable.
3656 @xref{Heterogeneous Debugging} for additional infomation related to
3657 threads in heterogeneous systems.
3660 @section Debugging Forks
3662 @cindex fork, debugging programs which call
3663 @cindex multiple processes
3664 @cindex processes, multiple
3665 On most systems, @value{GDBN} has no special support for debugging
3666 programs which create additional processes using the @code{fork}
3667 function. When a program forks, @value{GDBN} will continue to debug the
3668 parent process and the child process will run unimpeded. If you have
3669 set a breakpoint in any code which the child then executes, the child
3670 will get a @code{SIGTRAP} signal which (unless it catches the signal)
3671 will cause it to terminate.
3673 However, if you want to debug the child process there is a workaround
3674 which isn't too painful. Put a call to @code{sleep} in the code which
3675 the child process executes after the fork. It may be useful to sleep
3676 only if a certain environment variable is set, or a certain file exists,
3677 so that the delay need not occur when you don't want to run @value{GDBN}
3678 on the child. While the child is sleeping, use the @code{ps} program to
3679 get its process ID. Then tell @value{GDBN} (a new invocation of
3680 @value{GDBN} if you are also debugging the parent process) to attach to
3681 the child process (@pxref{Attach}). From that point on you can debug
3682 the child process just like any other process which you attached to.
3684 On some systems, @value{GDBN} provides support for debugging programs
3685 that create additional processes using the @code{fork} or @code{vfork}
3686 functions. On @sc{gnu}/Linux platforms, this feature is supported
3687 with kernel version 2.5.46 and later.
3689 The fork debugging commands are supported in native mode and when
3690 connected to @code{gdbserver} in either @code{target remote} mode or
3691 @code{target extended-remote} mode.
3693 By default, when a program forks, @value{GDBN} will continue to debug
3694 the parent process and the child process will run unimpeded.
3696 If you want to follow the child process instead of the parent process,
3697 use the command @w{@code{set follow-fork-mode}}.
3700 @kindex set follow-fork-mode
3701 @item set follow-fork-mode @var{mode}
3702 Set the debugger response to a program call of @code{fork} or
3703 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3704 process. The @var{mode} argument can be:
3708 The original process is debugged after a fork. The child process runs
3709 unimpeded. This is the default.
3712 The new process is debugged after a fork. The parent process runs
3717 @kindex show follow-fork-mode
3718 @item show follow-fork-mode
3719 Display the current debugger response to a @code{fork} or @code{vfork} call.
3722 @cindex debugging multiple processes
3723 On Linux, if you want to debug both the parent and child processes, use the
3724 command @w{@code{set detach-on-fork}}.
3727 @kindex set detach-on-fork
3728 @item set detach-on-fork @var{mode}
3729 Tells gdb whether to detach one of the processes after a fork, or
3730 retain debugger control over them both.
3734 The child process (or parent process, depending on the value of
3735 @code{follow-fork-mode}) will be detached and allowed to run
3736 independently. This is the default.
3739 Both processes will be held under the control of @value{GDBN}.
3740 One process (child or parent, depending on the value of
3741 @code{follow-fork-mode}) is debugged as usual, while the other
3746 @kindex show detach-on-fork
3747 @item show detach-on-fork
3748 Show whether detach-on-fork mode is on/off.
3751 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3752 will retain control of all forked processes (including nested forks).
3753 You can list the forked processes under the control of @value{GDBN} by
3754 using the @w{@code{info inferiors}} command, and switch from one fork
3755 to another by using the @code{inferior} command (@pxref{Inferiors and
3756 Programs, ,Debugging Multiple Inferiors and Programs}).
3758 To quit debugging one of the forked processes, you can either detach
3759 from it by using the @w{@code{detach inferiors}} command (allowing it
3760 to run independently), or kill it using the @w{@code{kill inferiors}}
3761 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3764 If you ask to debug a child process and a @code{vfork} is followed by an
3765 @code{exec}, @value{GDBN} executes the new target up to the first
3766 breakpoint in the new target. If you have a breakpoint set on
3767 @code{main} in your original program, the breakpoint will also be set on
3768 the child process's @code{main}.
3770 On some systems, when a child process is spawned by @code{vfork}, you
3771 cannot debug the child or parent until an @code{exec} call completes.
3773 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3774 call executes, the new target restarts. To restart the parent
3775 process, use the @code{file} command with the parent executable name
3776 as its argument. By default, after an @code{exec} call executes,
3777 @value{GDBN} discards the symbols of the previous executable image.
3778 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3782 @kindex set follow-exec-mode
3783 @item set follow-exec-mode @var{mode}
3785 Set debugger response to a program call of @code{exec}. An
3786 @code{exec} call replaces the program image of a process.
3788 @code{follow-exec-mode} can be:
3792 @value{GDBN} creates a new inferior and rebinds the process to this
3793 new inferior. The program the process was running before the
3794 @code{exec} call can be restarted afterwards by restarting the
3800 (@value{GDBP}) info inferiors
3801 (@value{GDBP}) info inferior
3802 Id Description Executable
3805 process 12020 is executing new program: prog2
3806 Program exited normally.
3807 (@value{GDBP}) info inferiors
3808 Id Description Executable
3814 @value{GDBN} keeps the process bound to the same inferior. The new
3815 executable image replaces the previous executable loaded in the
3816 inferior. Restarting the inferior after the @code{exec} call, with
3817 e.g., the @code{run} command, restarts the executable the process was
3818 running after the @code{exec} call. This is the default mode.
3823 (@value{GDBP}) info inferiors
3824 Id Description Executable
3827 process 12020 is executing new program: prog2
3828 Program exited normally.
3829 (@value{GDBP}) info inferiors
3830 Id Description Executable
3837 @code{follow-exec-mode} is supported in native mode and
3838 @code{target extended-remote} mode.
3840 You can use the @code{catch} command to make @value{GDBN} stop whenever
3841 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3842 Catchpoints, ,Setting Catchpoints}.
3844 @node Checkpoint/Restart
3845 @section Setting a @emph{Bookmark} to Return to Later
3850 @cindex snapshot of a process
3851 @cindex rewind program state
3853 On certain operating systems@footnote{Currently, only
3854 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3855 program's state, called a @dfn{checkpoint}, and come back to it
3858 Returning to a checkpoint effectively undoes everything that has
3859 happened in the program since the @code{checkpoint} was saved. This
3860 includes changes in memory, registers, and even (within some limits)
3861 system state. Effectively, it is like going back in time to the
3862 moment when the checkpoint was saved.
3864 Thus, if you're stepping thru a program and you think you're
3865 getting close to the point where things go wrong, you can save
3866 a checkpoint. Then, if you accidentally go too far and miss
3867 the critical statement, instead of having to restart your program
3868 from the beginning, you can just go back to the checkpoint and
3869 start again from there.
3871 This can be especially useful if it takes a lot of time or
3872 steps to reach the point where you think the bug occurs.
3874 To use the @code{checkpoint}/@code{restart} method of debugging:
3879 Save a snapshot of the debugged program's current execution state.
3880 The @code{checkpoint} command takes no arguments, but each checkpoint
3881 is assigned a small integer id, similar to a breakpoint id.
3883 @kindex info checkpoints
3884 @item info checkpoints
3885 List the checkpoints that have been saved in the current debugging
3886 session. For each checkpoint, the following information will be
3893 @item Source line, or label
3896 @kindex restart @var{checkpoint-id}
3897 @item restart @var{checkpoint-id}
3898 Restore the program state that was saved as checkpoint number
3899 @var{checkpoint-id}. All program variables, registers, stack frames
3900 etc.@: will be returned to the values that they had when the checkpoint
3901 was saved. In essence, gdb will ``wind back the clock'' to the point
3902 in time when the checkpoint was saved.
3904 Note that breakpoints, @value{GDBN} variables, command history etc.
3905 are not affected by restoring a checkpoint. In general, a checkpoint
3906 only restores things that reside in the program being debugged, not in
3909 @kindex delete checkpoint @var{checkpoint-id}
3910 @item delete checkpoint @var{checkpoint-id}
3911 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3915 Returning to a previously saved checkpoint will restore the user state
3916 of the program being debugged, plus a significant subset of the system
3917 (OS) state, including file pointers. It won't ``un-write'' data from
3918 a file, but it will rewind the file pointer to the previous location,
3919 so that the previously written data can be overwritten. For files
3920 opened in read mode, the pointer will also be restored so that the
3921 previously read data can be read again.
3923 Of course, characters that have been sent to a printer (or other
3924 external device) cannot be ``snatched back'', and characters received
3925 from eg.@: a serial device can be removed from internal program buffers,
3926 but they cannot be ``pushed back'' into the serial pipeline, ready to
3927 be received again. Similarly, the actual contents of files that have
3928 been changed cannot be restored (at this time).
3930 However, within those constraints, you actually can ``rewind'' your
3931 program to a previously saved point in time, and begin debugging it
3932 again --- and you can change the course of events so as to debug a
3933 different execution path this time.
3935 @cindex checkpoints and process id
3936 Finally, there is one bit of internal program state that will be
3937 different when you return to a checkpoint --- the program's process
3938 id. Each checkpoint will have a unique process id (or @var{pid}),
3939 and each will be different from the program's original @var{pid}.
3940 If your program has saved a local copy of its process id, this could
3941 potentially pose a problem.
3943 @subsection A Non-obvious Benefit of Using Checkpoints
3945 On some systems such as @sc{gnu}/Linux, address space randomization
3946 is performed on new processes for security reasons. This makes it
3947 difficult or impossible to set a breakpoint, or watchpoint, on an
3948 absolute address if you have to restart the program, since the
3949 absolute location of a symbol will change from one execution to the
3952 A checkpoint, however, is an @emph{identical} copy of a process.
3953 Therefore if you create a checkpoint at (eg.@:) the start of main,
3954 and simply return to that checkpoint instead of restarting the
3955 process, you can avoid the effects of address randomization and
3956 your symbols will all stay in the same place.
3959 @chapter Stopping and Continuing
3961 The principal purposes of using a debugger are so that you can stop your
3962 program before it terminates; or so that, if your program runs into
3963 trouble, you can investigate and find out why.
3965 Inside @value{GDBN}, your program may stop for any of several reasons,
3966 such as a signal, a breakpoint, or reaching a new line after a
3967 @value{GDBN} command such as @code{step}. You may then examine and
3968 change variables, set new breakpoints or remove old ones, and then
3969 continue execution. Usually, the messages shown by @value{GDBN} provide
3970 ample explanation of the status of your program---but you can also
3971 explicitly request this information at any time.
3974 @kindex info program
3976 Display information about the status of your program: whether it is
3977 running or not, what process it is, and why it stopped.
3981 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3982 * Continuing and Stepping:: Resuming execution
3983 * Skipping Over Functions and Files::
3984 Skipping over functions and files
3986 * Thread Stops:: Stopping and starting multi-thread programs
3990 @section Breakpoints, Watchpoints, and Catchpoints
3993 A @dfn{breakpoint} makes your program stop whenever a certain point in
3994 the program is reached. For each breakpoint, you can add conditions to
3995 control in finer detail whether your program stops. You can set
3996 breakpoints with the @code{break} command and its variants (@pxref{Set
3997 Breaks, ,Setting Breakpoints}), to specify the place where your program
3998 should stop by line number, function name or exact address in the
4001 On some systems, you can set breakpoints in shared libraries before
4002 the executable is run.
4005 @cindex data breakpoints
4006 @cindex memory tracing
4007 @cindex breakpoint on memory address
4008 @cindex breakpoint on variable modification
4009 A @dfn{watchpoint} is a special breakpoint that stops your program
4010 when the value of an expression changes. The expression may be a value
4011 of a variable, or it could involve values of one or more variables
4012 combined by operators, such as @samp{a + b}. This is sometimes called
4013 @dfn{data breakpoints}. You must use a different command to set
4014 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
4015 from that, you can manage a watchpoint like any other breakpoint: you
4016 enable, disable, and delete both breakpoints and watchpoints using the
4019 You can arrange to have values from your program displayed automatically
4020 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
4024 @cindex breakpoint on events
4025 A @dfn{catchpoint} is another special breakpoint that stops your program
4026 when a certain kind of event occurs, such as the throwing of a C@t{++}
4027 exception or the loading of a library. As with watchpoints, you use a
4028 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
4029 Catchpoints}), but aside from that, you can manage a catchpoint like any
4030 other breakpoint. (To stop when your program receives a signal, use the
4031 @code{handle} command; see @ref{Signals, ,Signals}.)
4033 @cindex breakpoint numbers
4034 @cindex numbers for breakpoints
4035 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
4036 catchpoint when you create it; these numbers are successive integers
4037 starting with one. In many of the commands for controlling various
4038 features of breakpoints you use the breakpoint number to say which
4039 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
4040 @dfn{disabled}; if disabled, it has no effect on your program until you
4043 @cindex breakpoint ranges
4044 @cindex breakpoint lists
4045 @cindex ranges of breakpoints
4046 @cindex lists of breakpoints
4047 Some @value{GDBN} commands accept a space-separated list of breakpoints
4048 on which to operate. A list element can be either a single breakpoint number,
4049 like @samp{5}, or a range of such numbers, like @samp{5-7}.
4050 When a breakpoint list is given to a command, all breakpoints in that list
4054 * Set Breaks:: Setting breakpoints
4055 * Set Watchpoints:: Setting watchpoints
4056 * Set Catchpoints:: Setting catchpoints
4057 * Delete Breaks:: Deleting breakpoints
4058 * Disabling:: Disabling breakpoints
4059 * Conditions:: Break conditions
4060 * Break Commands:: Breakpoint command lists
4061 * Dynamic Printf:: Dynamic printf
4062 * Save Breakpoints:: How to save breakpoints in a file
4063 * Static Probe Points:: Listing static probe points
4064 * Error in Breakpoints:: ``Cannot insert breakpoints''
4065 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
4069 @subsection Setting Breakpoints
4071 @c FIXME LMB what does GDB do if no code on line of breakpt?
4072 @c consider in particular declaration with/without initialization.
4074 @c FIXME 2 is there stuff on this already? break at fun start, already init?
4077 @kindex b @r{(@code{break})}
4078 @vindex $bpnum@r{, convenience variable}
4079 @cindex latest breakpoint
4080 Breakpoints are set with the @code{break} command (abbreviated
4081 @code{b}). The debugger convenience variable @samp{$bpnum} records the
4082 number of the breakpoint you've set most recently; see @ref{Convenience
4083 Vars,, Convenience Variables}, for a discussion of what you can do with
4084 convenience variables.
4087 @item break @var{location}
4088 Set a breakpoint at the given @var{location}, which can specify a
4089 function name, a line number, or an address of an instruction.
4090 (@xref{Specify Location}, for a list of all the possible ways to
4091 specify a @var{location}.) The breakpoint will stop your program just
4092 before it executes any of the code in the specified @var{location}.
4094 When using source languages that permit overloading of symbols, such as
4095 C@t{++}, a function name may refer to more than one possible place to break.
4096 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
4099 It is also possible to insert a breakpoint that will stop the program
4100 only if a specific thread (@pxref{Thread-Specific Breakpoints})
4101 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
4104 When called without any arguments, @code{break} sets a breakpoint at
4105 the next instruction to be executed in the selected stack frame
4106 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
4107 innermost, this makes your program stop as soon as control
4108 returns to that frame. This is similar to the effect of a
4109 @code{finish} command in the frame inside the selected frame---except
4110 that @code{finish} does not leave an active breakpoint. If you use
4111 @code{break} without an argument in the innermost frame, @value{GDBN} stops
4112 the next time it reaches the current location; this may be useful
4115 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
4116 least one instruction has been executed. If it did not do this, you
4117 would be unable to proceed past a breakpoint without first disabling the
4118 breakpoint. This rule applies whether or not the breakpoint already
4119 existed when your program stopped.
4121 @item break @dots{} if @var{cond}
4122 Set a breakpoint with condition @var{cond}; evaluate the expression
4123 @var{cond} each time the breakpoint is reached, and stop only if the
4124 value is nonzero---that is, if @var{cond} evaluates as true.
4125 @samp{@dots{}} stands for one of the possible arguments described
4126 above (or no argument) specifying where to break. @xref{Conditions,
4127 ,Break Conditions}, for more information on breakpoint conditions.
4130 @item tbreak @var{args}
4131 Set a breakpoint enabled only for one stop. The @var{args} are the
4132 same as for the @code{break} command, and the breakpoint is set in the same
4133 way, but the breakpoint is automatically deleted after the first time your
4134 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
4137 @cindex hardware breakpoints
4138 @item hbreak @var{args}
4139 Set a hardware-assisted breakpoint. The @var{args} are the same as for the
4140 @code{break} command and the breakpoint is set in the same way, but the
4141 breakpoint requires hardware support and some target hardware may not
4142 have this support. The main purpose of this is EPROM/ROM code
4143 debugging, so you can set a breakpoint at an instruction without
4144 changing the instruction. This can be used with the new trap-generation
4145 provided by SPARClite DSU and most x86-based targets. These targets
4146 will generate traps when a program accesses some data or instruction
4147 address that is assigned to the debug registers. However the hardware
4148 breakpoint registers can take a limited number of breakpoints. For
4149 example, on the DSU, only two data breakpoints can be set at a time, and
4150 @value{GDBN} will reject this command if more than two are used. Delete
4151 or disable unused hardware breakpoints before setting new ones
4152 (@pxref{Disabling, ,Disabling Breakpoints}).
4153 @xref{Conditions, ,Break Conditions}.
4154 For remote targets, you can restrict the number of hardware
4155 breakpoints @value{GDBN} will use, see @ref{set remote
4156 hardware-breakpoint-limit}.
4159 @item thbreak @var{args}
4160 Set a hardware-assisted breakpoint enabled only for one stop. The @var{args}
4161 are the same as for the @code{hbreak} command and the breakpoint is set in
4162 the same way. However, like the @code{tbreak} command,
4163 the breakpoint is automatically deleted after the
4164 first time your program stops there. Also, like the @code{hbreak}
4165 command, the breakpoint requires hardware support and some target hardware
4166 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
4167 See also @ref{Conditions, ,Break Conditions}.
4170 @cindex regular expression
4171 @cindex breakpoints at functions matching a regexp
4172 @cindex set breakpoints in many functions
4173 @item rbreak @var{regex}
4174 Set breakpoints on all functions matching the regular expression
4175 @var{regex}. This command sets an unconditional breakpoint on all
4176 matches, printing a list of all breakpoints it set. Once these
4177 breakpoints are set, they are treated just like the breakpoints set with
4178 the @code{break} command. You can delete them, disable them, or make
4179 them conditional the same way as any other breakpoint.
4181 In programs using different languages, @value{GDBN} chooses the syntax
4182 to print the list of all breakpoints it sets according to the
4183 @samp{set language} value: using @samp{set language auto}
4184 (see @ref{Automatically, ,Set Language Automatically}) means to use the
4185 language of the breakpoint's function, other values mean to use
4186 the manually specified language (see @ref{Manually, ,Set Language Manually}).
4188 The syntax of the regular expression is the standard one used with tools
4189 like @file{grep}. Note that this is different from the syntax used by
4190 shells, so for instance @code{foo*} matches all functions that include
4191 an @code{fo} followed by zero or more @code{o}s. There is an implicit
4192 @code{.*} leading and trailing the regular expression you supply, so to
4193 match only functions that begin with @code{foo}, use @code{^foo}.
4195 @cindex non-member C@t{++} functions, set breakpoint in
4196 When debugging C@t{++} programs, @code{rbreak} is useful for setting
4197 breakpoints on overloaded functions that are not members of any special
4200 @cindex set breakpoints on all functions
4201 The @code{rbreak} command can be used to set breakpoints in
4202 @strong{all} the functions in a program, like this:
4205 (@value{GDBP}) rbreak .
4208 @item rbreak @var{file}:@var{regex}
4209 If @code{rbreak} is called with a filename qualification, it limits
4210 the search for functions matching the given regular expression to the
4211 specified @var{file}. This can be used, for example, to set breakpoints on
4212 every function in a given file:
4215 (@value{GDBP}) rbreak file.c:.
4218 The colon separating the filename qualifier from the regex may
4219 optionally be surrounded by spaces.
4221 @kindex info breakpoints
4222 @cindex @code{$_} and @code{info breakpoints}
4223 @item info breakpoints @r{[}@var{list}@dots{}@r{]}
4224 @itemx info break @r{[}@var{list}@dots{}@r{]}
4225 Print a table of all breakpoints, watchpoints, and catchpoints set and
4226 not deleted. Optional argument @var{n} means print information only
4227 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
4228 For each breakpoint, following columns are printed:
4231 @item Breakpoint Numbers
4233 Breakpoint, watchpoint, or catchpoint.
4235 Whether the breakpoint is marked to be disabled or deleted when hit.
4236 @item Enabled or Disabled
4237 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
4238 that are not enabled.
4240 Where the breakpoint is in your program, as a memory address. For a
4241 pending breakpoint whose address is not yet known, this field will
4242 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
4243 library that has the symbol or line referred by breakpoint is loaded.
4244 See below for details. A breakpoint with several locations will
4245 have @samp{<MULTIPLE>} in this field---see below for details.
4247 Where the breakpoint is in the source for your program, as a file and
4248 line number. For a pending breakpoint, the original string passed to
4249 the breakpoint command will be listed as it cannot be resolved until
4250 the appropriate shared library is loaded in the future.
4254 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
4255 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
4256 @value{GDBN} on the host's side. If it is ``target'', then the condition
4257 is evaluated by the target. The @code{info break} command shows
4258 the condition on the line following the affected breakpoint, together with
4259 its condition evaluation mode in between parentheses.
4261 Breakpoint commands, if any, are listed after that. A pending breakpoint is
4262 allowed to have a condition specified for it. The condition is not parsed for
4263 validity until a shared library is loaded that allows the pending
4264 breakpoint to resolve to a valid location.
4267 @code{info break} with a breakpoint
4268 number @var{n} as argument lists only that breakpoint. The
4269 convenience variable @code{$_} and the default examining-address for
4270 the @code{x} command are set to the address of the last breakpoint
4271 listed (@pxref{Memory, ,Examining Memory}).
4274 @code{info break} displays a count of the number of times the breakpoint
4275 has been hit. This is especially useful in conjunction with the
4276 @code{ignore} command. You can ignore a large number of breakpoint
4277 hits, look at the breakpoint info to see how many times the breakpoint
4278 was hit, and then run again, ignoring one less than that number. This
4279 will get you quickly to the last hit of that breakpoint.
4282 For a breakpoints with an enable count (xref) greater than 1,
4283 @code{info break} also displays that count.
4287 @value{GDBN} allows you to set any number of breakpoints at the same place in
4288 your program. There is nothing silly or meaningless about this. When
4289 the breakpoints are conditional, this is even useful
4290 (@pxref{Conditions, ,Break Conditions}).
4292 @cindex multiple locations, breakpoints
4293 @cindex breakpoints, multiple locations
4294 It is possible that a breakpoint corresponds to several locations
4295 in your program. Examples of this situation are:
4299 Multiple functions in the program may have the same name.
4302 For a C@t{++} constructor, the @value{NGCC} compiler generates several
4303 instances of the function body, used in different cases.
4306 For a C@t{++} template function, a given line in the function can
4307 correspond to any number of instantiations.
4310 For an inlined function, a given source line can correspond to
4311 several places where that function is inlined.
4314 In all those cases, @value{GDBN} will insert a breakpoint at all
4315 the relevant locations.
4317 A breakpoint with multiple locations is displayed in the breakpoint
4318 table using several rows---one header row, followed by one row for
4319 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
4320 address column. The rows for individual locations contain the actual
4321 addresses for locations, and show the functions to which those
4322 locations belong. The number column for a location is of the form
4323 @var{breakpoint-number}.@var{location-number}.
4328 Num Type Disp Enb Address What
4329 1 breakpoint keep y <MULTIPLE>
4331 breakpoint already hit 1 time
4332 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
4333 1.2 y 0x080486ca in void foo<double>() at t.cc:8
4336 You cannot delete the individual locations from a breakpoint. However,
4337 each location can be individually enabled or disabled by passing
4338 @var{breakpoint-number}.@var{location-number} as argument to the
4339 @code{enable} and @code{disable} commands. It's also possible to
4340 @code{enable} and @code{disable} a range of @var{location-number}
4341 locations using a @var{breakpoint-number} and two @var{location-number}s,
4342 in increasing order, separated by a hyphen, like
4343 @kbd{@var{breakpoint-number}.@var{location-number1}-@var{location-number2}},
4344 in which case @value{GDBN} acts on all the locations in the range (inclusive).
4345 Disabling or enabling the parent breakpoint (@pxref{Disabling}) affects
4346 all of the locations that belong to that breakpoint.
4348 @cindex pending breakpoints
4349 It's quite common to have a breakpoint inside a shared library.
4350 Shared libraries can be loaded and unloaded explicitly,
4351 and possibly repeatedly, as the program is executed. To support
4352 this use case, @value{GDBN} updates breakpoint locations whenever
4353 any shared library is loaded or unloaded. Typically, you would
4354 set a breakpoint in a shared library at the beginning of your
4355 debugging session, when the library is not loaded, and when the
4356 symbols from the library are not available. When you try to set
4357 breakpoint, @value{GDBN} will ask you if you want to set
4358 a so called @dfn{pending breakpoint}---breakpoint whose address
4359 is not yet resolved.
4361 After the program is run, whenever a new shared library is loaded,
4362 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
4363 shared library contains the symbol or line referred to by some
4364 pending breakpoint, that breakpoint is resolved and becomes an
4365 ordinary breakpoint. When a library is unloaded, all breakpoints
4366 that refer to its symbols or source lines become pending again.
4368 This logic works for breakpoints with multiple locations, too. For
4369 example, if you have a breakpoint in a C@t{++} template function, and
4370 a newly loaded shared library has an instantiation of that template,
4371 a new location is added to the list of locations for the breakpoint.
4373 Except for having unresolved address, pending breakpoints do not
4374 differ from regular breakpoints. You can set conditions or commands,
4375 enable and disable them and perform other breakpoint operations.
4377 @value{GDBN} provides some additional commands for controlling what
4378 happens when the @samp{break} command cannot resolve breakpoint
4379 address specification to an address:
4381 @kindex set breakpoint pending
4382 @kindex show breakpoint pending
4384 @item set breakpoint pending auto
4385 This is the default behavior. When @value{GDBN} cannot find the breakpoint
4386 location, it queries you whether a pending breakpoint should be created.
4388 @item set breakpoint pending on
4389 This indicates that an unrecognized breakpoint location should automatically
4390 result in a pending breakpoint being created.
4392 @item set breakpoint pending off
4393 This indicates that pending breakpoints are not to be created. Any
4394 unrecognized breakpoint location results in an error. This setting does
4395 not affect any pending breakpoints previously created.
4397 @item show breakpoint pending
4398 Show the current behavior setting for creating pending breakpoints.
4401 The settings above only affect the @code{break} command and its
4402 variants. Once breakpoint is set, it will be automatically updated
4403 as shared libraries are loaded and unloaded.
4405 @cindex automatic hardware breakpoints
4406 For some targets, @value{GDBN} can automatically decide if hardware or
4407 software breakpoints should be used, depending on whether the
4408 breakpoint address is read-only or read-write. This applies to
4409 breakpoints set with the @code{break} command as well as to internal
4410 breakpoints set by commands like @code{next} and @code{finish}. For
4411 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
4414 You can control this automatic behaviour with the following commands:
4416 @kindex set breakpoint auto-hw
4417 @kindex show breakpoint auto-hw
4419 @item set breakpoint auto-hw on
4420 This is the default behavior. When @value{GDBN} sets a breakpoint, it
4421 will try to use the target memory map to decide if software or hardware
4422 breakpoint must be used.
4424 @item set breakpoint auto-hw off
4425 This indicates @value{GDBN} should not automatically select breakpoint
4426 type. If the target provides a memory map, @value{GDBN} will warn when
4427 trying to set software breakpoint at a read-only address.
4430 @value{GDBN} normally implements breakpoints by replacing the program code
4431 at the breakpoint address with a special instruction, which, when
4432 executed, given control to the debugger. By default, the program
4433 code is so modified only when the program is resumed. As soon as
4434 the program stops, @value{GDBN} restores the original instructions. This
4435 behaviour guards against leaving breakpoints inserted in the
4436 target should gdb abrubptly disconnect. However, with slow remote
4437 targets, inserting and removing breakpoint can reduce the performance.
4438 This behavior can be controlled with the following commands::
4440 @kindex set breakpoint always-inserted
4441 @kindex show breakpoint always-inserted
4443 @item set breakpoint always-inserted off
4444 All breakpoints, including newly added by the user, are inserted in
4445 the target only when the target is resumed. All breakpoints are
4446 removed from the target when it stops. This is the default mode.
4448 @item set breakpoint always-inserted on
4449 Causes all breakpoints to be inserted in the target at all times. If
4450 the user adds a new breakpoint, or changes an existing breakpoint, the
4451 breakpoints in the target are updated immediately. A breakpoint is
4452 removed from the target only when breakpoint itself is deleted.
4455 @value{GDBN} handles conditional breakpoints by evaluating these conditions
4456 when a breakpoint breaks. If the condition is true, then the process being
4457 debugged stops, otherwise the process is resumed.
4459 If the target supports evaluating conditions on its end, @value{GDBN} may
4460 download the breakpoint, together with its conditions, to it.
4462 This feature can be controlled via the following commands:
4464 @kindex set breakpoint condition-evaluation
4465 @kindex show breakpoint condition-evaluation
4467 @item set breakpoint condition-evaluation host
4468 This option commands @value{GDBN} to evaluate the breakpoint
4469 conditions on the host's side. Unconditional breakpoints are sent to
4470 the target which in turn receives the triggers and reports them back to GDB
4471 for condition evaluation. This is the standard evaluation mode.
4473 @item set breakpoint condition-evaluation target
4474 This option commands @value{GDBN} to download breakpoint conditions
4475 to the target at the moment of their insertion. The target
4476 is responsible for evaluating the conditional expression and reporting
4477 breakpoint stop events back to @value{GDBN} whenever the condition
4478 is true. Due to limitations of target-side evaluation, some conditions
4479 cannot be evaluated there, e.g., conditions that depend on local data
4480 that is only known to the host. Examples include
4481 conditional expressions involving convenience variables, complex types
4482 that cannot be handled by the agent expression parser and expressions
4483 that are too long to be sent over to the target, specially when the
4484 target is a remote system. In these cases, the conditions will be
4485 evaluated by @value{GDBN}.
4487 @item set breakpoint condition-evaluation auto
4488 This is the default mode. If the target supports evaluating breakpoint
4489 conditions on its end, @value{GDBN} will download breakpoint conditions to
4490 the target (limitations mentioned previously apply). If the target does
4491 not support breakpoint condition evaluation, then @value{GDBN} will fallback
4492 to evaluating all these conditions on the host's side.
4496 @cindex negative breakpoint numbers
4497 @cindex internal @value{GDBN} breakpoints
4498 @value{GDBN} itself sometimes sets breakpoints in your program for
4499 special purposes, such as proper handling of @code{longjmp} (in C
4500 programs). These internal breakpoints are assigned negative numbers,
4501 starting with @code{-1}; @samp{info breakpoints} does not display them.
4502 You can see these breakpoints with the @value{GDBN} maintenance command
4503 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
4506 @node Set Watchpoints
4507 @subsection Setting Watchpoints
4509 @cindex setting watchpoints
4510 You can use a watchpoint to stop execution whenever the value of an
4511 expression changes, without having to predict a particular place where
4512 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
4513 The expression may be as simple as the value of a single variable, or
4514 as complex as many variables combined by operators. Examples include:
4518 A reference to the value of a single variable.
4521 An address cast to an appropriate data type. For example,
4522 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
4523 address (assuming an @code{int} occupies 4 bytes).
4526 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
4527 expression can use any operators valid in the program's native
4528 language (@pxref{Languages}).
4531 You can set a watchpoint on an expression even if the expression can
4532 not be evaluated yet. For instance, you can set a watchpoint on
4533 @samp{*global_ptr} before @samp{global_ptr} is initialized.
4534 @value{GDBN} will stop when your program sets @samp{global_ptr} and
4535 the expression produces a valid value. If the expression becomes
4536 valid in some other way than changing a variable (e.g.@: if the memory
4537 pointed to by @samp{*global_ptr} becomes readable as the result of a
4538 @code{malloc} call), @value{GDBN} may not stop until the next time
4539 the expression changes.
4541 @cindex software watchpoints
4542 @cindex hardware watchpoints
4543 Depending on your system, watchpoints may be implemented in software or
4544 hardware. @value{GDBN} does software watchpointing by single-stepping your
4545 program and testing the variable's value each time, which is hundreds of
4546 times slower than normal execution. (But this may still be worth it, to
4547 catch errors where you have no clue what part of your program is the
4550 On some systems, such as most PowerPC or x86-based targets,
4551 @value{GDBN} includes support for hardware watchpoints, which do not
4552 slow down the running of your program.
4556 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4557 Set a watchpoint for an expression. @value{GDBN} will break when the
4558 expression @var{expr} is written into by the program and its value
4559 changes. The simplest (and the most popular) use of this command is
4560 to watch the value of a single variable:
4563 (@value{GDBP}) watch foo
4566 If the command includes a @code{@r{[}thread @var{thread-id}@r{]}}
4567 argument, @value{GDBN} breaks only when the thread identified by
4568 @var{thread-id} changes the value of @var{expr}. If any other threads
4569 change the value of @var{expr}, @value{GDBN} will not break. Note
4570 that watchpoints restricted to a single thread in this way only work
4571 with Hardware Watchpoints.
4573 Ordinarily a watchpoint respects the scope of variables in @var{expr}
4574 (see below). The @code{-location} argument tells @value{GDBN} to
4575 instead watch the memory referred to by @var{expr}. In this case,
4576 @value{GDBN} will evaluate @var{expr}, take the address of the result,
4577 and watch the memory at that address. The type of the result is used
4578 to determine the size of the watched memory. If the expression's
4579 result does not have an address, then @value{GDBN} will print an
4582 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
4583 of masked watchpoints, if the current architecture supports this
4584 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
4585 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
4586 to an address to watch. The mask specifies that some bits of an address
4587 (the bits which are reset in the mask) should be ignored when matching
4588 the address accessed by the inferior against the watchpoint address.
4589 Thus, a masked watchpoint watches many addresses simultaneously---those
4590 addresses whose unmasked bits are identical to the unmasked bits in the
4591 watchpoint address. The @code{mask} argument implies @code{-location}.
4595 (@value{GDBP}) watch foo mask 0xffff00ff
4596 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
4600 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4601 Set a watchpoint that will break when the value of @var{expr} is read
4605 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4606 Set a watchpoint that will break when @var{expr} is either read from
4607 or written into by the program.
4609 @kindex info watchpoints @r{[}@var{list}@dots{}@r{]}
4610 @item info watchpoints @r{[}@var{list}@dots{}@r{]}
4611 This command prints a list of watchpoints, using the same format as
4612 @code{info break} (@pxref{Set Breaks}).
4615 If you watch for a change in a numerically entered address you need to
4616 dereference it, as the address itself is just a constant number which will
4617 never change. @value{GDBN} refuses to create a watchpoint that watches
4618 a never-changing value:
4621 (@value{GDBP}) watch 0x600850
4622 Cannot watch constant value 0x600850.
4623 (@value{GDBP}) watch *(int *) 0x600850
4624 Watchpoint 1: *(int *) 6293584
4627 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
4628 watchpoints execute very quickly, and the debugger reports a change in
4629 value at the exact instruction where the change occurs. If @value{GDBN}
4630 cannot set a hardware watchpoint, it sets a software watchpoint, which
4631 executes more slowly and reports the change in value at the next
4632 @emph{statement}, not the instruction, after the change occurs.
4634 @cindex use only software watchpoints
4635 You can force @value{GDBN} to use only software watchpoints with the
4636 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
4637 zero, @value{GDBN} will never try to use hardware watchpoints, even if
4638 the underlying system supports them. (Note that hardware-assisted
4639 watchpoints that were set @emph{before} setting
4640 @code{can-use-hw-watchpoints} to zero will still use the hardware
4641 mechanism of watching expression values.)
4644 @item set can-use-hw-watchpoints
4645 @kindex set can-use-hw-watchpoints
4646 Set whether or not to use hardware watchpoints.
4648 @item show can-use-hw-watchpoints
4649 @kindex show can-use-hw-watchpoints
4650 Show the current mode of using hardware watchpoints.
4653 For remote targets, you can restrict the number of hardware
4654 watchpoints @value{GDBN} will use, see @ref{set remote
4655 hardware-breakpoint-limit}.
4657 When you issue the @code{watch} command, @value{GDBN} reports
4660 Hardware watchpoint @var{num}: @var{expr}
4664 if it was able to set a hardware watchpoint.
4666 Currently, the @code{awatch} and @code{rwatch} commands can only set
4667 hardware watchpoints, because accesses to data that don't change the
4668 value of the watched expression cannot be detected without examining
4669 every instruction as it is being executed, and @value{GDBN} does not do
4670 that currently. If @value{GDBN} finds that it is unable to set a
4671 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
4672 will print a message like this:
4675 Expression cannot be implemented with read/access watchpoint.
4678 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4679 data type of the watched expression is wider than what a hardware
4680 watchpoint on the target machine can handle. For example, some systems
4681 can only watch regions that are up to 4 bytes wide; on such systems you
4682 cannot set hardware watchpoints for an expression that yields a
4683 double-precision floating-point number (which is typically 8 bytes
4684 wide). As a work-around, it might be possible to break the large region
4685 into a series of smaller ones and watch them with separate watchpoints.
4687 If you set too many hardware watchpoints, @value{GDBN} might be unable
4688 to insert all of them when you resume the execution of your program.
4689 Since the precise number of active watchpoints is unknown until such
4690 time as the program is about to be resumed, @value{GDBN} might not be
4691 able to warn you about this when you set the watchpoints, and the
4692 warning will be printed only when the program is resumed:
4695 Hardware watchpoint @var{num}: Could not insert watchpoint
4699 If this happens, delete or disable some of the watchpoints.
4701 Watching complex expressions that reference many variables can also
4702 exhaust the resources available for hardware-assisted watchpoints.
4703 That's because @value{GDBN} needs to watch every variable in the
4704 expression with separately allocated resources.
4706 If you call a function interactively using @code{print} or @code{call},
4707 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4708 kind of breakpoint or the call completes.
4710 @value{GDBN} automatically deletes watchpoints that watch local
4711 (automatic) variables, or expressions that involve such variables, when
4712 they go out of scope, that is, when the execution leaves the block in
4713 which these variables were defined. In particular, when the program
4714 being debugged terminates, @emph{all} local variables go out of scope,
4715 and so only watchpoints that watch global variables remain set. If you
4716 rerun the program, you will need to set all such watchpoints again. One
4717 way of doing that would be to set a code breakpoint at the entry to the
4718 @code{main} function and when it breaks, set all the watchpoints.
4720 @cindex watchpoints and threads
4721 @cindex threads and watchpoints
4722 In multi-threaded programs, watchpoints will detect changes to the
4723 watched expression from every thread.
4726 @emph{Warning:} In multi-threaded programs, software watchpoints
4727 have only limited usefulness. If @value{GDBN} creates a software
4728 watchpoint, it can only watch the value of an expression @emph{in a
4729 single thread}. If you are confident that the expression can only
4730 change due to the current thread's activity (and if you are also
4731 confident that no other thread can become current), then you can use
4732 software watchpoints as usual. However, @value{GDBN} may not notice
4733 when a non-current thread's activity changes the expression. (Hardware
4734 watchpoints, in contrast, watch an expression in all threads.)
4737 @xref{set remote hardware-watchpoint-limit}.
4739 @node Set Catchpoints
4740 @subsection Setting Catchpoints
4741 @cindex catchpoints, setting
4742 @cindex exception handlers
4743 @cindex event handling
4745 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4746 kinds of program events, such as C@t{++} exceptions or the loading of a
4747 shared library. Use the @code{catch} command to set a catchpoint.
4751 @item catch @var{event}
4752 Stop when @var{event} occurs. The @var{event} can be any of the following:
4755 @item throw @r{[}@var{regexp}@r{]}
4756 @itemx rethrow @r{[}@var{regexp}@r{]}
4757 @itemx catch @r{[}@var{regexp}@r{]}
4759 @kindex catch rethrow
4761 @cindex stop on C@t{++} exceptions
4762 The throwing, re-throwing, or catching of a C@t{++} exception.
4764 If @var{regexp} is given, then only exceptions whose type matches the
4765 regular expression will be caught.
4767 @vindex $_exception@r{, convenience variable}
4768 The convenience variable @code{$_exception} is available at an
4769 exception-related catchpoint, on some systems. This holds the
4770 exception being thrown.
4772 There are currently some limitations to C@t{++} exception handling in
4777 The support for these commands is system-dependent. Currently, only
4778 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4782 The regular expression feature and the @code{$_exception} convenience
4783 variable rely on the presence of some SDT probes in @code{libstdc++}.
4784 If these probes are not present, then these features cannot be used.
4785 These probes were first available in the GCC 4.8 release, but whether
4786 or not they are available in your GCC also depends on how it was
4790 The @code{$_exception} convenience variable is only valid at the
4791 instruction at which an exception-related catchpoint is set.
4794 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4795 location in the system library which implements runtime exception
4796 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4797 (@pxref{Selection}) to get to your code.
4800 If you call a function interactively, @value{GDBN} normally returns
4801 control to you when the function has finished executing. If the call
4802 raises an exception, however, the call may bypass the mechanism that
4803 returns control to you and cause your program either to abort or to
4804 simply continue running until it hits a breakpoint, catches a signal
4805 that @value{GDBN} is listening for, or exits. This is the case even if
4806 you set a catchpoint for the exception; catchpoints on exceptions are
4807 disabled within interactive calls. @xref{Calling}, for information on
4808 controlling this with @code{set unwind-on-terminating-exception}.
4811 You cannot raise an exception interactively.
4814 You cannot install an exception handler interactively.
4817 @item exception @r{[}@var{name}@r{]}
4818 @kindex catch exception
4819 @cindex Ada exception catching
4820 @cindex catch Ada exceptions
4821 An Ada exception being raised. If an exception name is specified
4822 at the end of the command (eg @code{catch exception Program_Error}),
4823 the debugger will stop only when this specific exception is raised.
4824 Otherwise, the debugger stops execution when any Ada exception is raised.
4826 When inserting an exception catchpoint on a user-defined exception whose
4827 name is identical to one of the exceptions defined by the language, the
4828 fully qualified name must be used as the exception name. Otherwise,
4829 @value{GDBN} will assume that it should stop on the pre-defined exception
4830 rather than the user-defined one. For instance, assuming an exception
4831 called @code{Constraint_Error} is defined in package @code{Pck}, then
4832 the command to use to catch such exceptions is @kbd{catch exception
4833 Pck.Constraint_Error}.
4835 @vindex $_ada_exception@r{, convenience variable}
4836 The convenience variable @code{$_ada_exception} holds the address of
4837 the exception being thrown. This can be useful when setting a
4838 condition for such a catchpoint.
4840 @item exception unhandled
4841 @kindex catch exception unhandled
4842 An exception that was raised but is not handled by the program. The
4843 convenience variable @code{$_ada_exception} is set as for @code{catch
4846 @item handlers @r{[}@var{name}@r{]}
4847 @kindex catch handlers
4848 @cindex Ada exception handlers catching
4849 @cindex catch Ada exceptions when handled
4850 An Ada exception being handled. If an exception name is
4851 specified at the end of the command
4852 (eg @kbd{catch handlers Program_Error}), the debugger will stop
4853 only when this specific exception is handled.
4854 Otherwise, the debugger stops execution when any Ada exception is handled.
4856 When inserting a handlers catchpoint on a user-defined
4857 exception whose name is identical to one of the exceptions
4858 defined by the language, the fully qualified name must be used
4859 as the exception name. Otherwise, @value{GDBN} will assume that it
4860 should stop on the pre-defined exception rather than the
4861 user-defined one. For instance, assuming an exception called
4862 @code{Constraint_Error} is defined in package @code{Pck}, then the
4863 command to use to catch such exceptions handling is
4864 @kbd{catch handlers Pck.Constraint_Error}.
4866 The convenience variable @code{$_ada_exception} is set as for
4867 @code{catch exception}.
4870 @kindex catch assert
4871 A failed Ada assertion. Note that the convenience variable
4872 @code{$_ada_exception} is @emph{not} set by this catchpoint.
4876 @cindex break on fork/exec
4877 A call to @code{exec}.
4879 @anchor{catch syscall}
4881 @itemx syscall @r{[}@var{name} @r{|} @var{number} @r{|} @r{group:}@var{groupname} @r{|} @r{g:}@var{groupname}@r{]} @dots{}
4882 @kindex catch syscall
4883 @cindex break on a system call.
4884 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4885 syscall is a mechanism for application programs to request a service
4886 from the operating system (OS) or one of the OS system services.
4887 @value{GDBN} can catch some or all of the syscalls issued by the
4888 debuggee, and show the related information for each syscall. If no
4889 argument is specified, calls to and returns from all system calls
4892 @var{name} can be any system call name that is valid for the
4893 underlying OS. Just what syscalls are valid depends on the OS. On
4894 GNU and Unix systems, you can find the full list of valid syscall
4895 names on @file{/usr/include/asm/unistd.h}.
4897 @c For MS-Windows, the syscall names and the corresponding numbers
4898 @c can be found, e.g., on this URL:
4899 @c http://www.metasploit.com/users/opcode/syscalls.html
4900 @c but we don't support Windows syscalls yet.
4902 Normally, @value{GDBN} knows in advance which syscalls are valid for
4903 each OS, so you can use the @value{GDBN} command-line completion
4904 facilities (@pxref{Completion,, command completion}) to list the
4907 You may also specify the system call numerically. A syscall's
4908 number is the value passed to the OS's syscall dispatcher to
4909 identify the requested service. When you specify the syscall by its
4910 name, @value{GDBN} uses its database of syscalls to convert the name
4911 into the corresponding numeric code, but using the number directly
4912 may be useful if @value{GDBN}'s database does not have the complete
4913 list of syscalls on your system (e.g., because @value{GDBN} lags
4914 behind the OS upgrades).
4916 You may specify a group of related syscalls to be caught at once using
4917 the @code{group:} syntax (@code{g:} is a shorter equivalent). For
4918 instance, on some platforms @value{GDBN} allows you to catch all
4919 network related syscalls, by passing the argument @code{group:network}
4920 to @code{catch syscall}. Note that not all syscall groups are
4921 available in every system. You can use the command completion
4922 facilities (@pxref{Completion,, command completion}) to list the
4923 syscall groups available on your environment.
4925 The example below illustrates how this command works if you don't provide
4929 (@value{GDBP}) catch syscall
4930 Catchpoint 1 (syscall)
4932 Starting program: /tmp/catch-syscall
4934 Catchpoint 1 (call to syscall 'close'), \
4935 0xffffe424 in __kernel_vsyscall ()
4939 Catchpoint 1 (returned from syscall 'close'), \
4940 0xffffe424 in __kernel_vsyscall ()
4944 Here is an example of catching a system call by name:
4947 (@value{GDBP}) catch syscall chroot
4948 Catchpoint 1 (syscall 'chroot' [61])
4950 Starting program: /tmp/catch-syscall
4952 Catchpoint 1 (call to syscall 'chroot'), \
4953 0xffffe424 in __kernel_vsyscall ()
4957 Catchpoint 1 (returned from syscall 'chroot'), \
4958 0xffffe424 in __kernel_vsyscall ()
4962 An example of specifying a system call numerically. In the case
4963 below, the syscall number has a corresponding entry in the XML
4964 file, so @value{GDBN} finds its name and prints it:
4967 (@value{GDBP}) catch syscall 252
4968 Catchpoint 1 (syscall(s) 'exit_group')
4970 Starting program: /tmp/catch-syscall
4972 Catchpoint 1 (call to syscall 'exit_group'), \
4973 0xffffe424 in __kernel_vsyscall ()
4977 Program exited normally.
4981 Here is an example of catching a syscall group:
4984 (@value{GDBP}) catch syscall group:process
4985 Catchpoint 1 (syscalls 'exit' [1] 'fork' [2] 'waitpid' [7]
4986 'execve' [11] 'wait4' [114] 'clone' [120] 'vfork' [190]
4987 'exit_group' [252] 'waitid' [284] 'unshare' [310])
4989 Starting program: /tmp/catch-syscall
4991 Catchpoint 1 (call to syscall fork), 0x00007ffff7df4e27 in open64 ()
4992 from /lib64/ld-linux-x86-64.so.2
4998 However, there can be situations when there is no corresponding name
4999 in XML file for that syscall number. In this case, @value{GDBN} prints
5000 a warning message saying that it was not able to find the syscall name,
5001 but the catchpoint will be set anyway. See the example below:
5004 (@value{GDBP}) catch syscall 764
5005 warning: The number '764' does not represent a known syscall.
5006 Catchpoint 2 (syscall 764)
5010 If you configure @value{GDBN} using the @samp{--without-expat} option,
5011 it will not be able to display syscall names. Also, if your
5012 architecture does not have an XML file describing its system calls,
5013 you will not be able to see the syscall names. It is important to
5014 notice that these two features are used for accessing the syscall
5015 name database. In either case, you will see a warning like this:
5018 (@value{GDBP}) catch syscall
5019 warning: Could not open "syscalls/i386-linux.xml"
5020 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
5021 GDB will not be able to display syscall names.
5022 Catchpoint 1 (syscall)
5026 Of course, the file name will change depending on your architecture and system.
5028 Still using the example above, you can also try to catch a syscall by its
5029 number. In this case, you would see something like:
5032 (@value{GDBP}) catch syscall 252
5033 Catchpoint 1 (syscall(s) 252)
5036 Again, in this case @value{GDBN} would not be able to display syscall's names.
5040 A call to @code{fork}.
5044 A call to @code{vfork}.
5046 @item load @r{[}@var{regexp}@r{]}
5047 @itemx unload @r{[}@var{regexp}@r{]}
5049 @kindex catch unload
5050 The loading or unloading of a shared library. If @var{regexp} is
5051 given, then the catchpoint will stop only if the regular expression
5052 matches one of the affected libraries.
5054 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
5055 @kindex catch signal
5056 The delivery of a signal.
5058 With no arguments, this catchpoint will catch any signal that is not
5059 used internally by @value{GDBN}, specifically, all signals except
5060 @samp{SIGTRAP} and @samp{SIGINT}.
5062 With the argument @samp{all}, all signals, including those used by
5063 @value{GDBN}, will be caught. This argument cannot be used with other
5066 Otherwise, the arguments are a list of signal names as given to
5067 @code{handle} (@pxref{Signals}). Only signals specified in this list
5070 One reason that @code{catch signal} can be more useful than
5071 @code{handle} is that you can attach commands and conditions to the
5074 When a signal is caught by a catchpoint, the signal's @code{stop} and
5075 @code{print} settings, as specified by @code{handle}, are ignored.
5076 However, whether the signal is still delivered to the inferior depends
5077 on the @code{pass} setting; this can be changed in the catchpoint's
5082 @item tcatch @var{event}
5084 Set a catchpoint that is enabled only for one stop. The catchpoint is
5085 automatically deleted after the first time the event is caught.
5089 Use the @code{info break} command to list the current catchpoints.
5093 @subsection Deleting Breakpoints
5095 @cindex clearing breakpoints, watchpoints, catchpoints
5096 @cindex deleting breakpoints, watchpoints, catchpoints
5097 It is often necessary to eliminate a breakpoint, watchpoint, or
5098 catchpoint once it has done its job and you no longer want your program
5099 to stop there. This is called @dfn{deleting} the breakpoint. A
5100 breakpoint that has been deleted no longer exists; it is forgotten.
5102 With the @code{clear} command you can delete breakpoints according to
5103 where they are in your program. With the @code{delete} command you can
5104 delete individual breakpoints, watchpoints, or catchpoints by specifying
5105 their breakpoint numbers.
5107 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
5108 automatically ignores breakpoints on the first instruction to be executed
5109 when you continue execution without changing the execution address.
5114 Delete any breakpoints at the next instruction to be executed in the
5115 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
5116 the innermost frame is selected, this is a good way to delete a
5117 breakpoint where your program just stopped.
5119 @item clear @var{location}
5120 Delete any breakpoints set at the specified @var{location}.
5121 @xref{Specify Location}, for the various forms of @var{location}; the
5122 most useful ones are listed below:
5125 @item clear @var{function}
5126 @itemx clear @var{filename}:@var{function}
5127 Delete any breakpoints set at entry to the named @var{function}.
5129 @item clear @var{linenum}
5130 @itemx clear @var{filename}:@var{linenum}
5131 Delete any breakpoints set at or within the code of the specified
5132 @var{linenum} of the specified @var{filename}.
5135 @cindex delete breakpoints
5137 @kindex d @r{(@code{delete})}
5138 @item delete @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5139 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
5140 list specified as argument. If no argument is specified, delete all
5141 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
5142 confirm off}). You can abbreviate this command as @code{d}.
5146 @subsection Disabling Breakpoints
5148 @cindex enable/disable a breakpoint
5149 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
5150 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
5151 it had been deleted, but remembers the information on the breakpoint so
5152 that you can @dfn{enable} it again later.
5154 You disable and enable breakpoints, watchpoints, and catchpoints with
5155 the @code{enable} and @code{disable} commands, optionally specifying
5156 one or more breakpoint numbers as arguments. Use @code{info break} to
5157 print a list of all breakpoints, watchpoints, and catchpoints if you
5158 do not know which numbers to use.
5160 Disabling and enabling a breakpoint that has multiple locations
5161 affects all of its locations.
5163 A breakpoint, watchpoint, or catchpoint can have any of several
5164 different states of enablement:
5168 Enabled. The breakpoint stops your program. A breakpoint set
5169 with the @code{break} command starts out in this state.
5171 Disabled. The breakpoint has no effect on your program.
5173 Enabled once. The breakpoint stops your program, but then becomes
5176 Enabled for a count. The breakpoint stops your program for the next
5177 N times, then becomes disabled.
5179 Enabled for deletion. The breakpoint stops your program, but
5180 immediately after it does so it is deleted permanently. A breakpoint
5181 set with the @code{tbreak} command starts out in this state.
5184 You can use the following commands to enable or disable breakpoints,
5185 watchpoints, and catchpoints:
5189 @kindex dis @r{(@code{disable})}
5190 @item disable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5191 Disable the specified breakpoints---or all breakpoints, if none are
5192 listed. A disabled breakpoint has no effect but is not forgotten. All
5193 options such as ignore-counts, conditions and commands are remembered in
5194 case the breakpoint is enabled again later. You may abbreviate
5195 @code{disable} as @code{dis}.
5198 @item enable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5199 Enable the specified breakpoints (or all defined breakpoints). They
5200 become effective once again in stopping your program.
5202 @item enable @r{[}breakpoints@r{]} once @var{list}@dots{}
5203 Enable the specified breakpoints temporarily. @value{GDBN} disables any
5204 of these breakpoints immediately after stopping your program.
5206 @item enable @r{[}breakpoints@r{]} count @var{count} @var{list}@dots{}
5207 Enable the specified breakpoints temporarily. @value{GDBN} records
5208 @var{count} with each of the specified breakpoints, and decrements a
5209 breakpoint's count when it is hit. When any count reaches 0,
5210 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
5211 count (@pxref{Conditions, ,Break Conditions}), that will be
5212 decremented to 0 before @var{count} is affected.
5214 @item enable @r{[}breakpoints@r{]} delete @var{list}@dots{}
5215 Enable the specified breakpoints to work once, then die. @value{GDBN}
5216 deletes any of these breakpoints as soon as your program stops there.
5217 Breakpoints set by the @code{tbreak} command start out in this state.
5220 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
5221 @c confusing: tbreak is also initially enabled.
5222 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
5223 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
5224 subsequently, they become disabled or enabled only when you use one of
5225 the commands above. (The command @code{until} can set and delete a
5226 breakpoint of its own, but it does not change the state of your other
5227 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
5231 @subsection Break Conditions
5232 @cindex conditional breakpoints
5233 @cindex breakpoint conditions
5235 @c FIXME what is scope of break condition expr? Context where wanted?
5236 @c in particular for a watchpoint?
5237 The simplest sort of breakpoint breaks every time your program reaches a
5238 specified place. You can also specify a @dfn{condition} for a
5239 breakpoint. A condition is just a Boolean expression in your
5240 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
5241 a condition evaluates the expression each time your program reaches it,
5242 and your program stops only if the condition is @emph{true}.
5244 This is the converse of using assertions for program validation; in that
5245 situation, you want to stop when the assertion is violated---that is,
5246 when the condition is false. In C, if you want to test an assertion expressed
5247 by the condition @var{assert}, you should set the condition
5248 @samp{! @var{assert}} on the appropriate breakpoint.
5250 Conditions are also accepted for watchpoints; you may not need them,
5251 since a watchpoint is inspecting the value of an expression anyhow---but
5252 it might be simpler, say, to just set a watchpoint on a variable name,
5253 and specify a condition that tests whether the new value is an interesting
5256 Break conditions can have side effects, and may even call functions in
5257 your program. This can be useful, for example, to activate functions
5258 that log program progress, or to use your own print functions to
5259 format special data structures. The effects are completely predictable
5260 unless there is another enabled breakpoint at the same address. (In
5261 that case, @value{GDBN} might see the other breakpoint first and stop your
5262 program without checking the condition of this one.) Note that
5263 breakpoint commands are usually more convenient and flexible than break
5265 purpose of performing side effects when a breakpoint is reached
5266 (@pxref{Break Commands, ,Breakpoint Command Lists}).
5268 Breakpoint conditions can also be evaluated on the target's side if
5269 the target supports it. Instead of evaluating the conditions locally,
5270 @value{GDBN} encodes the expression into an agent expression
5271 (@pxref{Agent Expressions}) suitable for execution on the target,
5272 independently of @value{GDBN}. Global variables become raw memory
5273 locations, locals become stack accesses, and so forth.
5275 In this case, @value{GDBN} will only be notified of a breakpoint trigger
5276 when its condition evaluates to true. This mechanism may provide faster
5277 response times depending on the performance characteristics of the target
5278 since it does not need to keep @value{GDBN} informed about
5279 every breakpoint trigger, even those with false conditions.
5281 Break conditions can be specified when a breakpoint is set, by using
5282 @samp{if} in the arguments to the @code{break} command. @xref{Set
5283 Breaks, ,Setting Breakpoints}. They can also be changed at any time
5284 with the @code{condition} command.
5286 You can also use the @code{if} keyword with the @code{watch} command.
5287 The @code{catch} command does not recognize the @code{if} keyword;
5288 @code{condition} is the only way to impose a further condition on a
5293 @item condition @var{bnum} @var{expression}
5294 Specify @var{expression} as the break condition for breakpoint,
5295 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
5296 breakpoint @var{bnum} stops your program only if the value of
5297 @var{expression} is true (nonzero, in C). When you use
5298 @code{condition}, @value{GDBN} checks @var{expression} immediately for
5299 syntactic correctness, and to determine whether symbols in it have
5300 referents in the context of your breakpoint. If @var{expression} uses
5301 symbols not referenced in the context of the breakpoint, @value{GDBN}
5302 prints an error message:
5305 No symbol "foo" in current context.
5310 not actually evaluate @var{expression} at the time the @code{condition}
5311 command (or a command that sets a breakpoint with a condition, like
5312 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
5314 @item condition @var{bnum}
5315 Remove the condition from breakpoint number @var{bnum}. It becomes
5316 an ordinary unconditional breakpoint.
5319 @cindex ignore count (of breakpoint)
5320 A special case of a breakpoint condition is to stop only when the
5321 breakpoint has been reached a certain number of times. This is so
5322 useful that there is a special way to do it, using the @dfn{ignore
5323 count} of the breakpoint. Every breakpoint has an ignore count, which
5324 is an integer. Most of the time, the ignore count is zero, and
5325 therefore has no effect. But if your program reaches a breakpoint whose
5326 ignore count is positive, then instead of stopping, it just decrements
5327 the ignore count by one and continues. As a result, if the ignore count
5328 value is @var{n}, the breakpoint does not stop the next @var{n} times
5329 your program reaches it.
5333 @item ignore @var{bnum} @var{count}
5334 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
5335 The next @var{count} times the breakpoint is reached, your program's
5336 execution does not stop; other than to decrement the ignore count, @value{GDBN}
5339 To make the breakpoint stop the next time it is reached, specify
5342 When you use @code{continue} to resume execution of your program from a
5343 breakpoint, you can specify an ignore count directly as an argument to
5344 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
5345 Stepping,,Continuing and Stepping}.
5347 If a breakpoint has a positive ignore count and a condition, the
5348 condition is not checked. Once the ignore count reaches zero,
5349 @value{GDBN} resumes checking the condition.
5351 You could achieve the effect of the ignore count with a condition such
5352 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
5353 is decremented each time. @xref{Convenience Vars, ,Convenience
5357 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
5360 @node Break Commands
5361 @subsection Breakpoint Command Lists
5363 @cindex breakpoint commands
5364 You can give any breakpoint (or watchpoint or catchpoint) a series of
5365 commands to execute when your program stops due to that breakpoint. For
5366 example, you might want to print the values of certain expressions, or
5367 enable other breakpoints.
5371 @kindex end@r{ (breakpoint commands)}
5372 @item commands @r{[}@var{list}@dots{}@r{]}
5373 @itemx @dots{} @var{command-list} @dots{}
5375 Specify a list of commands for the given breakpoints. The commands
5376 themselves appear on the following lines. Type a line containing just
5377 @code{end} to terminate the commands.
5379 To remove all commands from a breakpoint, type @code{commands} and
5380 follow it immediately with @code{end}; that is, give no commands.
5382 With no argument, @code{commands} refers to the last breakpoint,
5383 watchpoint, or catchpoint set (not to the breakpoint most recently
5384 encountered). If the most recent breakpoints were set with a single
5385 command, then the @code{commands} will apply to all the breakpoints
5386 set by that command. This applies to breakpoints set by
5387 @code{rbreak}, and also applies when a single @code{break} command
5388 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
5392 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
5393 disabled within a @var{command-list}.
5395 You can use breakpoint commands to start your program up again. Simply
5396 use the @code{continue} command, or @code{step}, or any other command
5397 that resumes execution.
5399 Any other commands in the command list, after a command that resumes
5400 execution, are ignored. This is because any time you resume execution
5401 (even with a simple @code{next} or @code{step}), you may encounter
5402 another breakpoint---which could have its own command list, leading to
5403 ambiguities about which list to execute.
5406 If the first command you specify in a command list is @code{silent}, the
5407 usual message about stopping at a breakpoint is not printed. This may
5408 be desirable for breakpoints that are to print a specific message and
5409 then continue. If none of the remaining commands print anything, you
5410 see no sign that the breakpoint was reached. @code{silent} is
5411 meaningful only at the beginning of a breakpoint command list.
5413 The commands @code{echo}, @code{output}, and @code{printf} allow you to
5414 print precisely controlled output, and are often useful in silent
5415 breakpoints. @xref{Output, ,Commands for Controlled Output}.
5417 For example, here is how you could use breakpoint commands to print the
5418 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
5424 printf "x is %d\n",x
5429 One application for breakpoint commands is to compensate for one bug so
5430 you can test for another. Put a breakpoint just after the erroneous line
5431 of code, give it a condition to detect the case in which something
5432 erroneous has been done, and give it commands to assign correct values
5433 to any variables that need them. End with the @code{continue} command
5434 so that your program does not stop, and start with the @code{silent}
5435 command so that no output is produced. Here is an example:
5446 @node Dynamic Printf
5447 @subsection Dynamic Printf
5449 @cindex dynamic printf
5451 The dynamic printf command @code{dprintf} combines a breakpoint with
5452 formatted printing of your program's data to give you the effect of
5453 inserting @code{printf} calls into your program on-the-fly, without
5454 having to recompile it.
5456 In its most basic form, the output goes to the GDB console. However,
5457 you can set the variable @code{dprintf-style} for alternate handling.
5458 For instance, you can ask to format the output by calling your
5459 program's @code{printf} function. This has the advantage that the
5460 characters go to the program's output device, so they can recorded in
5461 redirects to files and so forth.
5463 If you are doing remote debugging with a stub or agent, you can also
5464 ask to have the printf handled by the remote agent. In addition to
5465 ensuring that the output goes to the remote program's device along
5466 with any other output the program might produce, you can also ask that
5467 the dprintf remain active even after disconnecting from the remote
5468 target. Using the stub/agent is also more efficient, as it can do
5469 everything without needing to communicate with @value{GDBN}.
5473 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
5474 Whenever execution reaches @var{location}, print the values of one or
5475 more @var{expressions} under the control of the string @var{template}.
5476 To print several values, separate them with commas.
5478 @item set dprintf-style @var{style}
5479 Set the dprintf output to be handled in one of several different
5480 styles enumerated below. A change of style affects all existing
5481 dynamic printfs immediately. (If you need individual control over the
5482 print commands, simply define normal breakpoints with
5483 explicitly-supplied command lists.)
5487 @kindex dprintf-style gdb
5488 Handle the output using the @value{GDBN} @code{printf} command.
5491 @kindex dprintf-style call
5492 Handle the output by calling a function in your program (normally
5496 @kindex dprintf-style agent
5497 Have the remote debugging agent (such as @code{gdbserver}) handle
5498 the output itself. This style is only available for agents that
5499 support running commands on the target.
5502 @item set dprintf-function @var{function}
5503 Set the function to call if the dprintf style is @code{call}. By
5504 default its value is @code{printf}. You may set it to any expression.
5505 that @value{GDBN} can evaluate to a function, as per the @code{call}
5508 @item set dprintf-channel @var{channel}
5509 Set a ``channel'' for dprintf. If set to a non-empty value,
5510 @value{GDBN} will evaluate it as an expression and pass the result as
5511 a first argument to the @code{dprintf-function}, in the manner of
5512 @code{fprintf} and similar functions. Otherwise, the dprintf format
5513 string will be the first argument, in the manner of @code{printf}.
5515 As an example, if you wanted @code{dprintf} output to go to a logfile
5516 that is a standard I/O stream assigned to the variable @code{mylog},
5517 you could do the following:
5520 (@value{GDBP}) set dprintf-style call
5521 (@value{GDBP}) set dprintf-function fprintf
5522 (@value{GDBP}) set dprintf-channel mylog
5523 (@value{GDBP}) dprintf 25,"at line 25, glob=%d\n",glob
5524 Dprintf 1 at 0x123456: file main.c, line 25.
5525 (@value{GDBP}) info break
5526 1 dprintf keep y 0x00123456 in main at main.c:25
5527 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
5532 Note that the @code{info break} displays the dynamic printf commands
5533 as normal breakpoint commands; you can thus easily see the effect of
5534 the variable settings.
5536 @item set disconnected-dprintf on
5537 @itemx set disconnected-dprintf off
5538 @kindex set disconnected-dprintf
5539 Choose whether @code{dprintf} commands should continue to run if
5540 @value{GDBN} has disconnected from the target. This only applies
5541 if the @code{dprintf-style} is @code{agent}.
5543 @item show disconnected-dprintf off
5544 @kindex show disconnected-dprintf
5545 Show the current choice for disconnected @code{dprintf}.
5549 @value{GDBN} does not check the validity of function and channel,
5550 relying on you to supply values that are meaningful for the contexts
5551 in which they are being used. For instance, the function and channel
5552 may be the values of local variables, but if that is the case, then
5553 all enabled dynamic prints must be at locations within the scope of
5554 those locals. If evaluation fails, @value{GDBN} will report an error.
5556 @node Save Breakpoints
5557 @subsection How to save breakpoints to a file
5559 To save breakpoint definitions to a file use the @w{@code{save
5560 breakpoints}} command.
5563 @kindex save breakpoints
5564 @cindex save breakpoints to a file for future sessions
5565 @item save breakpoints [@var{filename}]
5566 This command saves all current breakpoint definitions together with
5567 their commands and ignore counts, into a file @file{@var{filename}}
5568 suitable for use in a later debugging session. This includes all
5569 types of breakpoints (breakpoints, watchpoints, catchpoints,
5570 tracepoints). To read the saved breakpoint definitions, use the
5571 @code{source} command (@pxref{Command Files}). Note that watchpoints
5572 with expressions involving local variables may fail to be recreated
5573 because it may not be possible to access the context where the
5574 watchpoint is valid anymore. Because the saved breakpoint definitions
5575 are simply a sequence of @value{GDBN} commands that recreate the
5576 breakpoints, you can edit the file in your favorite editing program,
5577 and remove the breakpoint definitions you're not interested in, or
5578 that can no longer be recreated.
5581 @node Static Probe Points
5582 @subsection Static Probe Points
5584 @cindex static probe point, SystemTap
5585 @cindex static probe point, DTrace
5586 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
5587 for Statically Defined Tracing, and the probes are designed to have a tiny
5588 runtime code and data footprint, and no dynamic relocations.
5590 Currently, the following types of probes are supported on
5591 ELF-compatible systems:
5595 @item @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
5596 @acronym{SDT} probes@footnote{See
5597 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
5598 for more information on how to add @code{SystemTap} @acronym{SDT}
5599 probes in your applications.}. @code{SystemTap} probes are usable
5600 from assembly, C and C@t{++} languages@footnote{See
5601 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
5602 for a good reference on how the @acronym{SDT} probes are implemented.}.
5604 @item @code{DTrace} (@uref{http://oss.oracle.com/projects/DTrace})
5605 @acronym{USDT} probes. @code{DTrace} probes are usable from C and
5609 @cindex semaphores on static probe points
5610 Some @code{SystemTap} probes have an associated semaphore variable;
5611 for instance, this happens automatically if you defined your probe
5612 using a DTrace-style @file{.d} file. If your probe has a semaphore,
5613 @value{GDBN} will automatically enable it when you specify a
5614 breakpoint using the @samp{-probe-stap} notation. But, if you put a
5615 breakpoint at a probe's location by some other method (e.g.,
5616 @code{break file:line}), then @value{GDBN} will not automatically set
5617 the semaphore. @code{DTrace} probes do not support semaphores.
5619 You can examine the available static static probes using @code{info
5620 probes}, with optional arguments:
5624 @item info probes @r{[}@var{type}@r{]} @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5625 If given, @var{type} is either @code{stap} for listing
5626 @code{SystemTap} probes or @code{dtrace} for listing @code{DTrace}
5627 probes. If omitted all probes are listed regardless of their types.
5629 If given, @var{provider} is a regular expression used to match against provider
5630 names when selecting which probes to list. If omitted, probes by all
5631 probes from all providers are listed.
5633 If given, @var{name} is a regular expression to match against probe names
5634 when selecting which probes to list. If omitted, probe names are not
5635 considered when deciding whether to display them.
5637 If given, @var{objfile} is a regular expression used to select which
5638 object files (executable or shared libraries) to examine. If not
5639 given, all object files are considered.
5641 @item info probes all
5642 List the available static probes, from all types.
5645 @cindex enabling and disabling probes
5646 Some probe points can be enabled and/or disabled. The effect of
5647 enabling or disabling a probe depends on the type of probe being
5648 handled. Some @code{DTrace} probes can be enabled or
5649 disabled, but @code{SystemTap} probes cannot be disabled.
5651 You can enable (or disable) one or more probes using the following
5652 commands, with optional arguments:
5655 @kindex enable probes
5656 @item enable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5657 If given, @var{provider} is a regular expression used to match against
5658 provider names when selecting which probes to enable. If omitted,
5659 all probes from all providers are enabled.
5661 If given, @var{name} is a regular expression to match against probe
5662 names when selecting which probes to enable. If omitted, probe names
5663 are not considered when deciding whether to enable them.
5665 If given, @var{objfile} is a regular expression used to select which
5666 object files (executable or shared libraries) to examine. If not
5667 given, all object files are considered.
5669 @kindex disable probes
5670 @item disable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5671 See the @code{enable probes} command above for a description of the
5672 optional arguments accepted by this command.
5675 @vindex $_probe_arg@r{, convenience variable}
5676 A probe may specify up to twelve arguments. These are available at the
5677 point at which the probe is defined---that is, when the current PC is
5678 at the probe's location. The arguments are available using the
5679 convenience variables (@pxref{Convenience Vars})
5680 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. In @code{SystemTap}
5681 probes each probe argument is an integer of the appropriate size;
5682 types are not preserved. In @code{DTrace} probes types are preserved
5683 provided that they are recognized as such by @value{GDBN}; otherwise
5684 the value of the probe argument will be a long integer. The
5685 convenience variable @code{$_probe_argc} holds the number of arguments
5686 at the current probe point.
5688 These variables are always available, but attempts to access them at
5689 any location other than a probe point will cause @value{GDBN} to give
5693 @c @ifclear BARETARGET
5694 @node Error in Breakpoints
5695 @subsection ``Cannot insert breakpoints''
5697 If you request too many active hardware-assisted breakpoints and
5698 watchpoints, you will see this error message:
5700 @c FIXME: the precise wording of this message may change; the relevant
5701 @c source change is not committed yet (Sep 3, 1999).
5703 Stopped; cannot insert breakpoints.
5704 You may have requested too many hardware breakpoints and watchpoints.
5708 This message is printed when you attempt to resume the program, since
5709 only then @value{GDBN} knows exactly how many hardware breakpoints and
5710 watchpoints it needs to insert.
5712 When this message is printed, you need to disable or remove some of the
5713 hardware-assisted breakpoints and watchpoints, and then continue.
5715 @node Breakpoint-related Warnings
5716 @subsection ``Breakpoint address adjusted...''
5717 @cindex breakpoint address adjusted
5719 Some processor architectures place constraints on the addresses at
5720 which breakpoints may be placed. For architectures thus constrained,
5721 @value{GDBN} will attempt to adjust the breakpoint's address to comply
5722 with the constraints dictated by the architecture.
5724 One example of such an architecture is the Fujitsu FR-V. The FR-V is
5725 a VLIW architecture in which a number of RISC-like instructions may be
5726 bundled together for parallel execution. The FR-V architecture
5727 constrains the location of a breakpoint instruction within such a
5728 bundle to the instruction with the lowest address. @value{GDBN}
5729 honors this constraint by adjusting a breakpoint's address to the
5730 first in the bundle.
5732 It is not uncommon for optimized code to have bundles which contain
5733 instructions from different source statements, thus it may happen that
5734 a breakpoint's address will be adjusted from one source statement to
5735 another. Since this adjustment may significantly alter @value{GDBN}'s
5736 breakpoint related behavior from what the user expects, a warning is
5737 printed when the breakpoint is first set and also when the breakpoint
5740 A warning like the one below is printed when setting a breakpoint
5741 that's been subject to address adjustment:
5744 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
5747 Such warnings are printed both for user settable and @value{GDBN}'s
5748 internal breakpoints. If you see one of these warnings, you should
5749 verify that a breakpoint set at the adjusted address will have the
5750 desired affect. If not, the breakpoint in question may be removed and
5751 other breakpoints may be set which will have the desired behavior.
5752 E.g., it may be sufficient to place the breakpoint at a later
5753 instruction. A conditional breakpoint may also be useful in some
5754 cases to prevent the breakpoint from triggering too often.
5756 @value{GDBN} will also issue a warning when stopping at one of these
5757 adjusted breakpoints:
5760 warning: Breakpoint 1 address previously adjusted from 0x00010414
5764 When this warning is encountered, it may be too late to take remedial
5765 action except in cases where the breakpoint is hit earlier or more
5766 frequently than expected.
5768 @node Continuing and Stepping
5769 @section Continuing and Stepping
5773 @cindex resuming execution
5774 @dfn{Continuing} means resuming program execution until your program
5775 completes normally. In contrast, @dfn{stepping} means executing just
5776 one more ``step'' of your program, where ``step'' may mean either one
5777 line of source code, or one machine instruction (depending on what
5778 particular command you use). Either when continuing or when stepping,
5779 your program may stop even sooner, due to a breakpoint or a signal. (If
5780 it stops due to a signal, you may want to use @code{handle}, or use
5781 @samp{signal 0} to resume execution (@pxref{Signals, ,Signals}),
5782 or you may step into the signal's handler (@pxref{stepping and signal
5787 @kindex c @r{(@code{continue})}
5788 @kindex fg @r{(resume foreground execution)}
5789 @item continue @r{[}@var{ignore-count}@r{]}
5790 @itemx c @r{[}@var{ignore-count}@r{]}
5791 @itemx fg @r{[}@var{ignore-count}@r{]}
5792 Resume program execution, at the address where your program last stopped;
5793 any breakpoints set at that address are bypassed. The optional argument
5794 @var{ignore-count} allows you to specify a further number of times to
5795 ignore a breakpoint at this location; its effect is like that of
5796 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
5798 The argument @var{ignore-count} is meaningful only when your program
5799 stopped due to a breakpoint. At other times, the argument to
5800 @code{continue} is ignored.
5802 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5803 debugged program is deemed to be the foreground program) are provided
5804 purely for convenience, and have exactly the same behavior as
5808 To resume execution at a different place, you can use @code{return}
5809 (@pxref{Returning, ,Returning from a Function}) to go back to the
5810 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5811 Different Address}) to go to an arbitrary location in your program.
5813 A typical technique for using stepping is to set a breakpoint
5814 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5815 beginning of the function or the section of your program where a problem
5816 is believed to lie, run your program until it stops at that breakpoint,
5817 and then step through the suspect area, examining the variables that are
5818 interesting, until you see the problem happen.
5822 @kindex s @r{(@code{step})}
5824 Continue running your program until control reaches a different source
5825 line, then stop it and return control to @value{GDBN}. This command is
5826 abbreviated @code{s}.
5829 @c "without debugging information" is imprecise; actually "without line
5830 @c numbers in the debugging information". (gcc -g1 has debugging info but
5831 @c not line numbers). But it seems complex to try to make that
5832 @c distinction here.
5833 @emph{Warning:} If you use the @code{step} command while control is
5834 within a function that was compiled without debugging information,
5835 execution proceeds until control reaches a function that does have
5836 debugging information. Likewise, it will not step into a function which
5837 is compiled without debugging information. To step through functions
5838 without debugging information, use the @code{stepi} command, described
5842 The @code{step} command only stops at the first instruction of a source
5843 line. This prevents the multiple stops that could otherwise occur in
5844 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5845 to stop if a function that has debugging information is called within
5846 the line. In other words, @code{step} @emph{steps inside} any functions
5847 called within the line.
5849 Also, the @code{step} command only enters a function if there is line
5850 number information for the function. Otherwise it acts like the
5851 @code{next} command. This avoids problems when using @code{cc -gl}
5852 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5853 was any debugging information about the routine.
5855 @item step @var{count}
5856 Continue running as in @code{step}, but do so @var{count} times. If a
5857 breakpoint is reached, or a signal not related to stepping occurs before
5858 @var{count} steps, stepping stops right away.
5861 @kindex n @r{(@code{next})}
5862 @item next @r{[}@var{count}@r{]}
5863 Continue to the next source line in the current (innermost) stack frame.
5864 This is similar to @code{step}, but function calls that appear within
5865 the line of code are executed without stopping. Execution stops when
5866 control reaches a different line of code at the original stack level
5867 that was executing when you gave the @code{next} command. This command
5868 is abbreviated @code{n}.
5870 An argument @var{count} is a repeat count, as for @code{step}.
5873 @c FIX ME!! Do we delete this, or is there a way it fits in with
5874 @c the following paragraph? --- Vctoria
5876 @c @code{next} within a function that lacks debugging information acts like
5877 @c @code{step}, but any function calls appearing within the code of the
5878 @c function are executed without stopping.
5880 The @code{next} command only stops at the first instruction of a
5881 source line. This prevents multiple stops that could otherwise occur in
5882 @code{switch} statements, @code{for} loops, etc.
5884 @kindex set step-mode
5886 @cindex functions without line info, and stepping
5887 @cindex stepping into functions with no line info
5888 @itemx set step-mode on
5889 The @code{set step-mode on} command causes the @code{step} command to
5890 stop at the first instruction of a function which contains no debug line
5891 information rather than stepping over it.
5893 This is useful in cases where you may be interested in inspecting the
5894 machine instructions of a function which has no symbolic info and do not
5895 want @value{GDBN} to automatically skip over this function.
5897 @item set step-mode off
5898 Causes the @code{step} command to step over any functions which contains no
5899 debug information. This is the default.
5901 @item show step-mode
5902 Show whether @value{GDBN} will stop in or step over functions without
5903 source line debug information.
5906 @kindex fin @r{(@code{finish})}
5908 Continue running until just after function in the selected stack frame
5909 returns. Print the returned value (if any). This command can be
5910 abbreviated as @code{fin}.
5912 Contrast this with the @code{return} command (@pxref{Returning,
5913 ,Returning from a Function}).
5915 @kindex set print finish
5916 @kindex show print finish
5917 @item set print finish @r{[}on|off@r{]}
5918 @itemx show print finish
5919 By default the @code{finish} command will show the value that is
5920 returned by the function. This can be disabled using @code{set print
5921 finish off}. When disabled, the value is still entered into the value
5922 history (@pxref{Value History}), but not displayed.
5925 @kindex u @r{(@code{until})}
5926 @cindex run until specified location
5929 Continue running until a source line past the current line, in the
5930 current stack frame, is reached. This command is used to avoid single
5931 stepping through a loop more than once. It is like the @code{next}
5932 command, except that when @code{until} encounters a jump, it
5933 automatically continues execution until the program counter is greater
5934 than the address of the jump.
5936 This means that when you reach the end of a loop after single stepping
5937 though it, @code{until} makes your program continue execution until it
5938 exits the loop. In contrast, a @code{next} command at the end of a loop
5939 simply steps back to the beginning of the loop, which forces you to step
5940 through the next iteration.
5942 @code{until} always stops your program if it attempts to exit the current
5945 @code{until} may produce somewhat counterintuitive results if the order
5946 of machine code does not match the order of the source lines. For
5947 example, in the following excerpt from a debugging session, the @code{f}
5948 (@code{frame}) command shows that execution is stopped at line
5949 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5953 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5955 (@value{GDBP}) until
5956 195 for ( ; argc > 0; NEXTARG) @{
5959 This happened because, for execution efficiency, the compiler had
5960 generated code for the loop closure test at the end, rather than the
5961 start, of the loop---even though the test in a C @code{for}-loop is
5962 written before the body of the loop. The @code{until} command appeared
5963 to step back to the beginning of the loop when it advanced to this
5964 expression; however, it has not really gone to an earlier
5965 statement---not in terms of the actual machine code.
5967 @code{until} with no argument works by means of single
5968 instruction stepping, and hence is slower than @code{until} with an
5971 @item until @var{location}
5972 @itemx u @var{location}
5973 Continue running your program until either the specified @var{location} is
5974 reached, or the current stack frame returns. The location is any of
5975 the forms described in @ref{Specify Location}.
5976 This form of the command uses temporary breakpoints, and
5977 hence is quicker than @code{until} without an argument. The specified
5978 location is actually reached only if it is in the current frame. This
5979 implies that @code{until} can be used to skip over recursive function
5980 invocations. For instance in the code below, if the current location is
5981 line @code{96}, issuing @code{until 99} will execute the program up to
5982 line @code{99} in the same invocation of factorial, i.e., after the inner
5983 invocations have returned.
5986 94 int factorial (int value)
5988 96 if (value > 1) @{
5989 97 value *= factorial (value - 1);
5996 @kindex advance @var{location}
5997 @item advance @var{location}
5998 Continue running the program up to the given @var{location}. An argument is
5999 required, which should be of one of the forms described in
6000 @ref{Specify Location}.
6001 Execution will also stop upon exit from the current stack
6002 frame. This command is similar to @code{until}, but @code{advance} will
6003 not skip over recursive function calls, and the target location doesn't
6004 have to be in the same frame as the current one.
6008 @kindex si @r{(@code{stepi})}
6010 @itemx stepi @var{arg}
6012 Execute one machine instruction, then stop and return to the debugger.
6014 It is often useful to do @samp{display/i $pc} when stepping by machine
6015 instructions. This makes @value{GDBN} automatically display the next
6016 instruction to be executed, each time your program stops. @xref{Auto
6017 Display,, Automatic Display}.
6019 An argument is a repeat count, as in @code{step}.
6023 @kindex ni @r{(@code{nexti})}
6025 @itemx nexti @var{arg}
6027 Execute one machine instruction, but if it is a function call,
6028 proceed until the function returns.
6030 An argument is a repeat count, as in @code{next}.
6034 @anchor{range stepping}
6035 @cindex range stepping
6036 @cindex target-assisted range stepping
6037 By default, and if available, @value{GDBN} makes use of
6038 target-assisted @dfn{range stepping}. In other words, whenever you
6039 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
6040 tells the target to step the corresponding range of instruction
6041 addresses instead of issuing multiple single-steps. This speeds up
6042 line stepping, particularly for remote targets. Ideally, there should
6043 be no reason you would want to turn range stepping off. However, it's
6044 possible that a bug in the debug info, a bug in the remote stub (for
6045 remote targets), or even a bug in @value{GDBN} could make line
6046 stepping behave incorrectly when target-assisted range stepping is
6047 enabled. You can use the following command to turn off range stepping
6051 @kindex set range-stepping
6052 @kindex show range-stepping
6053 @item set range-stepping
6054 @itemx show range-stepping
6055 Control whether range stepping is enabled.
6057 If @code{on}, and the target supports it, @value{GDBN} tells the
6058 target to step a range of addresses itself, instead of issuing
6059 multiple single-steps. If @code{off}, @value{GDBN} always issues
6060 single-steps, even if range stepping is supported by the target. The
6061 default is @code{on}.
6065 @node Skipping Over Functions and Files
6066 @section Skipping Over Functions and Files
6067 @cindex skipping over functions and files
6069 The program you are debugging may contain some functions which are
6070 uninteresting to debug. The @code{skip} command lets you tell @value{GDBN} to
6071 skip a function, all functions in a file or a particular function in
6072 a particular file when stepping.
6074 For example, consider the following C function:
6085 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
6086 are not interested in stepping through @code{boring}. If you run @code{step}
6087 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
6088 step over both @code{foo} and @code{boring}!
6090 One solution is to @code{step} into @code{boring} and use the @code{finish}
6091 command to immediately exit it. But this can become tedious if @code{boring}
6092 is called from many places.
6094 A more flexible solution is to execute @kbd{skip boring}. This instructs
6095 @value{GDBN} never to step into @code{boring}. Now when you execute
6096 @code{step} at line 103, you'll step over @code{boring} and directly into
6099 Functions may be skipped by providing either a function name, linespec
6100 (@pxref{Specify Location}), regular expression that matches the function's
6101 name, file name or a @code{glob}-style pattern that matches the file name.
6103 On Posix systems the form of the regular expression is
6104 ``Extended Regular Expressions''. See for example @samp{man 7 regex}
6105 on @sc{gnu}/Linux systems. On non-Posix systems the form of the regular
6106 expression is whatever is provided by the @code{regcomp} function of
6107 the underlying system.
6108 See for example @samp{man 7 glob} on @sc{gnu}/Linux systems for a
6109 description of @code{glob}-style patterns.
6113 @item skip @r{[}@var{options}@r{]}
6114 The basic form of the @code{skip} command takes zero or more options
6115 that specify what to skip.
6116 The @var{options} argument is any useful combination of the following:
6119 @item -file @var{file}
6120 @itemx -fi @var{file}
6121 Functions in @var{file} will be skipped over when stepping.
6123 @item -gfile @var{file-glob-pattern}
6124 @itemx -gfi @var{file-glob-pattern}
6125 @cindex skipping over files via glob-style patterns
6126 Functions in files matching @var{file-glob-pattern} will be skipped
6130 (@value{GDBP}) skip -gfi utils/*.c
6133 @item -function @var{linespec}
6134 @itemx -fu @var{linespec}
6135 Functions named by @var{linespec} or the function containing the line
6136 named by @var{linespec} will be skipped over when stepping.
6137 @xref{Specify Location}.
6139 @item -rfunction @var{regexp}
6140 @itemx -rfu @var{regexp}
6141 @cindex skipping over functions via regular expressions
6142 Functions whose name matches @var{regexp} will be skipped over when stepping.
6144 This form is useful for complex function names.
6145 For example, there is generally no need to step into C@t{++} @code{std::string}
6146 constructors or destructors. Plus with C@t{++} templates it can be hard to
6147 write out the full name of the function, and often it doesn't matter what
6148 the template arguments are. Specifying the function to be skipped as a
6149 regular expression makes this easier.
6152 (@value{GDBP}) skip -rfu ^std::(allocator|basic_string)<.*>::~?\1 *\(
6155 If you want to skip every templated C@t{++} constructor and destructor
6156 in the @code{std} namespace you can do:
6159 (@value{GDBP}) skip -rfu ^std::([a-zA-z0-9_]+)<.*>::~?\1 *\(
6163 If no options are specified, the function you're currently debugging
6166 @kindex skip function
6167 @item skip function @r{[}@var{linespec}@r{]}
6168 After running this command, the function named by @var{linespec} or the
6169 function containing the line named by @var{linespec} will be skipped over when
6170 stepping. @xref{Specify Location}.
6172 If you do not specify @var{linespec}, the function you're currently debugging
6175 (If you have a function called @code{file} that you want to skip, use
6176 @kbd{skip function file}.)
6179 @item skip file @r{[}@var{filename}@r{]}
6180 After running this command, any function whose source lives in @var{filename}
6181 will be skipped over when stepping.
6184 (@value{GDBP}) skip file boring.c
6185 File boring.c will be skipped when stepping.
6188 If you do not specify @var{filename}, functions whose source lives in the file
6189 you're currently debugging will be skipped.
6192 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
6193 These are the commands for managing your list of skips:
6197 @item info skip @r{[}@var{range}@r{]}
6198 Print details about the specified skip(s). If @var{range} is not specified,
6199 print a table with details about all functions and files marked for skipping.
6200 @code{info skip} prints the following information about each skip:
6204 A number identifying this skip.
6205 @item Enabled or Disabled
6206 Enabled skips are marked with @samp{y}.
6207 Disabled skips are marked with @samp{n}.
6209 If the file name is a @samp{glob} pattern this is @samp{y}.
6210 Otherwise it is @samp{n}.
6212 The name or @samp{glob} pattern of the file to be skipped.
6213 If no file is specified this is @samp{<none>}.
6215 If the function name is a @samp{regular expression} this is @samp{y}.
6216 Otherwise it is @samp{n}.
6218 The name or regular expression of the function to skip.
6219 If no function is specified this is @samp{<none>}.
6223 @item skip delete @r{[}@var{range}@r{]}
6224 Delete the specified skip(s). If @var{range} is not specified, delete all
6228 @item skip enable @r{[}@var{range}@r{]}
6229 Enable the specified skip(s). If @var{range} is not specified, enable all
6232 @kindex skip disable
6233 @item skip disable @r{[}@var{range}@r{]}
6234 Disable the specified skip(s). If @var{range} is not specified, disable all
6237 @kindex set debug skip
6238 @item set debug skip @r{[}on|off@r{]}
6239 Set whether to print the debug output about skipping files and functions.
6241 @kindex show debug skip
6242 @item show debug skip
6243 Show whether the debug output about skipping files and functions is printed.
6251 A signal is an asynchronous event that can happen in a program. The
6252 operating system defines the possible kinds of signals, and gives each
6253 kind a name and a number. For example, in Unix @code{SIGINT} is the
6254 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
6255 @code{SIGSEGV} is the signal a program gets from referencing a place in
6256 memory far away from all the areas in use; @code{SIGALRM} occurs when
6257 the alarm clock timer goes off (which happens only if your program has
6258 requested an alarm).
6260 @cindex fatal signals
6261 Some signals, including @code{SIGALRM}, are a normal part of the
6262 functioning of your program. Others, such as @code{SIGSEGV}, indicate
6263 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
6264 program has not specified in advance some other way to handle the signal.
6265 @code{SIGINT} does not indicate an error in your program, but it is normally
6266 fatal so it can carry out the purpose of the interrupt: to kill the program.
6268 @value{GDBN} has the ability to detect any occurrence of a signal in your
6269 program. You can tell @value{GDBN} in advance what to do for each kind of
6272 @cindex handling signals
6273 Normally, @value{GDBN} is set up to let the non-erroneous signals like
6274 @code{SIGALRM} be silently passed to your program
6275 (so as not to interfere with their role in the program's functioning)
6276 but to stop your program immediately whenever an error signal happens.
6277 You can change these settings with the @code{handle} command.
6280 @kindex info signals
6284 Print a table of all the kinds of signals and how @value{GDBN} has been told to
6285 handle each one. You can use this to see the signal numbers of all
6286 the defined types of signals.
6288 @item info signals @var{sig}
6289 Similar, but print information only about the specified signal number.
6291 @code{info handle} is an alias for @code{info signals}.
6293 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
6294 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
6295 for details about this command.
6298 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
6299 Change the way @value{GDBN} handles signal @var{signal}. The @var{signal}
6300 can be the number of a signal or its name (with or without the
6301 @samp{SIG} at the beginning); a list of signal numbers of the form
6302 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
6303 known signals. Optional arguments @var{keywords}, described below,
6304 say what change to make.
6308 The keywords allowed by the @code{handle} command can be abbreviated.
6309 Their full names are:
6313 @value{GDBN} should not stop your program when this signal happens. It may
6314 still print a message telling you that the signal has come in.
6317 @value{GDBN} should stop your program when this signal happens. This implies
6318 the @code{print} keyword as well.
6321 @value{GDBN} should print a message when this signal happens.
6324 @value{GDBN} should not mention the occurrence of the signal at all. This
6325 implies the @code{nostop} keyword as well.
6329 @value{GDBN} should allow your program to see this signal; your program
6330 can handle the signal, or else it may terminate if the signal is fatal
6331 and not handled. @code{pass} and @code{noignore} are synonyms.
6335 @value{GDBN} should not allow your program to see this signal.
6336 @code{nopass} and @code{ignore} are synonyms.
6340 When a signal stops your program, the signal is not visible to the
6342 continue. Your program sees the signal then, if @code{pass} is in
6343 effect for the signal in question @emph{at that time}. In other words,
6344 after @value{GDBN} reports a signal, you can use the @code{handle}
6345 command with @code{pass} or @code{nopass} to control whether your
6346 program sees that signal when you continue.
6348 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
6349 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
6350 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
6353 You can also use the @code{signal} command to prevent your program from
6354 seeing a signal, or cause it to see a signal it normally would not see,
6355 or to give it any signal at any time. For example, if your program stopped
6356 due to some sort of memory reference error, you might store correct
6357 values into the erroneous variables and continue, hoping to see more
6358 execution; but your program would probably terminate immediately as
6359 a result of the fatal signal once it saw the signal. To prevent this,
6360 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
6363 @cindex stepping and signal handlers
6364 @anchor{stepping and signal handlers}
6366 @value{GDBN} optimizes for stepping the mainline code. If a signal
6367 that has @code{handle nostop} and @code{handle pass} set arrives while
6368 a stepping command (e.g., @code{stepi}, @code{step}, @code{next}) is
6369 in progress, @value{GDBN} lets the signal handler run and then resumes
6370 stepping the mainline code once the signal handler returns. In other
6371 words, @value{GDBN} steps over the signal handler. This prevents
6372 signals that you've specified as not interesting (with @code{handle
6373 nostop}) from changing the focus of debugging unexpectedly. Note that
6374 the signal handler itself may still hit a breakpoint, stop for another
6375 signal that has @code{handle stop} in effect, or for any other event
6376 that normally results in stopping the stepping command sooner. Also
6377 note that @value{GDBN} still informs you that the program received a
6378 signal if @code{handle print} is set.
6380 @anchor{stepping into signal handlers}
6382 If you set @code{handle pass} for a signal, and your program sets up a
6383 handler for it, then issuing a stepping command, such as @code{step}
6384 or @code{stepi}, when your program is stopped due to the signal will
6385 step @emph{into} the signal handler (if the target supports that).
6387 Likewise, if you use the @code{queue-signal} command to queue a signal
6388 to be delivered to the current thread when execution of the thread
6389 resumes (@pxref{Signaling, ,Giving your Program a Signal}), then a
6390 stepping command will step into the signal handler.
6392 Here's an example, using @code{stepi} to step to the first instruction
6393 of @code{SIGUSR1}'s handler:
6396 (@value{GDBP}) handle SIGUSR1
6397 Signal Stop Print Pass to program Description
6398 SIGUSR1 Yes Yes Yes User defined signal 1
6402 Program received signal SIGUSR1, User defined signal 1.
6403 main () sigusr1.c:28
6406 sigusr1_handler () at sigusr1.c:9
6410 The same, but using @code{queue-signal} instead of waiting for the
6411 program to receive the signal first:
6416 (@value{GDBP}) queue-signal SIGUSR1
6418 sigusr1_handler () at sigusr1.c:9
6423 @cindex extra signal information
6424 @anchor{extra signal information}
6426 On some targets, @value{GDBN} can inspect extra signal information
6427 associated with the intercepted signal, before it is actually
6428 delivered to the program being debugged. This information is exported
6429 by the convenience variable @code{$_siginfo}, and consists of data
6430 that is passed by the kernel to the signal handler at the time of the
6431 receipt of a signal. The data type of the information itself is
6432 target dependent. You can see the data type using the @code{ptype
6433 $_siginfo} command. On Unix systems, it typically corresponds to the
6434 standard @code{siginfo_t} type, as defined in the @file{signal.h}
6437 Here's an example, on a @sc{gnu}/Linux system, printing the stray
6438 referenced address that raised a segmentation fault.
6442 (@value{GDBP}) continue
6443 Program received signal SIGSEGV, Segmentation fault.
6444 0x0000000000400766 in main ()
6446 (@value{GDBP}) ptype $_siginfo
6453 struct @{...@} _kill;
6454 struct @{...@} _timer;
6456 struct @{...@} _sigchld;
6457 struct @{...@} _sigfault;
6458 struct @{...@} _sigpoll;
6461 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
6465 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
6466 $1 = (void *) 0x7ffff7ff7000
6470 Depending on target support, @code{$_siginfo} may also be writable.
6472 @cindex Intel MPX boundary violations
6473 @cindex boundary violations, Intel MPX
6474 On some targets, a @code{SIGSEGV} can be caused by a boundary
6475 violation, i.e., accessing an address outside of the allowed range.
6476 In those cases @value{GDBN} may displays additional information,
6477 depending on how @value{GDBN} has been told to handle the signal.
6478 With @code{handle stop SIGSEGV}, @value{GDBN} displays the violation
6479 kind: "Upper" or "Lower", the memory address accessed and the
6480 bounds, while with @code{handle nostop SIGSEGV} no additional
6481 information is displayed.
6483 The usual output of a segfault is:
6485 Program received signal SIGSEGV, Segmentation fault
6486 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6487 68 value = *(p + len);
6490 While a bound violation is presented as:
6492 Program received signal SIGSEGV, Segmentation fault
6493 Upper bound violation while accessing address 0x7fffffffc3b3
6494 Bounds: [lower = 0x7fffffffc390, upper = 0x7fffffffc3a3]
6495 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6496 68 value = *(p + len);
6500 @section Stopping and Starting Multi-thread Programs
6502 @cindex stopped threads
6503 @cindex threads, stopped
6505 @cindex continuing threads
6506 @cindex threads, continuing
6508 @value{GDBN} supports debugging programs with multiple threads
6509 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
6510 are two modes of controlling execution of your program within the
6511 debugger. In the default mode, referred to as @dfn{all-stop mode},
6512 when any thread in your program stops (for example, at a breakpoint
6513 or while being stepped), all other threads in the program are also stopped by
6514 @value{GDBN}. On some targets, @value{GDBN} also supports
6515 @dfn{non-stop mode}, in which other threads can continue to run freely while
6516 you examine the stopped thread in the debugger.
6519 * All-Stop Mode:: All threads stop when GDB takes control
6520 * Non-Stop Mode:: Other threads continue to execute
6521 * Background Execution:: Running your program asynchronously
6522 * Thread-Specific Breakpoints:: Controlling breakpoints
6523 * Interrupted System Calls:: GDB may interfere with system calls
6524 * Observer Mode:: GDB does not alter program behavior
6528 @subsection All-Stop Mode
6530 @cindex all-stop mode
6532 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
6533 @emph{all} threads of execution stop, not just the current thread. This
6534 allows you to examine the overall state of the program, including
6535 switching between threads, without worrying that things may change
6538 Conversely, whenever you restart the program, @emph{all} threads start
6539 executing. @emph{This is true even when single-stepping} with commands
6540 like @code{step} or @code{next}.
6542 In particular, @value{GDBN} cannot single-step all threads in lockstep.
6543 Since thread scheduling is up to your debugging target's operating
6544 system (not controlled by @value{GDBN}), other threads may
6545 execute more than one statement while the current thread completes a
6546 single step. Moreover, in general other threads stop in the middle of a
6547 statement, rather than at a clean statement boundary, when the program
6550 You might even find your program stopped in another thread after
6551 continuing or even single-stepping. This happens whenever some other
6552 thread runs into a breakpoint, a signal, or an exception before the
6553 first thread completes whatever you requested.
6555 @cindex automatic thread selection
6556 @cindex switching threads automatically
6557 @cindex threads, automatic switching
6558 Whenever @value{GDBN} stops your program, due to a breakpoint or a
6559 signal, it automatically selects the thread where that breakpoint or
6560 signal happened. @value{GDBN} alerts you to the context switch with a
6561 message such as @samp{[Switching to Thread @var{n}]} to identify the
6564 On some OSes, you can modify @value{GDBN}'s default behavior by
6565 locking the OS scheduler to allow only a single thread to run.
6568 @item set scheduler-locking @var{mode}
6569 @cindex scheduler locking mode
6570 @cindex lock scheduler
6571 Set the scheduler locking mode. It applies to normal execution,
6572 record mode, and replay mode. If it is @code{off}, then there is no
6573 locking and any thread may run at any time. If @code{on}, then only
6574 the current thread may run when the inferior is resumed. The
6575 @code{step} mode optimizes for single-stepping; it prevents other
6576 threads from preempting the current thread while you are stepping, so
6577 that the focus of debugging does not change unexpectedly. Other
6578 threads never get a chance to run when you step, and they are
6579 completely free to run when you use commands like @samp{continue},
6580 @samp{until}, or @samp{finish}. However, unless another thread hits a
6581 breakpoint during its timeslice, @value{GDBN} does not change the
6582 current thread away from the thread that you are debugging. The
6583 @code{replay} mode behaves like @code{off} in record mode and like
6584 @code{on} in replay mode.
6586 @item show scheduler-locking
6587 Display the current scheduler locking mode.
6590 @cindex resume threads of multiple processes simultaneously
6591 By default, when you issue one of the execution commands such as
6592 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
6593 threads of the current inferior to run. For example, if @value{GDBN}
6594 is attached to two inferiors, each with two threads, the
6595 @code{continue} command resumes only the two threads of the current
6596 inferior. This is useful, for example, when you debug a program that
6597 forks and you want to hold the parent stopped (so that, for instance,
6598 it doesn't run to exit), while you debug the child. In other
6599 situations, you may not be interested in inspecting the current state
6600 of any of the processes @value{GDBN} is attached to, and you may want
6601 to resume them all until some breakpoint is hit. In the latter case,
6602 you can instruct @value{GDBN} to allow all threads of all the
6603 inferiors to run with the @w{@code{set schedule-multiple}} command.
6606 @kindex set schedule-multiple
6607 @item set schedule-multiple
6608 Set the mode for allowing threads of multiple processes to be resumed
6609 when an execution command is issued. When @code{on}, all threads of
6610 all processes are allowed to run. When @code{off}, only the threads
6611 of the current process are resumed. The default is @code{off}. The
6612 @code{scheduler-locking} mode takes precedence when set to @code{on},
6613 or while you are stepping and set to @code{step}.
6615 @item show schedule-multiple
6616 Display the current mode for resuming the execution of threads of
6621 @subsection Non-Stop Mode
6623 @cindex non-stop mode
6625 @c This section is really only a place-holder, and needs to be expanded
6626 @c with more details.
6628 For some multi-threaded targets, @value{GDBN} supports an optional
6629 mode of operation in which you can examine stopped program threads in
6630 the debugger while other threads continue to execute freely. This
6631 minimizes intrusion when debugging live systems, such as programs
6632 where some threads have real-time constraints or must continue to
6633 respond to external events. This is referred to as @dfn{non-stop} mode.
6635 In non-stop mode, when a thread stops to report a debugging event,
6636 @emph{only} that thread is stopped; @value{GDBN} does not stop other
6637 threads as well, in contrast to the all-stop mode behavior. Additionally,
6638 execution commands such as @code{continue} and @code{step} apply by default
6639 only to the current thread in non-stop mode, rather than all threads as
6640 in all-stop mode. This allows you to control threads explicitly in
6641 ways that are not possible in all-stop mode --- for example, stepping
6642 one thread while allowing others to run freely, stepping
6643 one thread while holding all others stopped, or stepping several threads
6644 independently and simultaneously.
6646 To enter non-stop mode, use this sequence of commands before you run
6647 or attach to your program:
6650 # If using the CLI, pagination breaks non-stop.
6653 # Finally, turn it on!
6657 You can use these commands to manipulate the non-stop mode setting:
6660 @kindex set non-stop
6661 @item set non-stop on
6662 Enable selection of non-stop mode.
6663 @item set non-stop off
6664 Disable selection of non-stop mode.
6665 @kindex show non-stop
6667 Show the current non-stop enablement setting.
6670 Note these commands only reflect whether non-stop mode is enabled,
6671 not whether the currently-executing program is being run in non-stop mode.
6672 In particular, the @code{set non-stop} preference is only consulted when
6673 @value{GDBN} starts or connects to the target program, and it is generally
6674 not possible to switch modes once debugging has started. Furthermore,
6675 since not all targets support non-stop mode, even when you have enabled
6676 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
6679 In non-stop mode, all execution commands apply only to the current thread
6680 by default. That is, @code{continue} only continues one thread.
6681 To continue all threads, issue @code{continue -a} or @code{c -a}.
6683 You can use @value{GDBN}'s background execution commands
6684 (@pxref{Background Execution}) to run some threads in the background
6685 while you continue to examine or step others from @value{GDBN}.
6686 The MI execution commands (@pxref{GDB/MI Program Execution}) are
6687 always executed asynchronously in non-stop mode.
6689 Suspending execution is done with the @code{interrupt} command when
6690 running in the background, or @kbd{Ctrl-c} during foreground execution.
6691 In all-stop mode, this stops the whole process;
6692 but in non-stop mode the interrupt applies only to the current thread.
6693 To stop the whole program, use @code{interrupt -a}.
6695 Other execution commands do not currently support the @code{-a} option.
6697 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
6698 that thread current, as it does in all-stop mode. This is because the
6699 thread stop notifications are asynchronous with respect to @value{GDBN}'s
6700 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
6701 changed to a different thread just as you entered a command to operate on the
6702 previously current thread.
6704 @node Background Execution
6705 @subsection Background Execution
6707 @cindex foreground execution
6708 @cindex background execution
6709 @cindex asynchronous execution
6710 @cindex execution, foreground, background and asynchronous
6712 @value{GDBN}'s execution commands have two variants: the normal
6713 foreground (synchronous) behavior, and a background
6714 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
6715 the program to report that some thread has stopped before prompting for
6716 another command. In background execution, @value{GDBN} immediately gives
6717 a command prompt so that you can issue other commands while your program runs.
6719 If the target doesn't support async mode, @value{GDBN} issues an error
6720 message if you attempt to use the background execution commands.
6722 @cindex @code{&}, background execution of commands
6723 To specify background execution, add a @code{&} to the command. For example,
6724 the background form of the @code{continue} command is @code{continue&}, or
6725 just @code{c&}. The execution commands that accept background execution
6731 @xref{Starting, , Starting your Program}.
6735 @xref{Attach, , Debugging an Already-running Process}.
6739 @xref{Continuing and Stepping, step}.
6743 @xref{Continuing and Stepping, stepi}.
6747 @xref{Continuing and Stepping, next}.
6751 @xref{Continuing and Stepping, nexti}.
6755 @xref{Continuing and Stepping, continue}.
6759 @xref{Continuing and Stepping, finish}.
6763 @xref{Continuing and Stepping, until}.
6767 Background execution is especially useful in conjunction with non-stop
6768 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
6769 However, you can also use these commands in the normal all-stop mode with
6770 the restriction that you cannot issue another execution command until the
6771 previous one finishes. Examples of commands that are valid in all-stop
6772 mode while the program is running include @code{help} and @code{info break}.
6774 You can interrupt your program while it is running in the background by
6775 using the @code{interrupt} command.
6782 Suspend execution of the running program. In all-stop mode,
6783 @code{interrupt} stops the whole process, but in non-stop mode, it stops
6784 only the current thread. To stop the whole program in non-stop mode,
6785 use @code{interrupt -a}.
6788 @node Thread-Specific Breakpoints
6789 @subsection Thread-Specific Breakpoints
6791 When your program has multiple threads (@pxref{Threads,, Debugging
6792 Programs with Multiple Threads}), you can choose whether to set
6793 breakpoints on all threads, or on a particular thread.
6796 @cindex breakpoints and threads
6797 @cindex thread breakpoints
6798 @kindex break @dots{} thread @var{thread-id}
6799 @item break @var{location} thread @var{thread-id}
6800 @itemx break @var{location} thread @var{thread-id} if @dots{}
6801 @var{location} specifies source lines; there are several ways of
6802 writing them (@pxref{Specify Location}), but the effect is always to
6803 specify some source line.
6805 Use the qualifier @samp{thread @var{thread-id}} with a breakpoint command
6806 to specify that you only want @value{GDBN} to stop the program when a
6807 particular thread reaches this breakpoint. The @var{thread-id} specifier
6808 is one of the thread identifiers assigned by @value{GDBN}, shown
6809 in the first column of the @samp{info threads} display.
6811 If you do not specify @samp{thread @var{thread-id}} when you set a
6812 breakpoint, the breakpoint applies to @emph{all} threads of your
6815 You can use the @code{thread} qualifier on conditional breakpoints as
6816 well; in this case, place @samp{thread @var{thread-id}} before or
6817 after the breakpoint condition, like this:
6820 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
6825 Thread-specific breakpoints are automatically deleted when
6826 @value{GDBN} detects the corresponding thread is no longer in the
6827 thread list. For example:
6831 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
6834 There are several ways for a thread to disappear, such as a regular
6835 thread exit, but also when you detach from the process with the
6836 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
6837 Process}), or if @value{GDBN} loses the remote connection
6838 (@pxref{Remote Debugging}), etc. Note that with some targets,
6839 @value{GDBN} is only able to detect a thread has exited when the user
6840 explictly asks for the thread list with the @code{info threads}
6843 @node Interrupted System Calls
6844 @subsection Interrupted System Calls
6846 @cindex thread breakpoints and system calls
6847 @cindex system calls and thread breakpoints
6848 @cindex premature return from system calls
6849 There is an unfortunate side effect when using @value{GDBN} to debug
6850 multi-threaded programs. If one thread stops for a
6851 breakpoint, or for some other reason, and another thread is blocked in a
6852 system call, then the system call may return prematurely. This is a
6853 consequence of the interaction between multiple threads and the signals
6854 that @value{GDBN} uses to implement breakpoints and other events that
6857 To handle this problem, your program should check the return value of
6858 each system call and react appropriately. This is good programming
6861 For example, do not write code like this:
6867 The call to @code{sleep} will return early if a different thread stops
6868 at a breakpoint or for some other reason.
6870 Instead, write this:
6875 unslept = sleep (unslept);
6878 A system call is allowed to return early, so the system is still
6879 conforming to its specification. But @value{GDBN} does cause your
6880 multi-threaded program to behave differently than it would without
6883 Also, @value{GDBN} uses internal breakpoints in the thread library to
6884 monitor certain events such as thread creation and thread destruction.
6885 When such an event happens, a system call in another thread may return
6886 prematurely, even though your program does not appear to stop.
6889 @subsection Observer Mode
6891 If you want to build on non-stop mode and observe program behavior
6892 without any chance of disruption by @value{GDBN}, you can set
6893 variables to disable all of the debugger's attempts to modify state,
6894 whether by writing memory, inserting breakpoints, etc. These operate
6895 at a low level, intercepting operations from all commands.
6897 When all of these are set to @code{off}, then @value{GDBN} is said to
6898 be @dfn{observer mode}. As a convenience, the variable
6899 @code{observer} can be set to disable these, plus enable non-stop
6902 Note that @value{GDBN} will not prevent you from making nonsensical
6903 combinations of these settings. For instance, if you have enabled
6904 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
6905 then breakpoints that work by writing trap instructions into the code
6906 stream will still not be able to be placed.
6911 @item set observer on
6912 @itemx set observer off
6913 When set to @code{on}, this disables all the permission variables
6914 below (except for @code{insert-fast-tracepoints}), plus enables
6915 non-stop debugging. Setting this to @code{off} switches back to
6916 normal debugging, though remaining in non-stop mode.
6919 Show whether observer mode is on or off.
6921 @kindex may-write-registers
6922 @item set may-write-registers on
6923 @itemx set may-write-registers off
6924 This controls whether @value{GDBN} will attempt to alter the values of
6925 registers, such as with assignment expressions in @code{print}, or the
6926 @code{jump} command. It defaults to @code{on}.
6928 @item show may-write-registers
6929 Show the current permission to write registers.
6931 @kindex may-write-memory
6932 @item set may-write-memory on
6933 @itemx set may-write-memory off
6934 This controls whether @value{GDBN} will attempt to alter the contents
6935 of memory, such as with assignment expressions in @code{print}. It
6936 defaults to @code{on}.
6938 @item show may-write-memory
6939 Show the current permission to write memory.
6941 @kindex may-insert-breakpoints
6942 @item set may-insert-breakpoints on
6943 @itemx set may-insert-breakpoints off
6944 This controls whether @value{GDBN} will attempt to insert breakpoints.
6945 This affects all breakpoints, including internal breakpoints defined
6946 by @value{GDBN}. It defaults to @code{on}.
6948 @item show may-insert-breakpoints
6949 Show the current permission to insert breakpoints.
6951 @kindex may-insert-tracepoints
6952 @item set may-insert-tracepoints on
6953 @itemx set may-insert-tracepoints off
6954 This controls whether @value{GDBN} will attempt to insert (regular)
6955 tracepoints at the beginning of a tracing experiment. It affects only
6956 non-fast tracepoints, fast tracepoints being under the control of
6957 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
6959 @item show may-insert-tracepoints
6960 Show the current permission to insert tracepoints.
6962 @kindex may-insert-fast-tracepoints
6963 @item set may-insert-fast-tracepoints on
6964 @itemx set may-insert-fast-tracepoints off
6965 This controls whether @value{GDBN} will attempt to insert fast
6966 tracepoints at the beginning of a tracing experiment. It affects only
6967 fast tracepoints, regular (non-fast) tracepoints being under the
6968 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
6970 @item show may-insert-fast-tracepoints
6971 Show the current permission to insert fast tracepoints.
6973 @kindex may-interrupt
6974 @item set may-interrupt on
6975 @itemx set may-interrupt off
6976 This controls whether @value{GDBN} will attempt to interrupt or stop
6977 program execution. When this variable is @code{off}, the
6978 @code{interrupt} command will have no effect, nor will
6979 @kbd{Ctrl-c}. It defaults to @code{on}.
6981 @item show may-interrupt
6982 Show the current permission to interrupt or stop the program.
6986 @node Reverse Execution
6987 @chapter Running programs backward
6988 @cindex reverse execution
6989 @cindex running programs backward
6991 When you are debugging a program, it is not unusual to realize that
6992 you have gone too far, and some event of interest has already happened.
6993 If the target environment supports it, @value{GDBN} can allow you to
6994 ``rewind'' the program by running it backward.
6996 A target environment that supports reverse execution should be able
6997 to ``undo'' the changes in machine state that have taken place as the
6998 program was executing normally. Variables, registers etc.@: should
6999 revert to their previous values. Obviously this requires a great
7000 deal of sophistication on the part of the target environment; not
7001 all target environments can support reverse execution.
7003 When a program is executed in reverse, the instructions that
7004 have most recently been executed are ``un-executed'', in reverse
7005 order. The program counter runs backward, following the previous
7006 thread of execution in reverse. As each instruction is ``un-executed'',
7007 the values of memory and/or registers that were changed by that
7008 instruction are reverted to their previous states. After executing
7009 a piece of source code in reverse, all side effects of that code
7010 should be ``undone'', and all variables should be returned to their
7011 prior values@footnote{
7012 Note that some side effects are easier to undo than others. For instance,
7013 memory and registers are relatively easy, but device I/O is hard. Some
7014 targets may be able undo things like device I/O, and some may not.
7016 The contract between @value{GDBN} and the reverse executing target
7017 requires only that the target do something reasonable when
7018 @value{GDBN} tells it to execute backwards, and then report the
7019 results back to @value{GDBN}. Whatever the target reports back to
7020 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
7021 assumes that the memory and registers that the target reports are in a
7022 consistent state, but @value{GDBN} accepts whatever it is given.
7025 On some platforms, @value{GDBN} has built-in support for reverse
7026 execution, activated with the @code{record} or @code{record btrace}
7027 commands. @xref{Process Record and Replay}. Some remote targets,
7028 typically full system emulators, support reverse execution directly
7029 without requiring any special command.
7031 If you are debugging in a target environment that supports
7032 reverse execution, @value{GDBN} provides the following commands.
7035 @kindex reverse-continue
7036 @kindex rc @r{(@code{reverse-continue})}
7037 @item reverse-continue @r{[}@var{ignore-count}@r{]}
7038 @itemx rc @r{[}@var{ignore-count}@r{]}
7039 Beginning at the point where your program last stopped, start executing
7040 in reverse. Reverse execution will stop for breakpoints and synchronous
7041 exceptions (signals), just like normal execution. Behavior of
7042 asynchronous signals depends on the target environment.
7044 @kindex reverse-step
7045 @kindex rs @r{(@code{step})}
7046 @item reverse-step @r{[}@var{count}@r{]}
7047 Run the program backward until control reaches the start of a
7048 different source line; then stop it, and return control to @value{GDBN}.
7050 Like the @code{step} command, @code{reverse-step} will only stop
7051 at the beginning of a source line. It ``un-executes'' the previously
7052 executed source line. If the previous source line included calls to
7053 debuggable functions, @code{reverse-step} will step (backward) into
7054 the called function, stopping at the beginning of the @emph{last}
7055 statement in the called function (typically a return statement).
7057 Also, as with the @code{step} command, if non-debuggable functions are
7058 called, @code{reverse-step} will run thru them backward without stopping.
7060 @kindex reverse-stepi
7061 @kindex rsi @r{(@code{reverse-stepi})}
7062 @item reverse-stepi @r{[}@var{count}@r{]}
7063 Reverse-execute one machine instruction. Note that the instruction
7064 to be reverse-executed is @emph{not} the one pointed to by the program
7065 counter, but the instruction executed prior to that one. For instance,
7066 if the last instruction was a jump, @code{reverse-stepi} will take you
7067 back from the destination of the jump to the jump instruction itself.
7069 @kindex reverse-next
7070 @kindex rn @r{(@code{reverse-next})}
7071 @item reverse-next @r{[}@var{count}@r{]}
7072 Run backward to the beginning of the previous line executed in
7073 the current (innermost) stack frame. If the line contains function
7074 calls, they will be ``un-executed'' without stopping. Starting from
7075 the first line of a function, @code{reverse-next} will take you back
7076 to the caller of that function, @emph{before} the function was called,
7077 just as the normal @code{next} command would take you from the last
7078 line of a function back to its return to its caller
7079 @footnote{Unless the code is too heavily optimized.}.
7081 @kindex reverse-nexti
7082 @kindex rni @r{(@code{reverse-nexti})}
7083 @item reverse-nexti @r{[}@var{count}@r{]}
7084 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
7085 in reverse, except that called functions are ``un-executed'' atomically.
7086 That is, if the previously executed instruction was a return from
7087 another function, @code{reverse-nexti} will continue to execute
7088 in reverse until the call to that function (from the current stack
7091 @kindex reverse-finish
7092 @item reverse-finish
7093 Just as the @code{finish} command takes you to the point where the
7094 current function returns, @code{reverse-finish} takes you to the point
7095 where it was called. Instead of ending up at the end of the current
7096 function invocation, you end up at the beginning.
7098 @kindex set exec-direction
7099 @item set exec-direction
7100 Set the direction of target execution.
7101 @item set exec-direction reverse
7102 @cindex execute forward or backward in time
7103 @value{GDBN} will perform all execution commands in reverse, until the
7104 exec-direction mode is changed to ``forward''. Affected commands include
7105 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
7106 command cannot be used in reverse mode.
7107 @item set exec-direction forward
7108 @value{GDBN} will perform all execution commands in the normal fashion.
7109 This is the default.
7113 @node Process Record and Replay
7114 @chapter Recording Inferior's Execution and Replaying It
7115 @cindex process record and replay
7116 @cindex recording inferior's execution and replaying it
7118 On some platforms, @value{GDBN} provides a special @dfn{process record
7119 and replay} target that can record a log of the process execution, and
7120 replay it later with both forward and reverse execution commands.
7123 When this target is in use, if the execution log includes the record
7124 for the next instruction, @value{GDBN} will debug in @dfn{replay
7125 mode}. In the replay mode, the inferior does not really execute code
7126 instructions. Instead, all the events that normally happen during
7127 code execution are taken from the execution log. While code is not
7128 really executed in replay mode, the values of registers (including the
7129 program counter register) and the memory of the inferior are still
7130 changed as they normally would. Their contents are taken from the
7134 If the record for the next instruction is not in the execution log,
7135 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
7136 inferior executes normally, and @value{GDBN} records the execution log
7139 The process record and replay target supports reverse execution
7140 (@pxref{Reverse Execution}), even if the platform on which the
7141 inferior runs does not. However, the reverse execution is limited in
7142 this case by the range of the instructions recorded in the execution
7143 log. In other words, reverse execution on platforms that don't
7144 support it directly can only be done in the replay mode.
7146 When debugging in the reverse direction, @value{GDBN} will work in
7147 replay mode as long as the execution log includes the record for the
7148 previous instruction; otherwise, it will work in record mode, if the
7149 platform supports reverse execution, or stop if not.
7151 Currently, process record and replay is supported on ARM, Aarch64,
7152 Moxie, PowerPC, PowerPC64, S/390, and x86 (i386/amd64) running
7153 GNU/Linux. Process record and replay can be used both when native
7154 debugging, and when remote debugging via @code{gdbserver}.
7156 For architecture environments that support process record and replay,
7157 @value{GDBN} provides the following commands:
7160 @kindex target record
7161 @kindex target record-full
7162 @kindex target record-btrace
7165 @kindex record btrace
7166 @kindex record btrace bts
7167 @kindex record btrace pt
7173 @kindex rec btrace bts
7174 @kindex rec btrace pt
7177 @item record @var{method}
7178 This command starts the process record and replay target. The
7179 recording method can be specified as parameter. Without a parameter
7180 the command uses the @code{full} recording method. The following
7181 recording methods are available:
7185 Full record/replay recording using @value{GDBN}'s software record and
7186 replay implementation. This method allows replaying and reverse
7189 @item btrace @var{format}
7190 Hardware-supported instruction recording, supported on Intel
7191 processors. This method does not record data. Further, the data is
7192 collected in a ring buffer so old data will be overwritten when the
7193 buffer is full. It allows limited reverse execution. Variables and
7194 registers are not available during reverse execution. In remote
7195 debugging, recording continues on disconnect. Recorded data can be
7196 inspected after reconnecting. The recording may be stopped using
7199 The recording format can be specified as parameter. Without a parameter
7200 the command chooses the recording format. The following recording
7201 formats are available:
7205 @cindex branch trace store
7206 Use the @dfn{Branch Trace Store} (@acronym{BTS}) recording format. In
7207 this format, the processor stores a from/to record for each executed
7208 branch in the btrace ring buffer.
7211 @cindex Intel Processor Trace
7212 Use the @dfn{Intel Processor Trace} recording format. In this
7213 format, the processor stores the execution trace in a compressed form
7214 that is afterwards decoded by @value{GDBN}.
7216 The trace can be recorded with very low overhead. The compressed
7217 trace format also allows small trace buffers to already contain a big
7218 number of instructions compared to @acronym{BTS}.
7220 Decoding the recorded execution trace, on the other hand, is more
7221 expensive than decoding @acronym{BTS} trace. This is mostly due to the
7222 increased number of instructions to process. You should increase the
7223 buffer-size with care.
7226 Not all recording formats may be available on all processors.
7229 The process record and replay target can only debug a process that is
7230 already running. Therefore, you need first to start the process with
7231 the @kbd{run} or @kbd{start} commands, and then start the recording
7232 with the @kbd{record @var{method}} command.
7234 @cindex displaced stepping, and process record and replay
7235 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
7236 will be automatically disabled when process record and replay target
7237 is started. That's because the process record and replay target
7238 doesn't support displaced stepping.
7240 @cindex non-stop mode, and process record and replay
7241 @cindex asynchronous execution, and process record and replay
7242 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
7243 the asynchronous execution mode (@pxref{Background Execution}), not
7244 all recording methods are available. The @code{full} recording method
7245 does not support these two modes.
7250 Stop the process record and replay target. When process record and
7251 replay target stops, the entire execution log will be deleted and the
7252 inferior will either be terminated, or will remain in its final state.
7254 When you stop the process record and replay target in record mode (at
7255 the end of the execution log), the inferior will be stopped at the
7256 next instruction that would have been recorded. In other words, if
7257 you record for a while and then stop recording, the inferior process
7258 will be left in the same state as if the recording never happened.
7260 On the other hand, if the process record and replay target is stopped
7261 while in replay mode (that is, not at the end of the execution log,
7262 but at some earlier point), the inferior process will become ``live''
7263 at that earlier state, and it will then be possible to continue the
7264 usual ``live'' debugging of the process from that state.
7266 When the inferior process exits, or @value{GDBN} detaches from it,
7267 process record and replay target will automatically stop itself.
7271 Go to a specific location in the execution log. There are several
7272 ways to specify the location to go to:
7275 @item record goto begin
7276 @itemx record goto start
7277 Go to the beginning of the execution log.
7279 @item record goto end
7280 Go to the end of the execution log.
7282 @item record goto @var{n}
7283 Go to instruction number @var{n} in the execution log.
7287 @item record save @var{filename}
7288 Save the execution log to a file @file{@var{filename}}.
7289 Default filename is @file{gdb_record.@var{process_id}}, where
7290 @var{process_id} is the process ID of the inferior.
7292 This command may not be available for all recording methods.
7294 @kindex record restore
7295 @item record restore @var{filename}
7296 Restore the execution log from a file @file{@var{filename}}.
7297 File must have been created with @code{record save}.
7299 @kindex set record full
7300 @item set record full insn-number-max @var{limit}
7301 @itemx set record full insn-number-max unlimited
7302 Set the limit of instructions to be recorded for the @code{full}
7303 recording method. Default value is 200000.
7305 If @var{limit} is a positive number, then @value{GDBN} will start
7306 deleting instructions from the log once the number of the record
7307 instructions becomes greater than @var{limit}. For every new recorded
7308 instruction, @value{GDBN} will delete the earliest recorded
7309 instruction to keep the number of recorded instructions at the limit.
7310 (Since deleting recorded instructions loses information, @value{GDBN}
7311 lets you control what happens when the limit is reached, by means of
7312 the @code{stop-at-limit} option, described below.)
7314 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
7315 delete recorded instructions from the execution log. The number of
7316 recorded instructions is limited only by the available memory.
7318 @kindex show record full
7319 @item show record full insn-number-max
7320 Show the limit of instructions to be recorded with the @code{full}
7323 @item set record full stop-at-limit
7324 Control the behavior of the @code{full} recording method when the
7325 number of recorded instructions reaches the limit. If ON (the
7326 default), @value{GDBN} will stop when the limit is reached for the
7327 first time and ask you whether you want to stop the inferior or
7328 continue running it and recording the execution log. If you decide
7329 to continue recording, each new recorded instruction will cause the
7330 oldest one to be deleted.
7332 If this option is OFF, @value{GDBN} will automatically delete the
7333 oldest record to make room for each new one, without asking.
7335 @item show record full stop-at-limit
7336 Show the current setting of @code{stop-at-limit}.
7338 @item set record full memory-query
7339 Control the behavior when @value{GDBN} is unable to record memory
7340 changes caused by an instruction for the @code{full} recording method.
7341 If ON, @value{GDBN} will query whether to stop the inferior in that
7344 If this option is OFF (the default), @value{GDBN} will automatically
7345 ignore the effect of such instructions on memory. Later, when
7346 @value{GDBN} replays this execution log, it will mark the log of this
7347 instruction as not accessible, and it will not affect the replay
7350 @item show record full memory-query
7351 Show the current setting of @code{memory-query}.
7353 @kindex set record btrace
7354 The @code{btrace} record target does not trace data. As a
7355 convenience, when replaying, @value{GDBN} reads read-only memory off
7356 the live program directly, assuming that the addresses of the
7357 read-only areas don't change. This for example makes it possible to
7358 disassemble code while replaying, but not to print variables.
7359 In some cases, being able to inspect variables might be useful.
7360 You can use the following command for that:
7362 @item set record btrace replay-memory-access
7363 Control the behavior of the @code{btrace} recording method when
7364 accessing memory during replay. If @code{read-only} (the default),
7365 @value{GDBN} will only allow accesses to read-only memory.
7366 If @code{read-write}, @value{GDBN} will allow accesses to read-only
7367 and to read-write memory. Beware that the accessed memory corresponds
7368 to the live target and not necessarily to the current replay
7371 @item set record btrace cpu @var{identifier}
7372 Set the processor to be used for enabling workarounds for processor
7373 errata when decoding the trace.
7375 Processor errata are defects in processor operation, caused by its
7376 design or manufacture. They can cause a trace not to match the
7377 specification. This, in turn, may cause trace decode to fail.
7378 @value{GDBN} can detect erroneous trace packets and correct them, thus
7379 avoiding the decoding failures. These corrections are known as
7380 @dfn{errata workarounds}, and are enabled based on the processor on
7381 which the trace was recorded.
7383 By default, @value{GDBN} attempts to detect the processor
7384 automatically, and apply the necessary workarounds for it. However,
7385 you may need to specify the processor if @value{GDBN} does not yet
7386 support it. This command allows you to do that, and also allows to
7387 disable the workarounds.
7389 The argument @var{identifier} identifies the @sc{cpu} and is of the
7390 form: @code{@var{vendor}:@var{processor identifier}}. In addition,
7391 there are two special identifiers, @code{none} and @code{auto}
7394 The following vendor identifiers and corresponding processor
7395 identifiers are currently supported:
7397 @multitable @columnfractions .1 .9
7400 @tab @var{family}/@var{model}[/@var{stepping}]
7404 On GNU/Linux systems, the processor @var{family}, @var{model}, and
7405 @var{stepping} can be obtained from @code{/proc/cpuinfo}.
7407 If @var{identifier} is @code{auto}, enable errata workarounds for the
7408 processor on which the trace was recorded. If @var{identifier} is
7409 @code{none}, errata workarounds are disabled.
7411 For example, when using an old @value{GDBN} on a new system, decode
7412 may fail because @value{GDBN} does not support the new processor. It
7413 often suffices to specify an older processor that @value{GDBN}
7417 (@value{GDBP}) info record
7418 Active record target: record-btrace
7419 Recording format: Intel Processor Trace.
7421 Failed to configure the Intel Processor Trace decoder: unknown cpu.
7422 (@value{GDBP}) set record btrace cpu intel:6/158
7423 (@value{GDBP}) info record
7424 Active record target: record-btrace
7425 Recording format: Intel Processor Trace.
7427 Recorded 84872 instructions in 3189 functions (0 gaps) for thread 1 (...).
7430 @kindex show record btrace
7431 @item show record btrace replay-memory-access
7432 Show the current setting of @code{replay-memory-access}.
7434 @item show record btrace cpu
7435 Show the processor to be used for enabling trace decode errata
7438 @kindex set record btrace bts
7439 @item set record btrace bts buffer-size @var{size}
7440 @itemx set record btrace bts buffer-size unlimited
7441 Set the requested ring buffer size for branch tracing in @acronym{BTS}
7442 format. Default is 64KB.
7444 If @var{size} is a positive number, then @value{GDBN} will try to
7445 allocate a buffer of at least @var{size} bytes for each new thread
7446 that uses the btrace recording method and the @acronym{BTS} format.
7447 The actually obtained buffer size may differ from the requested
7448 @var{size}. Use the @code{info record} command to see the actual
7449 buffer size for each thread that uses the btrace recording method and
7450 the @acronym{BTS} format.
7452 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7453 allocate a buffer of 4MB.
7455 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7456 also need longer to process the branch trace data before it can be used.
7458 @item show record btrace bts buffer-size @var{size}
7459 Show the current setting of the requested ring buffer size for branch
7460 tracing in @acronym{BTS} format.
7462 @kindex set record btrace pt
7463 @item set record btrace pt buffer-size @var{size}
7464 @itemx set record btrace pt buffer-size unlimited
7465 Set the requested ring buffer size for branch tracing in Intel
7466 Processor Trace format. Default is 16KB.
7468 If @var{size} is a positive number, then @value{GDBN} will try to
7469 allocate a buffer of at least @var{size} bytes for each new thread
7470 that uses the btrace recording method and the Intel Processor Trace
7471 format. The actually obtained buffer size may differ from the
7472 requested @var{size}. Use the @code{info record} command to see the
7473 actual buffer size for each thread.
7475 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7476 allocate a buffer of 4MB.
7478 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7479 also need longer to process the branch trace data before it can be used.
7481 @item show record btrace pt buffer-size @var{size}
7482 Show the current setting of the requested ring buffer size for branch
7483 tracing in Intel Processor Trace format.
7487 Show various statistics about the recording depending on the recording
7492 For the @code{full} recording method, it shows the state of process
7493 record and its in-memory execution log buffer, including:
7497 Whether in record mode or replay mode.
7499 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
7501 Highest recorded instruction number.
7503 Current instruction about to be replayed (if in replay mode).
7505 Number of instructions contained in the execution log.
7507 Maximum number of instructions that may be contained in the execution log.
7511 For the @code{btrace} recording method, it shows:
7517 Number of instructions that have been recorded.
7519 Number of blocks of sequential control-flow formed by the recorded
7522 Whether in record mode or replay mode.
7525 For the @code{bts} recording format, it also shows:
7528 Size of the perf ring buffer.
7531 For the @code{pt} recording format, it also shows:
7534 Size of the perf ring buffer.
7538 @kindex record delete
7541 When record target runs in replay mode (``in the past''), delete the
7542 subsequent execution log and begin to record a new execution log starting
7543 from the current address. This means you will abandon the previously
7544 recorded ``future'' and begin recording a new ``future''.
7546 @kindex record instruction-history
7547 @kindex rec instruction-history
7548 @item record instruction-history
7549 Disassembles instructions from the recorded execution log. By
7550 default, ten instructions are disassembled. This can be changed using
7551 the @code{set record instruction-history-size} command. Instructions
7552 are printed in execution order.
7554 It can also print mixed source+disassembly if you specify the the
7555 @code{/m} or @code{/s} modifier, and print the raw instructions in hex
7556 as well as in symbolic form by specifying the @code{/r} modifier.
7558 The current position marker is printed for the instruction at the
7559 current program counter value. This instruction can appear multiple
7560 times in the trace and the current position marker will be printed
7561 every time. To omit the current position marker, specify the
7564 To better align the printed instructions when the trace contains
7565 instructions from more than one function, the function name may be
7566 omitted by specifying the @code{/f} modifier.
7568 Speculatively executed instructions are prefixed with @samp{?}. This
7569 feature is not available for all recording formats.
7571 There are several ways to specify what part of the execution log to
7575 @item record instruction-history @var{insn}
7576 Disassembles ten instructions starting from instruction number
7579 @item record instruction-history @var{insn}, +/-@var{n}
7580 Disassembles @var{n} instructions around instruction number
7581 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
7582 @var{n} instructions after instruction number @var{insn}. If
7583 @var{n} is preceded with @code{-}, disassembles @var{n}
7584 instructions before instruction number @var{insn}.
7586 @item record instruction-history
7587 Disassembles ten more instructions after the last disassembly.
7589 @item record instruction-history -
7590 Disassembles ten more instructions before the last disassembly.
7592 @item record instruction-history @var{begin}, @var{end}
7593 Disassembles instructions beginning with instruction number
7594 @var{begin} until instruction number @var{end}. The instruction
7595 number @var{end} is included.
7598 This command may not be available for all recording methods.
7601 @item set record instruction-history-size @var{size}
7602 @itemx set record instruction-history-size unlimited
7603 Define how many instructions to disassemble in the @code{record
7604 instruction-history} command. The default value is 10.
7605 A @var{size} of @code{unlimited} means unlimited instructions.
7608 @item show record instruction-history-size
7609 Show how many instructions to disassemble in the @code{record
7610 instruction-history} command.
7612 @kindex record function-call-history
7613 @kindex rec function-call-history
7614 @item record function-call-history
7615 Prints the execution history at function granularity. It prints one
7616 line for each sequence of instructions that belong to the same
7617 function giving the name of that function, the source lines
7618 for this instruction sequence (if the @code{/l} modifier is
7619 specified), and the instructions numbers that form the sequence (if
7620 the @code{/i} modifier is specified). The function names are indented
7621 to reflect the call stack depth if the @code{/c} modifier is
7622 specified. The @code{/l}, @code{/i}, and @code{/c} modifiers can be
7626 (@value{GDBP}) @b{list 1, 10}
7637 (@value{GDBP}) @b{record function-call-history /ilc}
7638 1 bar inst 1,4 at foo.c:6,8
7639 2 foo inst 5,10 at foo.c:2,3
7640 3 bar inst 11,13 at foo.c:9,10
7643 By default, ten lines are printed. This can be changed using the
7644 @code{set record function-call-history-size} command. Functions are
7645 printed in execution order. There are several ways to specify what
7649 @item record function-call-history @var{func}
7650 Prints ten functions starting from function number @var{func}.
7652 @item record function-call-history @var{func}, +/-@var{n}
7653 Prints @var{n} functions around function number @var{func}. If
7654 @var{n} is preceded with @code{+}, prints @var{n} functions after
7655 function number @var{func}. If @var{n} is preceded with @code{-},
7656 prints @var{n} functions before function number @var{func}.
7658 @item record function-call-history
7659 Prints ten more functions after the last ten-line print.
7661 @item record function-call-history -
7662 Prints ten more functions before the last ten-line print.
7664 @item record function-call-history @var{begin}, @var{end}
7665 Prints functions beginning with function number @var{begin} until
7666 function number @var{end}. The function number @var{end} is included.
7669 This command may not be available for all recording methods.
7671 @item set record function-call-history-size @var{size}
7672 @itemx set record function-call-history-size unlimited
7673 Define how many lines to print in the
7674 @code{record function-call-history} command. The default value is 10.
7675 A size of @code{unlimited} means unlimited lines.
7677 @item show record function-call-history-size
7678 Show how many lines to print in the
7679 @code{record function-call-history} command.
7684 @chapter Examining the Stack
7686 When your program has stopped, the first thing you need to know is where it
7687 stopped and how it got there.
7690 Each time your program performs a function call, information about the call
7692 That information includes the location of the call in your program,
7693 the arguments of the call,
7694 and the local variables of the function being called.
7695 The information is saved in a block of data called a @dfn{stack frame}.
7696 The stack frames are allocated in a region of memory called the @dfn{call
7699 When your program stops, the @value{GDBN} commands for examining the
7700 stack allow you to see all of this information.
7702 @cindex selected frame
7703 One of the stack frames is @dfn{selected} by @value{GDBN} and many
7704 @value{GDBN} commands refer implicitly to the selected frame. In
7705 particular, whenever you ask @value{GDBN} for the value of a variable in
7706 your program, the value is found in the selected frame. There are
7707 special @value{GDBN} commands to select whichever frame you are
7708 interested in. @xref{Selection, ,Selecting a Frame}.
7710 When your program stops, @value{GDBN} automatically selects the
7711 currently executing frame and describes it briefly, similar to the
7712 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
7715 * Frames:: Stack frames
7716 * Backtrace:: Backtraces
7717 * Selection:: Selecting a frame
7718 * Frame Info:: Information on a frame
7719 * Frame Apply:: Applying a command to several frames
7720 * Frame Filter Management:: Managing frame filters
7725 @section Stack Frames
7727 @cindex frame, definition
7729 The call stack is divided up into contiguous pieces called @dfn{stack
7730 frames}, or @dfn{frames} for short; each frame is the data associated
7731 with one call to one function. The frame contains the arguments given
7732 to the function, the function's local variables, and the address at
7733 which the function is executing.
7735 @cindex initial frame
7736 @cindex outermost frame
7737 @cindex innermost frame
7738 When your program is started, the stack has only one frame, that of the
7739 function @code{main}. This is called the @dfn{initial} frame or the
7740 @dfn{outermost} frame. Each time a function is called, a new frame is
7741 made. Each time a function returns, the frame for that function invocation
7742 is eliminated. If a function is recursive, there can be many frames for
7743 the same function. The frame for the function in which execution is
7744 actually occurring is called the @dfn{innermost} frame. This is the most
7745 recently created of all the stack frames that still exist.
7747 @cindex frame pointer
7748 Inside your program, stack frames are identified by their addresses. A
7749 stack frame consists of many bytes, each of which has its own address; each
7750 kind of computer has a convention for choosing one byte whose
7751 address serves as the address of the frame. Usually this address is kept
7752 in a register called the @dfn{frame pointer register}
7753 (@pxref{Registers, $fp}) while execution is going on in that frame.
7756 @cindex frame number
7757 @value{GDBN} labels each existing stack frame with a @dfn{level}, a
7758 number that is zero for the innermost frame, one for the frame that
7759 called it, and so on upward. These level numbers give you a way of
7760 designating stack frames in @value{GDBN} commands. The terms
7761 @dfn{frame number} and @dfn{frame level} can be used interchangeably to
7762 describe this number.
7764 @c The -fomit-frame-pointer below perennially causes hbox overflow
7765 @c underflow problems.
7766 @cindex frameless execution
7767 Some compilers provide a way to compile functions so that they operate
7768 without stack frames. (For example, the @value{NGCC} option
7770 @samp{-fomit-frame-pointer}
7772 generates functions without a frame.)
7773 This is occasionally done with heavily used library functions to save
7774 the frame setup time. @value{GDBN} has limited facilities for dealing
7775 with these function invocations. If the innermost function invocation
7776 has no stack frame, @value{GDBN} nevertheless regards it as though
7777 it had a separate frame, which is numbered zero as usual, allowing
7778 correct tracing of the function call chain. However, @value{GDBN} has
7779 no provision for frameless functions elsewhere in the stack.
7785 @cindex call stack traces
7786 A backtrace is a summary of how your program got where it is. It shows one
7787 line per frame, for many frames, starting with the currently executing
7788 frame (frame zero), followed by its caller (frame one), and on up the
7791 @anchor{backtrace-command}
7793 @kindex bt @r{(@code{backtrace})}
7794 To print a backtrace of the entire stack, use the @code{backtrace}
7795 command, or its alias @code{bt}. This command will print one line per
7796 frame for frames in the stack. By default, all stack frames are
7797 printed. You can stop the backtrace at any time by typing the system
7798 interrupt character, normally @kbd{Ctrl-c}.
7801 @item backtrace [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
7802 @itemx bt [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
7803 Print the backtrace of the entire stack.
7805 The optional @var{count} can be one of the following:
7810 Print only the innermost @var{n} frames, where @var{n} is a positive
7815 Print only the outermost @var{n} frames, where @var{n} is a positive
7823 Print the values of the local variables also. This can be combined
7824 with the optional @var{count} to limit the number of frames shown.
7827 Do not run Python frame filters on this backtrace. @xref{Frame
7828 Filter API}, for more information. Additionally use @ref{disable
7829 frame-filter all} to turn off all frame filters. This is only
7830 relevant when @value{GDBN} has been configured with @code{Python}
7834 A Python frame filter might decide to ``elide'' some frames. Normally
7835 such elided frames are still printed, but they are indented relative
7836 to the filtered frames that cause them to be elided. The @code{-hide}
7837 option causes elided frames to not be printed at all.
7840 The @code{backtrace} command also supports a number of options that
7841 allow overriding relevant global print settings as set by @code{set
7842 backtrace} and @code{set print} subcommands:
7845 @item -past-main [@code{on}|@code{off}]
7846 Set whether backtraces should continue past @code{main}. Related setting:
7847 @ref{set backtrace past-main}.
7849 @item -past-entry [@code{on}|@code{off}]
7850 Set whether backtraces should continue past the entry point of a program.
7851 Related setting: @ref{set backtrace past-entry}.
7853 @item -entry-values @code{no}|@code{only}|@code{preferred}|@code{if-needed}|@code{both}|@code{compact}|@code{default}
7854 Set printing of function arguments at function entry.
7855 Related setting: @ref{set print entry-values}.
7857 @item -frame-arguments @code{all}|@code{scalars}|@code{none}
7858 Set printing of non-scalar frame arguments.
7859 Related setting: @ref{set print frame-arguments}.
7861 @item -raw-frame-arguments [@code{on}|@code{off}]
7862 Set whether to print frame arguments in raw form.
7863 Related setting: @ref{set print raw-frame-arguments}.
7865 @item -frame-info @code{auto}|@code{source-line}|@code{location}|@code{source-and-location}|@code{location-and-address}|@code{short-location}
7866 Set printing of frame information.
7867 Related setting: @ref{set print frame-info}.
7870 The optional @var{qualifier} is maintained for backward compatibility.
7871 It can be one of the following:
7875 Equivalent to the @code{-full} option.
7878 Equivalent to the @code{-no-filters} option.
7881 Equivalent to the @code{-hide} option.
7888 The names @code{where} and @code{info stack} (abbreviated @code{info s})
7889 are additional aliases for @code{backtrace}.
7891 @cindex multiple threads, backtrace
7892 In a multi-threaded program, @value{GDBN} by default shows the
7893 backtrace only for the current thread. To display the backtrace for
7894 several or all of the threads, use the command @code{thread apply}
7895 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
7896 apply all backtrace}, @value{GDBN} will display the backtrace for all
7897 the threads; this is handy when you debug a core dump of a
7898 multi-threaded program.
7900 Each line in the backtrace shows the frame number and the function name.
7901 The program counter value is also shown---unless you use @code{set
7902 print address off}. The backtrace also shows the source file name and
7903 line number, as well as the arguments to the function. The program
7904 counter value is omitted if it is at the beginning of the code for that
7907 Here is an example of a backtrace. It was made with the command
7908 @samp{bt 3}, so it shows the innermost three frames.
7912 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7914 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
7915 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
7917 (More stack frames follow...)
7922 The display for frame zero does not begin with a program counter
7923 value, indicating that your program has stopped at the beginning of the
7924 code for line @code{993} of @code{builtin.c}.
7927 The value of parameter @code{data} in frame 1 has been replaced by
7928 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
7929 only if it is a scalar (integer, pointer, enumeration, etc). See command
7930 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
7931 on how to configure the way function parameter values are printed.
7932 The command @kbd{set print frame-info} (@pxref{Print Settings}) controls
7933 what frame information is printed.
7935 @cindex optimized out, in backtrace
7936 @cindex function call arguments, optimized out
7937 If your program was compiled with optimizations, some compilers will
7938 optimize away arguments passed to functions if those arguments are
7939 never used after the call. Such optimizations generate code that
7940 passes arguments through registers, but doesn't store those arguments
7941 in the stack frame. @value{GDBN} has no way of displaying such
7942 arguments in stack frames other than the innermost one. Here's what
7943 such a backtrace might look like:
7947 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7949 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
7950 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
7952 (More stack frames follow...)
7957 The values of arguments that were not saved in their stack frames are
7958 shown as @samp{<optimized out>}.
7960 If you need to display the values of such optimized-out arguments,
7961 either deduce that from other variables whose values depend on the one
7962 you are interested in, or recompile without optimizations.
7964 @cindex backtrace beyond @code{main} function
7965 @cindex program entry point
7966 @cindex startup code, and backtrace
7967 Most programs have a standard user entry point---a place where system
7968 libraries and startup code transition into user code. For C this is
7969 @code{main}@footnote{
7970 Note that embedded programs (the so-called ``free-standing''
7971 environment) are not required to have a @code{main} function as the
7972 entry point. They could even have multiple entry points.}.
7973 When @value{GDBN} finds the entry function in a backtrace
7974 it will terminate the backtrace, to avoid tracing into highly
7975 system-specific (and generally uninteresting) code.
7977 If you need to examine the startup code, or limit the number of levels
7978 in a backtrace, you can change this behavior:
7981 @item set backtrace past-main
7982 @itemx set backtrace past-main on
7983 @anchor{set backtrace past-main}
7984 @kindex set backtrace
7985 Backtraces will continue past the user entry point.
7987 @item set backtrace past-main off
7988 Backtraces will stop when they encounter the user entry point. This is the
7991 @item show backtrace past-main
7992 @kindex show backtrace
7993 Display the current user entry point backtrace policy.
7995 @item set backtrace past-entry
7996 @itemx set backtrace past-entry on
7997 @anchor{set backtrace past-entry}
7998 Backtraces will continue past the internal entry point of an application.
7999 This entry point is encoded by the linker when the application is built,
8000 and is likely before the user entry point @code{main} (or equivalent) is called.
8002 @item set backtrace past-entry off
8003 Backtraces will stop when they encounter the internal entry point of an
8004 application. This is the default.
8006 @item show backtrace past-entry
8007 Display the current internal entry point backtrace policy.
8009 @item set backtrace limit @var{n}
8010 @itemx set backtrace limit 0
8011 @itemx set backtrace limit unlimited
8012 @anchor{set backtrace limit}
8013 @cindex backtrace limit
8014 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
8015 or zero means unlimited levels.
8017 @item show backtrace limit
8018 Display the current limit on backtrace levels.
8021 You can control how file names are displayed.
8024 @item set filename-display
8025 @itemx set filename-display relative
8026 @cindex filename-display
8027 Display file names relative to the compilation directory. This is the default.
8029 @item set filename-display basename
8030 Display only basename of a filename.
8032 @item set filename-display absolute
8033 Display an absolute filename.
8035 @item show filename-display
8036 Show the current way to display filenames.
8040 @section Selecting a Frame
8042 Most commands for examining the stack and other data in your program work on
8043 whichever stack frame is selected at the moment. Here are the commands for
8044 selecting a stack frame; all of them finish by printing a brief description
8045 of the stack frame just selected.
8048 @kindex frame@r{, selecting}
8049 @kindex f @r{(@code{frame})}
8050 @item frame @r{[} @var{frame-selection-spec} @r{]}
8051 @item f @r{[} @var{frame-selection-spec} @r{]}
8052 The @command{frame} command allows different stack frames to be
8053 selected. The @var{frame-selection-spec} can be any of the following:
8058 @item level @var{num}
8059 Select frame level @var{num}. Recall that frame zero is the innermost
8060 (currently executing) frame, frame one is the frame that called the
8061 innermost one, and so on. The highest level frame is usually the one
8064 As this is the most common method of navigating the frame stack, the
8065 string @command{level} can be omitted. For example, the following two
8066 commands are equivalent:
8069 (@value{GDBP}) frame 3
8070 (@value{GDBP}) frame level 3
8073 @kindex frame address
8074 @item address @var{stack-address}
8075 Select the frame with stack address @var{stack-address}. The
8076 @var{stack-address} for a frame can be seen in the output of
8077 @command{info frame}, for example:
8080 (@value{GDBP}) info frame
8081 Stack level 1, frame at 0x7fffffffda30:
8082 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
8083 tail call frame, caller of frame at 0x7fffffffda30
8084 source language c++.
8085 Arglist at unknown address.
8086 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
8089 The @var{stack-address} for this frame is @code{0x7fffffffda30} as
8090 indicated by the line:
8093 Stack level 1, frame at 0x7fffffffda30:
8096 @kindex frame function
8097 @item function @var{function-name}
8098 Select the stack frame for function @var{function-name}. If there are
8099 multiple stack frames for function @var{function-name} then the inner
8100 most stack frame is selected.
8103 @item view @var{stack-address} @r{[} @var{pc-addr} @r{]}
8104 View a frame that is not part of @value{GDBN}'s backtrace. The frame
8105 viewed has stack address @var{stack-addr}, and optionally, a program
8106 counter address of @var{pc-addr}.
8108 This is useful mainly if the chaining of stack frames has been
8109 damaged by a bug, making it impossible for @value{GDBN} to assign
8110 numbers properly to all frames. In addition, this can be useful
8111 when your program has multiple stacks and switches between them.
8113 When viewing a frame outside the current backtrace using
8114 @command{frame view} then you can always return to the original
8115 stack using one of the previous stack frame selection instructions,
8116 for example @command{frame level 0}.
8122 Move @var{n} frames up the stack; @var{n} defaults to 1. For positive
8123 numbers @var{n}, this advances toward the outermost frame, to higher
8124 frame numbers, to frames that have existed longer.
8127 @kindex do @r{(@code{down})}
8129 Move @var{n} frames down the stack; @var{n} defaults to 1. For
8130 positive numbers @var{n}, this advances toward the innermost frame, to
8131 lower frame numbers, to frames that were created more recently.
8132 You may abbreviate @code{down} as @code{do}.
8135 All of these commands end by printing two lines of output describing the
8136 frame. The first line shows the frame number, the function name, the
8137 arguments, and the source file and line number of execution in that
8138 frame. The second line shows the text of that source line.
8146 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
8148 10 read_input_file (argv[i]);
8152 After such a printout, the @code{list} command with no arguments
8153 prints ten lines centered on the point of execution in the frame.
8154 You can also edit the program at the point of execution with your favorite
8155 editing program by typing @code{edit}.
8156 @xref{List, ,Printing Source Lines},
8160 @kindex select-frame
8161 @item select-frame @r{[} @var{frame-selection-spec} @r{]}
8162 The @code{select-frame} command is a variant of @code{frame} that does
8163 not display the new frame after selecting it. This command is
8164 intended primarily for use in @value{GDBN} command scripts, where the
8165 output might be unnecessary and distracting. The
8166 @var{frame-selection-spec} is as for the @command{frame} command
8167 described in @ref{Selection, ,Selecting a Frame}.
8169 @kindex down-silently
8171 @item up-silently @var{n}
8172 @itemx down-silently @var{n}
8173 These two commands are variants of @code{up} and @code{down},
8174 respectively; they differ in that they do their work silently, without
8175 causing display of the new frame. They are intended primarily for use
8176 in @value{GDBN} command scripts, where the output might be unnecessary and
8181 @section Information About a Frame
8183 There are several other commands to print information about the selected
8189 When used without any argument, this command does not change which
8190 frame is selected, but prints a brief description of the currently
8191 selected stack frame. It can be abbreviated @code{f}. With an
8192 argument, this command is used to select a stack frame.
8193 @xref{Selection, ,Selecting a Frame}.
8196 @kindex info f @r{(@code{info frame})}
8199 This command prints a verbose description of the selected stack frame,
8204 the address of the frame
8206 the address of the next frame down (called by this frame)
8208 the address of the next frame up (caller of this frame)
8210 the language in which the source code corresponding to this frame is written
8212 the address of the frame's arguments
8214 the address of the frame's local variables
8216 the program counter saved in it (the address of execution in the caller frame)
8218 which registers were saved in the frame
8221 @noindent The verbose description is useful when
8222 something has gone wrong that has made the stack format fail to fit
8223 the usual conventions.
8225 @item info frame @r{[} @var{frame-selection-spec} @r{]}
8226 @itemx info f @r{[} @var{frame-selection-spec} @r{]}
8227 Print a verbose description of the frame selected by
8228 @var{frame-selection-spec}. The @var{frame-selection-spec} is the
8229 same as for the @command{frame} command (@pxref{Selection, ,Selecting
8230 a Frame}). The selected frame remains unchanged by this command.
8233 @item info args [-q]
8234 Print the arguments of the selected frame, each on a separate line.
8236 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
8237 printing header information and messages explaining why no argument
8240 @item info args [-q] [-t @var{type_regexp}] [@var{regexp}]
8241 Like @kbd{info args}, but only print the arguments selected
8242 with the provided regexp(s).
8244 If @var{regexp} is provided, print only the arguments whose names
8245 match the regular expression @var{regexp}.
8247 If @var{type_regexp} is provided, print only the arguments whose
8248 types, as printed by the @code{whatis} command, match
8249 the regular expression @var{type_regexp}.
8250 If @var{type_regexp} contains space(s), it should be enclosed in
8251 quote characters. If needed, use backslash to escape the meaning
8252 of special characters or quotes.
8254 If both @var{regexp} and @var{type_regexp} are provided, an argument
8255 is printed only if its name matches @var{regexp} and its type matches
8258 @item info locals [-q]
8260 Print the local variables of the selected frame, each on a separate
8261 line. These are all variables (declared either static or automatic)
8262 accessible at the point of execution of the selected frame.
8264 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
8265 printing header information and messages explaining why no local variables
8268 @item info locals [-q] [-t @var{type_regexp}] [@var{regexp}]
8269 Like @kbd{info locals}, but only print the local variables selected
8270 with the provided regexp(s).
8272 If @var{regexp} is provided, print only the local variables whose names
8273 match the regular expression @var{regexp}.
8275 If @var{type_regexp} is provided, print only the local variables whose
8276 types, as printed by the @code{whatis} command, match
8277 the regular expression @var{type_regexp}.
8278 If @var{type_regexp} contains space(s), it should be enclosed in
8279 quote characters. If needed, use backslash to escape the meaning
8280 of special characters or quotes.
8282 If both @var{regexp} and @var{type_regexp} are provided, a local variable
8283 is printed only if its name matches @var{regexp} and its type matches
8286 The command @kbd{info locals -q -t @var{type_regexp}} can usefully be
8287 combined with the commands @kbd{frame apply} and @kbd{thread apply}.
8288 For example, your program might use Resource Acquisition Is
8289 Initialization types (RAII) such as @code{lock_something_t}: each
8290 local variable of type @code{lock_something_t} automatically places a
8291 lock that is destroyed when the variable goes out of scope. You can
8292 then list all acquired locks in your program by doing
8294 thread apply all -s frame apply all -s info locals -q -t lock_something_t
8297 or the equivalent shorter form
8299 tfaas i lo -q -t lock_something_t
8305 @section Applying a Command to Several Frames.
8306 @anchor{frame apply}
8308 @cindex apply command to several frames
8310 @item frame apply [all | @var{count} | @var{-count} | level @var{level}@dots{}] [@var{option}]@dots{} @var{command}
8311 The @code{frame apply} command allows you to apply the named
8312 @var{command} to one or more frames.
8316 Specify @code{all} to apply @var{command} to all frames.
8319 Use @var{count} to apply @var{command} to the innermost @var{count}
8320 frames, where @var{count} is a positive number.
8323 Use @var{-count} to apply @var{command} to the outermost @var{count}
8324 frames, where @var{count} is a positive number.
8327 Use @code{level} to apply @var{command} to the set of frames identified
8328 by the @var{level} list. @var{level} is a frame level or a range of frame
8329 levels as @var{level1}-@var{level2}. The frame level is the number shown
8330 in the first field of the @samp{backtrace} command output.
8331 E.g., @samp{2-4 6-8 3} indicates to apply @var{command} for the frames
8332 at levels 2, 3, 4, 6, 7, 8, and then again on frame at level 3.
8336 Note that the frames on which @code{frame apply} applies a command are
8337 also influenced by the @code{set backtrace} settings such as @code{set
8338 backtrace past-main} and @code{set backtrace limit N}.
8339 @xref{Backtrace,,Backtraces}.
8341 The @code{frame apply} command also supports a number of options that
8342 allow overriding relevant @code{set backtrace} settings:
8345 @item -past-main [@code{on}|@code{off}]
8346 Whether backtraces should continue past @code{main}.
8347 Related setting: @ref{set backtrace past-main}.
8349 @item -past-entry [@code{on}|@code{off}]
8350 Whether backtraces should continue past the entry point of a program.
8351 Related setting: @ref{set backtrace past-entry}.
8354 By default, @value{GDBN} displays some frame information before the
8355 output produced by @var{command}, and an error raised during the
8356 execution of a @var{command} will abort @code{frame apply}. The
8357 following options can be used to fine-tune these behaviors:
8361 The flag @code{-c}, which stands for @samp{continue}, causes any
8362 errors in @var{command} to be displayed, and the execution of
8363 @code{frame apply} then continues.
8365 The flag @code{-s}, which stands for @samp{silent}, causes any errors
8366 or empty output produced by a @var{command} to be silently ignored.
8367 That is, the execution continues, but the frame information and errors
8370 The flag @code{-q} (@samp{quiet}) disables printing the frame
8374 The following example shows how the flags @code{-c} and @code{-s} are
8375 working when applying the command @code{p j} to all frames, where
8376 variable @code{j} can only be successfully printed in the outermost
8377 @code{#1 main} frame.
8381 (@value{GDBP}) frame apply all p j
8382 #0 some_function (i=5) at fun.c:4
8383 No symbol "j" in current context.
8384 (@value{GDBP}) frame apply all -c p j
8385 #0 some_function (i=5) at fun.c:4
8386 No symbol "j" in current context.
8387 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8389 (@value{GDBP}) frame apply all -s p j
8390 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8396 By default, @samp{frame apply}, prints the frame location
8397 information before the command output:
8401 (@value{GDBP}) frame apply all p $sp
8402 #0 some_function (i=5) at fun.c:4
8403 $4 = (void *) 0xffffd1e0
8404 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8405 $5 = (void *) 0xffffd1f0
8410 If the flag @code{-q} is given, no frame information is printed:
8413 (@value{GDBP}) frame apply all -q p $sp
8414 $12 = (void *) 0xffffd1e0
8415 $13 = (void *) 0xffffd1f0
8425 @cindex apply a command to all frames (ignoring errors and empty output)
8426 @item faas @var{command}
8427 Shortcut for @code{frame apply all -s @var{command}}.
8428 Applies @var{command} on all frames, ignoring errors and empty output.
8430 It can for example be used to print a local variable or a function
8431 argument without knowing the frame where this variable or argument
8434 (@value{GDBP}) faas p some_local_var_i_do_not_remember_where_it_is
8437 The @code{faas} command accepts the same options as the @code{frame
8438 apply} command. @xref{frame apply}.
8440 Note that the command @code{tfaas @var{command}} applies @var{command}
8441 on all frames of all threads. See @xref{Threads,,Threads}.
8445 @node Frame Filter Management
8446 @section Management of Frame Filters.
8447 @cindex managing frame filters
8449 Frame filters are Python based utilities to manage and decorate the
8450 output of frames. @xref{Frame Filter API}, for further information.
8452 Managing frame filters is performed by several commands available
8453 within @value{GDBN}, detailed here.
8456 @kindex info frame-filter
8457 @item info frame-filter
8458 Print a list of installed frame filters from all dictionaries, showing
8459 their name, priority and enabled status.
8461 @kindex disable frame-filter
8462 @anchor{disable frame-filter all}
8463 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
8464 Disable a frame filter in the dictionary matching
8465 @var{filter-dictionary} and @var{filter-name}. The
8466 @var{filter-dictionary} may be @code{all}, @code{global},
8467 @code{progspace}, or the name of the object file where the frame filter
8468 dictionary resides. When @code{all} is specified, all frame filters
8469 across all dictionaries are disabled. The @var{filter-name} is the name
8470 of the frame filter and is used when @code{all} is not the option for
8471 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
8472 may be enabled again later.
8474 @kindex enable frame-filter
8475 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
8476 Enable a frame filter in the dictionary matching
8477 @var{filter-dictionary} and @var{filter-name}. The
8478 @var{filter-dictionary} may be @code{all}, @code{global},
8479 @code{progspace} or the name of the object file where the frame filter
8480 dictionary resides. When @code{all} is specified, all frame filters across
8481 all dictionaries are enabled. The @var{filter-name} is the name of the frame
8482 filter and is used when @code{all} is not the option for
8483 @var{filter-dictionary}.
8488 (@value{GDBP}) info frame-filter
8490 global frame-filters:
8491 Priority Enabled Name
8492 1000 No PrimaryFunctionFilter
8495 progspace /build/test frame-filters:
8496 Priority Enabled Name
8497 100 Yes ProgspaceFilter
8499 objfile /build/test frame-filters:
8500 Priority Enabled Name
8501 999 Yes BuildProgramFilter
8503 (@value{GDBP}) disable frame-filter /build/test BuildProgramFilter
8504 (@value{GDBP}) info frame-filter
8506 global frame-filters:
8507 Priority Enabled Name
8508 1000 No PrimaryFunctionFilter
8511 progspace /build/test frame-filters:
8512 Priority Enabled Name
8513 100 Yes ProgspaceFilter
8515 objfile /build/test frame-filters:
8516 Priority Enabled Name
8517 999 No BuildProgramFilter
8519 (@value{GDBP}) enable frame-filter global PrimaryFunctionFilter
8520 (@value{GDBP}) info frame-filter
8522 global frame-filters:
8523 Priority Enabled Name
8524 1000 Yes PrimaryFunctionFilter
8527 progspace /build/test frame-filters:
8528 Priority Enabled Name
8529 100 Yes ProgspaceFilter
8531 objfile /build/test frame-filters:
8532 Priority Enabled Name
8533 999 No BuildProgramFilter
8536 @kindex set frame-filter priority
8537 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
8538 Set the @var{priority} of a frame filter in the dictionary matching
8539 @var{filter-dictionary}, and the frame filter name matching
8540 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8541 @code{progspace} or the name of the object file where the frame filter
8542 dictionary resides. The @var{priority} is an integer.
8544 @kindex show frame-filter priority
8545 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
8546 Show the @var{priority} of a frame filter in the dictionary matching
8547 @var{filter-dictionary}, and the frame filter name matching
8548 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8549 @code{progspace} or the name of the object file where the frame filter
8555 (@value{GDBP}) info frame-filter
8557 global frame-filters:
8558 Priority Enabled Name
8559 1000 Yes PrimaryFunctionFilter
8562 progspace /build/test frame-filters:
8563 Priority Enabled Name
8564 100 Yes ProgspaceFilter
8566 objfile /build/test frame-filters:
8567 Priority Enabled Name
8568 999 No BuildProgramFilter
8570 (@value{GDBP}) set frame-filter priority global Reverse 50
8571 (@value{GDBP}) info frame-filter
8573 global frame-filters:
8574 Priority Enabled Name
8575 1000 Yes PrimaryFunctionFilter
8578 progspace /build/test frame-filters:
8579 Priority Enabled Name
8580 100 Yes ProgspaceFilter
8582 objfile /build/test frame-filters:
8583 Priority Enabled Name
8584 999 No BuildProgramFilter
8589 @chapter Examining Source Files
8591 @value{GDBN} can print parts of your program's source, since the debugging
8592 information recorded in the program tells @value{GDBN} what source files were
8593 used to build it. When your program stops, @value{GDBN} spontaneously prints
8594 the line where it stopped. Likewise, when you select a stack frame
8595 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
8596 execution in that frame has stopped. You can print other portions of
8597 source files by explicit command.
8599 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
8600 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
8601 @value{GDBN} under @sc{gnu} Emacs}.
8604 * List:: Printing source lines
8605 * Specify Location:: How to specify code locations
8606 * Edit:: Editing source files
8607 * Search:: Searching source files
8608 * Source Path:: Specifying source directories
8609 * Machine Code:: Source and machine code
8613 @section Printing Source Lines
8616 @kindex l @r{(@code{list})}
8617 To print lines from a source file, use the @code{list} command
8618 (abbreviated @code{l}). By default, ten lines are printed.
8619 There are several ways to specify what part of the file you want to
8620 print; see @ref{Specify Location}, for the full list.
8622 Here are the forms of the @code{list} command most commonly used:
8625 @item list @var{linenum}
8626 Print lines centered around line number @var{linenum} in the
8627 current source file.
8629 @item list @var{function}
8630 Print lines centered around the beginning of function
8634 Print more lines. If the last lines printed were printed with a
8635 @code{list} command, this prints lines following the last lines
8636 printed; however, if the last line printed was a solitary line printed
8637 as part of displaying a stack frame (@pxref{Stack, ,Examining the
8638 Stack}), this prints lines centered around that line.
8641 Print lines just before the lines last printed.
8644 @cindex @code{list}, how many lines to display
8645 By default, @value{GDBN} prints ten source lines with any of these forms of
8646 the @code{list} command. You can change this using @code{set listsize}:
8649 @kindex set listsize
8650 @item set listsize @var{count}
8651 @itemx set listsize unlimited
8652 Make the @code{list} command display @var{count} source lines (unless
8653 the @code{list} argument explicitly specifies some other number).
8654 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
8656 @kindex show listsize
8658 Display the number of lines that @code{list} prints.
8661 Repeating a @code{list} command with @key{RET} discards the argument,
8662 so it is equivalent to typing just @code{list}. This is more useful
8663 than listing the same lines again. An exception is made for an
8664 argument of @samp{-}; that argument is preserved in repetition so that
8665 each repetition moves up in the source file.
8667 In general, the @code{list} command expects you to supply zero, one or two
8668 @dfn{locations}. Locations specify source lines; there are several ways
8669 of writing them (@pxref{Specify Location}), but the effect is always
8670 to specify some source line.
8672 Here is a complete description of the possible arguments for @code{list}:
8675 @item list @var{location}
8676 Print lines centered around the line specified by @var{location}.
8678 @item list @var{first},@var{last}
8679 Print lines from @var{first} to @var{last}. Both arguments are
8680 locations. When a @code{list} command has two locations, and the
8681 source file of the second location is omitted, this refers to
8682 the same source file as the first location.
8684 @item list ,@var{last}
8685 Print lines ending with @var{last}.
8687 @item list @var{first},
8688 Print lines starting with @var{first}.
8691 Print lines just after the lines last printed.
8694 Print lines just before the lines last printed.
8697 As described in the preceding table.
8700 @node Specify Location
8701 @section Specifying a Location
8702 @cindex specifying location
8704 @cindex source location
8707 * Linespec Locations:: Linespec locations
8708 * Explicit Locations:: Explicit locations
8709 * Address Locations:: Address locations
8712 Several @value{GDBN} commands accept arguments that specify a location
8713 of your program's code. Since @value{GDBN} is a source-level
8714 debugger, a location usually specifies some line in the source code.
8715 Locations may be specified using three different formats:
8716 linespec locations, explicit locations, or address locations.
8718 @node Linespec Locations
8719 @subsection Linespec Locations
8720 @cindex linespec locations
8722 A @dfn{linespec} is a colon-separated list of source location parameters such
8723 as file name, function name, etc. Here are all the different ways of
8724 specifying a linespec:
8728 Specifies the line number @var{linenum} of the current source file.
8731 @itemx +@var{offset}
8732 Specifies the line @var{offset} lines before or after the @dfn{current
8733 line}. For the @code{list} command, the current line is the last one
8734 printed; for the breakpoint commands, this is the line at which
8735 execution stopped in the currently selected @dfn{stack frame}
8736 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
8737 used as the second of the two linespecs in a @code{list} command,
8738 this specifies the line @var{offset} lines up or down from the first
8741 @item @var{filename}:@var{linenum}
8742 Specifies the line @var{linenum} in the source file @var{filename}.
8743 If @var{filename} is a relative file name, then it will match any
8744 source file name with the same trailing components. For example, if
8745 @var{filename} is @samp{gcc/expr.c}, then it will match source file
8746 name of @file{/build/trunk/gcc/expr.c}, but not
8747 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
8749 @item @var{function}
8750 Specifies the line that begins the body of the function @var{function}.
8751 For example, in C, this is the line with the open brace.
8753 By default, in C@t{++} and Ada, @var{function} is interpreted as
8754 specifying all functions named @var{function} in all scopes. For
8755 C@t{++}, this means in all namespaces and classes. For Ada, this
8756 means in all packages.
8758 For example, assuming a program with C@t{++} symbols named
8759 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8760 func}} and @w{@kbd{break B::func}} set a breakpoint on both symbols.
8762 Commands that accept a linespec let you override this with the
8763 @code{-qualified} option. For example, @w{@kbd{break -qualified
8764 func}} sets a breakpoint on a free-function named @code{func} ignoring
8765 any C@t{++} class methods and namespace functions called @code{func}.
8767 @xref{Explicit Locations}.
8769 @item @var{function}:@var{label}
8770 Specifies the line where @var{label} appears in @var{function}.
8772 @item @var{filename}:@var{function}
8773 Specifies the line that begins the body of the function @var{function}
8774 in the file @var{filename}. You only need the file name with a
8775 function name to avoid ambiguity when there are identically named
8776 functions in different source files.
8779 Specifies the line at which the label named @var{label} appears
8780 in the function corresponding to the currently selected stack frame.
8781 If there is no current selected stack frame (for instance, if the inferior
8782 is not running), then @value{GDBN} will not search for a label.
8784 @cindex breakpoint at static probe point
8785 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
8786 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
8787 applications to embed static probes. @xref{Static Probe Points}, for more
8788 information on finding and using static probes. This form of linespec
8789 specifies the location of such a static probe.
8791 If @var{objfile} is given, only probes coming from that shared library
8792 or executable matching @var{objfile} as a regular expression are considered.
8793 If @var{provider} is given, then only probes from that provider are considered.
8794 If several probes match the spec, @value{GDBN} will insert a breakpoint at
8795 each one of those probes.
8798 @node Explicit Locations
8799 @subsection Explicit Locations
8800 @cindex explicit locations
8802 @dfn{Explicit locations} allow the user to directly specify the source
8803 location's parameters using option-value pairs.
8805 Explicit locations are useful when several functions, labels, or
8806 file names have the same name (base name for files) in the program's
8807 sources. In these cases, explicit locations point to the source
8808 line you meant more accurately and unambiguously. Also, using
8809 explicit locations might be faster in large programs.
8811 For example, the linespec @samp{foo:bar} may refer to a function @code{bar}
8812 defined in the file named @file{foo} or the label @code{bar} in a function
8813 named @code{foo}. @value{GDBN} must search either the file system or
8814 the symbol table to know.
8816 The list of valid explicit location options is summarized in the
8820 @item -source @var{filename}
8821 The value specifies the source file name. To differentiate between
8822 files with the same base name, prepend as many directories as is necessary
8823 to uniquely identify the desired file, e.g., @file{foo/bar/baz.c}. Otherwise
8824 @value{GDBN} will use the first file it finds with the given base
8825 name. This option requires the use of either @code{-function} or @code{-line}.
8827 @item -function @var{function}
8828 The value specifies the name of a function. Operations
8829 on function locations unmodified by other options (such as @code{-label}
8830 or @code{-line}) refer to the line that begins the body of the function.
8831 In C, for example, this is the line with the open brace.
8833 By default, in C@t{++} and Ada, @var{function} is interpreted as
8834 specifying all functions named @var{function} in all scopes. For
8835 C@t{++}, this means in all namespaces and classes. For Ada, this
8836 means in all packages.
8838 For example, assuming a program with C@t{++} symbols named
8839 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8840 -function func}} and @w{@kbd{break -function B::func}} set a
8841 breakpoint on both symbols.
8843 You can use the @kbd{-qualified} flag to override this (see below).
8847 This flag makes @value{GDBN} interpret a function name specified with
8848 @kbd{-function} as a complete fully-qualified name.
8850 For example, assuming a C@t{++} program with symbols named
8851 @code{A::B::func} and @code{B::func}, the @w{@kbd{break -qualified
8852 -function B::func}} command sets a breakpoint on @code{B::func}, only.
8854 (Note: the @kbd{-qualified} option can precede a linespec as well
8855 (@pxref{Linespec Locations}), so the particular example above could be
8856 simplified as @w{@kbd{break -qualified B::func}}.)
8858 @item -label @var{label}
8859 The value specifies the name of a label. When the function
8860 name is not specified, the label is searched in the function of the currently
8861 selected stack frame.
8863 @item -line @var{number}
8864 The value specifies a line offset for the location. The offset may either
8865 be absolute (@code{-line 3}) or relative (@code{-line +3}), depending on
8866 the command. When specified without any other options, the line offset is
8867 relative to the current line.
8870 Explicit location options may be abbreviated by omitting any non-unique
8871 trailing characters from the option name, e.g., @w{@kbd{break -s main.c -li 3}}.
8873 @node Address Locations
8874 @subsection Address Locations
8875 @cindex address locations
8877 @dfn{Address locations} indicate a specific program address. They have
8878 the generalized form *@var{address}.
8880 For line-oriented commands, such as @code{list} and @code{edit}, this
8881 specifies a source line that contains @var{address}. For @code{break} and
8882 other breakpoint-oriented commands, this can be used to set breakpoints in
8883 parts of your program which do not have debugging information or
8886 Here @var{address} may be any expression valid in the current working
8887 language (@pxref{Languages, working language}) that specifies a code
8888 address. In addition, as a convenience, @value{GDBN} extends the
8889 semantics of expressions used in locations to cover several situations
8890 that frequently occur during debugging. Here are the various forms
8894 @item @var{expression}
8895 Any expression valid in the current working language.
8897 @item @var{funcaddr}
8898 An address of a function or procedure derived from its name. In C,
8899 C@t{++}, Objective-C, Fortran, minimal, and assembly, this is
8900 simply the function's name @var{function} (and actually a special case
8901 of a valid expression). In Pascal and Modula-2, this is
8902 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
8903 (although the Pascal form also works).
8905 This form specifies the address of the function's first instruction,
8906 before the stack frame and arguments have been set up.
8908 @item '@var{filename}':@var{funcaddr}
8909 Like @var{funcaddr} above, but also specifies the name of the source
8910 file explicitly. This is useful if the name of the function does not
8911 specify the function unambiguously, e.g., if there are several
8912 functions with identical names in different source files.
8916 @section Editing Source Files
8917 @cindex editing source files
8920 @kindex e @r{(@code{edit})}
8921 To edit the lines in a source file, use the @code{edit} command.
8922 The editing program of your choice
8923 is invoked with the current line set to
8924 the active line in the program.
8925 Alternatively, there are several ways to specify what part of the file you
8926 want to print if you want to see other parts of the program:
8929 @item edit @var{location}
8930 Edit the source file specified by @code{location}. Editing starts at
8931 that @var{location}, e.g., at the specified source line of the
8932 specified file. @xref{Specify Location}, for all the possible forms
8933 of the @var{location} argument; here are the forms of the @code{edit}
8934 command most commonly used:
8937 @item edit @var{number}
8938 Edit the current source file with @var{number} as the active line number.
8940 @item edit @var{function}
8941 Edit the file containing @var{function} at the beginning of its definition.
8946 @subsection Choosing your Editor
8947 You can customize @value{GDBN} to use any editor you want
8949 The only restriction is that your editor (say @code{ex}), recognizes the
8950 following command-line syntax:
8952 ex +@var{number} file
8954 The optional numeric value +@var{number} specifies the number of the line in
8955 the file where to start editing.}.
8956 By default, it is @file{@value{EDITOR}}, but you can change this
8957 by setting the environment variable @code{EDITOR} before using
8958 @value{GDBN}. For example, to configure @value{GDBN} to use the
8959 @code{vi} editor, you could use these commands with the @code{sh} shell:
8965 or in the @code{csh} shell,
8967 setenv EDITOR /usr/bin/vi
8972 @section Searching Source Files
8973 @cindex searching source files
8975 There are two commands for searching through the current source file for a
8980 @kindex forward-search
8981 @kindex fo @r{(@code{forward-search})}
8982 @item forward-search @var{regexp}
8983 @itemx search @var{regexp}
8984 The command @samp{forward-search @var{regexp}} checks each line,
8985 starting with the one following the last line listed, for a match for
8986 @var{regexp}. It lists the line that is found. You can use the
8987 synonym @samp{search @var{regexp}} or abbreviate the command name as
8990 @kindex reverse-search
8991 @item reverse-search @var{regexp}
8992 The command @samp{reverse-search @var{regexp}} checks each line, starting
8993 with the one before the last line listed and going backward, for a match
8994 for @var{regexp}. It lists the line that is found. You can abbreviate
8995 this command as @code{rev}.
8999 @section Specifying Source Directories
9002 @cindex directories for source files
9003 Executable programs sometimes do not record the directories of the source
9004 files from which they were compiled, just the names. Even when they do,
9005 the directories could be moved between the compilation and your debugging
9006 session. @value{GDBN} has a list of directories to search for source files;
9007 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
9008 it tries all the directories in the list, in the order they are present
9009 in the list, until it finds a file with the desired name.
9011 For example, suppose an executable references the file
9012 @file{/usr/src/foo-1.0/lib/foo.c}, does not record a compilation
9013 directory, and the @dfn{source path} is @file{/mnt/cross}.
9014 @value{GDBN} would look for the source file in the following
9019 @item @file{/usr/src/foo-1.0/lib/foo.c}
9020 @item @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c}
9021 @item @file{/mnt/cross/foo.c}
9025 If the source file is not present at any of the above locations then
9026 an error is printed. @value{GDBN} does not look up the parts of the
9027 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
9028 Likewise, the subdirectories of the source path are not searched: if
9029 the source path is @file{/mnt/cross}, and the binary refers to
9030 @file{foo.c}, @value{GDBN} would not find it under
9031 @file{/mnt/cross/usr/src/foo-1.0/lib}.
9033 Plain file names, relative file names with leading directories, file
9034 names containing dots, etc.@: are all treated as described above,
9035 except that non-absolute file names are not looked up literally. If
9036 the @dfn{source path} is @file{/mnt/cross}, the source file is
9037 recorded as @file{../lib/foo.c}, and no compilation directory is
9038 recorded, then @value{GDBN} will search in the following locations:
9042 @item @file{/mnt/cross/../lib/foo.c}
9043 @item @file{/mnt/cross/foo.c}
9049 @vindex $cdir@r{, convenience variable}
9050 @vindex $cwd@r{, convenience variable}
9051 @cindex compilation directory
9052 @cindex current directory
9053 @cindex working directory
9054 @cindex directory, current
9055 @cindex directory, compilation
9056 The @dfn{source path} will always include two special entries
9057 @samp{$cdir} and @samp{$cwd}, these refer to the compilation directory
9058 (if one is recorded) and the current working directory respectively.
9060 @samp{$cdir} causes @value{GDBN} to search within the compilation
9061 directory, if one is recorded in the debug information. If no
9062 compilation directory is recorded in the debug information then
9063 @samp{$cdir} is ignored.
9065 @samp{$cwd} is not the same as @samp{.}---the former tracks the
9066 current working directory as it changes during your @value{GDBN}
9067 session, while the latter is immediately expanded to the current
9068 directory at the time you add an entry to the source path.
9070 If a compilation directory is recorded in the debug information, and
9071 @value{GDBN} has not found the source file after the first search
9072 using @dfn{source path}, then @value{GDBN} will combine the
9073 compilation directory and the filename, and then search for the source
9074 file again using the @dfn{source path}.
9076 For example, if the executable records the source file as
9077 @file{/usr/src/foo-1.0/lib/foo.c}, the compilation directory is
9078 recorded as @file{/project/build}, and the @dfn{source path} is
9079 @file{/mnt/cross:$cdir:$cwd} while the current working directory of
9080 the @value{GDBN} session is @file{/home/user}, then @value{GDBN} will
9081 search for the source file in the following locations:
9085 @item @file{/usr/src/foo-1.0/lib/foo.c}
9086 @item @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c}
9087 @item @file{/project/build/usr/src/foo-1.0/lib/foo.c}
9088 @item @file{/home/user/usr/src/foo-1.0/lib/foo.c}
9089 @item @file{/mnt/cross/project/build/usr/src/foo-1.0/lib/foo.c}
9090 @item @file{/project/build/project/build/usr/src/foo-1.0/lib/foo.c}
9091 @item @file{/home/user/project/build/usr/src/foo-1.0/lib/foo.c}
9092 @item @file{/mnt/cross/foo.c}
9093 @item @file{/project/build/foo.c}
9094 @item @file{/home/user/foo.c}
9098 If the file name in the previous example had been recorded in the
9099 executable as a relative path rather than an absolute path, then the
9100 first look up would not have occurred, but all of the remaining steps
9103 When searching for source files on MS-DOS and MS-Windows, where
9104 absolute paths start with a drive letter (e.g.
9105 @file{C:/project/foo.c}), @value{GDBN} will remove the drive letter
9106 from the file name before appending it to a search directory from
9107 @dfn{source path}; for instance if the executable references the
9108 source file @file{C:/project/foo.c} and @dfn{source path} is set to
9109 @file{D:/mnt/cross}, then @value{GDBN} will search in the following
9110 locations for the source file:
9114 @item @file{C:/project/foo.c}
9115 @item @file{D:/mnt/cross/project/foo.c}
9116 @item @file{D:/mnt/cross/foo.c}
9120 Note that the executable search path is @emph{not} used to locate the
9123 Whenever you reset or rearrange the source path, @value{GDBN} clears out
9124 any information it has cached about where source files are found and where
9125 each line is in the file.
9129 When you start @value{GDBN}, its source path includes only @samp{$cdir}
9130 and @samp{$cwd}, in that order.
9131 To add other directories, use the @code{directory} command.
9133 The search path is used to find both program source files and @value{GDBN}
9134 script files (read using the @samp{-command} option and @samp{source} command).
9136 In addition to the source path, @value{GDBN} provides a set of commands
9137 that manage a list of source path substitution rules. A @dfn{substitution
9138 rule} specifies how to rewrite source directories stored in the program's
9139 debug information in case the sources were moved to a different
9140 directory between compilation and debugging. A rule is made of
9141 two strings, the first specifying what needs to be rewritten in
9142 the path, and the second specifying how it should be rewritten.
9143 In @ref{set substitute-path}, we name these two parts @var{from} and
9144 @var{to} respectively. @value{GDBN} does a simple string replacement
9145 of @var{from} with @var{to} at the start of the directory part of the
9146 source file name, and uses that result instead of the original file
9147 name to look up the sources.
9149 Using the previous example, suppose the @file{foo-1.0} tree has been
9150 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
9151 @value{GDBN} to replace @file{/usr/src} in all source path names with
9152 @file{/mnt/cross}. The first lookup will then be
9153 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
9154 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
9155 substitution rule, use the @code{set substitute-path} command
9156 (@pxref{set substitute-path}).
9158 To avoid unexpected substitution results, a rule is applied only if the
9159 @var{from} part of the directory name ends at a directory separator.
9160 For instance, a rule substituting @file{/usr/source} into
9161 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
9162 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
9163 is applied only at the beginning of the directory name, this rule will
9164 not be applied to @file{/root/usr/source/baz.c} either.
9166 In many cases, you can achieve the same result using the @code{directory}
9167 command. However, @code{set substitute-path} can be more efficient in
9168 the case where the sources are organized in a complex tree with multiple
9169 subdirectories. With the @code{directory} command, you need to add each
9170 subdirectory of your project. If you moved the entire tree while
9171 preserving its internal organization, then @code{set substitute-path}
9172 allows you to direct the debugger to all the sources with one single
9175 @code{set substitute-path} is also more than just a shortcut command.
9176 The source path is only used if the file at the original location no
9177 longer exists. On the other hand, @code{set substitute-path} modifies
9178 the debugger behavior to look at the rewritten location instead. So, if
9179 for any reason a source file that is not relevant to your executable is
9180 located at the original location, a substitution rule is the only
9181 method available to point @value{GDBN} at the new location.
9183 @cindex @samp{--with-relocated-sources}
9184 @cindex default source path substitution
9185 You can configure a default source path substitution rule by
9186 configuring @value{GDBN} with the
9187 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
9188 should be the name of a directory under @value{GDBN}'s configured
9189 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
9190 directory names in debug information under @var{dir} will be adjusted
9191 automatically if the installed @value{GDBN} is moved to a new
9192 location. This is useful if @value{GDBN}, libraries or executables
9193 with debug information and corresponding source code are being moved
9197 @item directory @var{dirname} @dots{}
9198 @item dir @var{dirname} @dots{}
9199 Add directory @var{dirname} to the front of the source path. Several
9200 directory names may be given to this command, separated by @samp{:}
9201 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
9202 part of absolute file names) or
9203 whitespace. You may specify a directory that is already in the source
9204 path; this moves it forward, so @value{GDBN} searches it sooner.
9206 The special strings @samp{$cdir} (to refer to the compilation
9207 directory, if one is recorded), and @samp{$cwd} (to refer to the
9208 current working directory) can also be included in the list of
9209 directories @var{dirname}. Though these will already be in the source
9210 path they will be moved forward in the list so @value{GDBN} searches
9214 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
9216 @c RET-repeat for @code{directory} is explicitly disabled, but since
9217 @c repeating it would be a no-op we do not say that. (thanks to RMS)
9219 @item set directories @var{path-list}
9220 @kindex set directories
9221 Set the source path to @var{path-list}.
9222 @samp{$cdir:$cwd} are added if missing.
9224 @item show directories
9225 @kindex show directories
9226 Print the source path: show which directories it contains.
9228 @anchor{set substitute-path}
9229 @item set substitute-path @var{from} @var{to}
9230 @kindex set substitute-path
9231 Define a source path substitution rule, and add it at the end of the
9232 current list of existing substitution rules. If a rule with the same
9233 @var{from} was already defined, then the old rule is also deleted.
9235 For example, if the file @file{/foo/bar/baz.c} was moved to
9236 @file{/mnt/cross/baz.c}, then the command
9239 (@value{GDBP}) set substitute-path /foo/bar /mnt/cross
9243 will tell @value{GDBN} to replace @samp{/foo/bar} with
9244 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
9245 @file{baz.c} even though it was moved.
9247 In the case when more than one substitution rule have been defined,
9248 the rules are evaluated one by one in the order where they have been
9249 defined. The first one matching, if any, is selected to perform
9252 For instance, if we had entered the following commands:
9255 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
9256 (@value{GDBP}) set substitute-path /usr/src /mnt/src
9260 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
9261 @file{/mnt/include/defs.h} by using the first rule. However, it would
9262 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
9263 @file{/mnt/src/lib/foo.c}.
9266 @item unset substitute-path [path]
9267 @kindex unset substitute-path
9268 If a path is specified, search the current list of substitution rules
9269 for a rule that would rewrite that path. Delete that rule if found.
9270 A warning is emitted by the debugger if no rule could be found.
9272 If no path is specified, then all substitution rules are deleted.
9274 @item show substitute-path [path]
9275 @kindex show substitute-path
9276 If a path is specified, then print the source path substitution rule
9277 which would rewrite that path, if any.
9279 If no path is specified, then print all existing source path substitution
9284 If your source path is cluttered with directories that are no longer of
9285 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
9286 versions of source. You can correct the situation as follows:
9290 Use @code{directory} with no argument to reset the source path to its default value.
9293 Use @code{directory} with suitable arguments to reinstall the
9294 directories you want in the source path. You can add all the
9295 directories in one command.
9299 @section Source and Machine Code
9300 @cindex source line and its code address
9302 You can use the command @code{info line} to map source lines to program
9303 addresses (and vice versa), and the command @code{disassemble} to display
9304 a range of addresses as machine instructions. You can use the command
9305 @code{set disassemble-next-line} to set whether to disassemble next
9306 source line when execution stops. When run under @sc{gnu} Emacs
9307 mode, the @code{info line} command causes the arrow to point to the
9308 line specified. Also, @code{info line} prints addresses in symbolic form as
9314 @itemx info line @var{location}
9315 Print the starting and ending addresses of the compiled code for
9316 source line @var{location}. You can specify source lines in any of
9317 the ways documented in @ref{Specify Location}. With no @var{location}
9318 information about the current source line is printed.
9321 For example, we can use @code{info line} to discover the location of
9322 the object code for the first line of function
9323 @code{m4_changequote}:
9326 (@value{GDBP}) info line m4_changequote
9327 Line 895 of "builtin.c" starts at pc 0x634c <m4_changequote> and \
9328 ends at 0x6350 <m4_changequote+4>.
9332 @cindex code address and its source line
9333 We can also inquire (using @code{*@var{addr}} as the form for
9334 @var{location}) what source line covers a particular address:
9336 (@value{GDBP}) info line *0x63ff
9337 Line 926 of "builtin.c" starts at pc 0x63e4 <m4_changequote+152> and \
9338 ends at 0x6404 <m4_changequote+184>.
9341 @cindex @code{$_} and @code{info line}
9342 @cindex @code{x} command, default address
9343 @kindex x@r{(examine), and} info line
9344 After @code{info line}, the default address for the @code{x} command
9345 is changed to the starting address of the line, so that @samp{x/i} is
9346 sufficient to begin examining the machine code (@pxref{Memory,
9347 ,Examining Memory}). Also, this address is saved as the value of the
9348 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
9351 @cindex info line, repeated calls
9352 After @code{info line}, using @code{info line} again without
9353 specifying a location will display information about the next source
9358 @cindex assembly instructions
9359 @cindex instructions, assembly
9360 @cindex machine instructions
9361 @cindex listing machine instructions
9363 @itemx disassemble /m
9364 @itemx disassemble /s
9365 @itemx disassemble /r
9366 This specialized command dumps a range of memory as machine
9367 instructions. It can also print mixed source+disassembly by specifying
9368 the @code{/m} or @code{/s} modifier and print the raw instructions in hex
9369 as well as in symbolic form by specifying the @code{/r} modifier.
9370 The default memory range is the function surrounding the
9371 program counter of the selected frame. A single argument to this
9372 command is a program counter value; @value{GDBN} dumps the function
9373 surrounding this value. When two arguments are given, they should
9374 be separated by a comma, possibly surrounded by whitespace. The
9375 arguments specify a range of addresses to dump, in one of two forms:
9378 @item @var{start},@var{end}
9379 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
9380 @item @var{start},+@var{length}
9381 the addresses from @var{start} (inclusive) to
9382 @code{@var{start}+@var{length}} (exclusive).
9386 When 2 arguments are specified, the name of the function is also
9387 printed (since there could be several functions in the given range).
9389 The argument(s) can be any expression yielding a numeric value, such as
9390 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
9392 If the range of memory being disassembled contains current program counter,
9393 the instruction at that location is shown with a @code{=>} marker.
9396 The following example shows the disassembly of a range of addresses of
9397 HP PA-RISC 2.0 code:
9400 (@value{GDBP}) disas 0x32c4, 0x32e4
9401 Dump of assembler code from 0x32c4 to 0x32e4:
9402 0x32c4 <main+204>: addil 0,dp
9403 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
9404 0x32cc <main+212>: ldil 0x3000,r31
9405 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
9406 0x32d4 <main+220>: ldo 0(r31),rp
9407 0x32d8 <main+224>: addil -0x800,dp
9408 0x32dc <main+228>: ldo 0x588(r1),r26
9409 0x32e0 <main+232>: ldil 0x3000,r31
9410 End of assembler dump.
9413 Here is an example showing mixed source+assembly for Intel x86
9414 with @code{/m} or @code{/s}, when the program is stopped just after
9415 function prologue in a non-optimized function with no inline code.
9418 (@value{GDBP}) disas /m main
9419 Dump of assembler code for function main:
9421 0x08048330 <+0>: push %ebp
9422 0x08048331 <+1>: mov %esp,%ebp
9423 0x08048333 <+3>: sub $0x8,%esp
9424 0x08048336 <+6>: and $0xfffffff0,%esp
9425 0x08048339 <+9>: sub $0x10,%esp
9427 6 printf ("Hello.\n");
9428 => 0x0804833c <+12>: movl $0x8048440,(%esp)
9429 0x08048343 <+19>: call 0x8048284 <puts@@plt>
9433 0x08048348 <+24>: mov $0x0,%eax
9434 0x0804834d <+29>: leave
9435 0x0804834e <+30>: ret
9437 End of assembler dump.
9440 The @code{/m} option is deprecated as its output is not useful when
9441 there is either inlined code or re-ordered code.
9442 The @code{/s} option is the preferred choice.
9443 Here is an example for AMD x86-64 showing the difference between
9444 @code{/m} output and @code{/s} output.
9445 This example has one inline function defined in a header file,
9446 and the code is compiled with @samp{-O2} optimization.
9447 Note how the @code{/m} output is missing the disassembly of
9448 several instructions that are present in the @code{/s} output.
9478 (@value{GDBP}) disas /m main
9479 Dump of assembler code for function main:
9483 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9484 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9488 0x000000000040041d <+29>: xor %eax,%eax
9489 0x000000000040041f <+31>: retq
9490 0x0000000000400420 <+32>: add %eax,%eax
9491 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9493 End of assembler dump.
9494 (@value{GDBP}) disas /s main
9495 Dump of assembler code for function main:
9499 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9503 0x0000000000400406 <+6>: test %eax,%eax
9504 0x0000000000400408 <+8>: js 0x400420 <main+32>
9509 0x000000000040040a <+10>: lea 0xa(%rax),%edx
9510 0x000000000040040d <+13>: test %eax,%eax
9511 0x000000000040040f <+15>: mov $0x1,%eax
9512 0x0000000000400414 <+20>: cmovne %edx,%eax
9516 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9520 0x000000000040041d <+29>: xor %eax,%eax
9521 0x000000000040041f <+31>: retq
9525 0x0000000000400420 <+32>: add %eax,%eax
9526 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9527 End of assembler dump.
9530 Here is another example showing raw instructions in hex for AMD x86-64,
9533 (@value{GDBP}) disas /r 0x400281,+10
9534 Dump of assembler code from 0x400281 to 0x40028b:
9535 0x0000000000400281: 38 36 cmp %dh,(%rsi)
9536 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
9537 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
9538 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
9539 End of assembler dump.
9542 Addresses cannot be specified as a location (@pxref{Specify Location}).
9543 So, for example, if you want to disassemble function @code{bar}
9544 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
9545 and not @samp{disassemble foo.c:bar}.
9547 Some architectures have more than one commonly-used set of instruction
9548 mnemonics or other syntax.
9550 For programs that were dynamically linked and use shared libraries,
9551 instructions that call functions or branch to locations in the shared
9552 libraries might show a seemingly bogus location---it's actually a
9553 location of the relocation table. On some architectures, @value{GDBN}
9554 might be able to resolve these to actual function names.
9557 @kindex set disassembler-options
9558 @cindex disassembler options
9559 @item set disassembler-options @var{option1}[,@var{option2}@dots{}]
9560 This command controls the passing of target specific information to
9561 the disassembler. For a list of valid options, please refer to the
9562 @code{-M}/@code{--disassembler-options} section of the @samp{objdump}
9563 manual and/or the output of @kbd{objdump --help}
9564 (@pxref{objdump,,objdump,binutils,The GNU Binary Utilities}).
9565 The default value is the empty string.
9567 If it is necessary to specify more than one disassembler option, then
9568 multiple options can be placed together into a comma separated list.
9569 Currently this command is only supported on targets ARM, MIPS, PowerPC
9572 @kindex show disassembler-options
9573 @item show disassembler-options
9574 Show the current setting of the disassembler options.
9578 @kindex set disassembly-flavor
9579 @cindex Intel disassembly flavor
9580 @cindex AT&T disassembly flavor
9581 @item set disassembly-flavor @var{instruction-set}
9582 Select the instruction set to use when disassembling the
9583 program via the @code{disassemble} or @code{x/i} commands.
9585 Currently this command is only defined for the Intel x86 family. You
9586 can set @var{instruction-set} to either @code{intel} or @code{att}.
9587 The default is @code{att}, the AT&T flavor used by default by Unix
9588 assemblers for x86-based targets.
9590 @kindex show disassembly-flavor
9591 @item show disassembly-flavor
9592 Show the current setting of the disassembly flavor.
9596 @kindex set disassemble-next-line
9597 @kindex show disassemble-next-line
9598 @item set disassemble-next-line
9599 @itemx show disassemble-next-line
9600 Control whether or not @value{GDBN} will disassemble the next source
9601 line or instruction when execution stops. If ON, @value{GDBN} will
9602 display disassembly of the next source line when execution of the
9603 program being debugged stops. This is @emph{in addition} to
9604 displaying the source line itself, which @value{GDBN} always does if
9605 possible. If the next source line cannot be displayed for some reason
9606 (e.g., if @value{GDBN} cannot find the source file, or there's no line
9607 info in the debug info), @value{GDBN} will display disassembly of the
9608 next @emph{instruction} instead of showing the next source line. If
9609 AUTO, @value{GDBN} will display disassembly of next instruction only
9610 if the source line cannot be displayed. This setting causes
9611 @value{GDBN} to display some feedback when you step through a function
9612 with no line info or whose source file is unavailable. The default is
9613 OFF, which means never display the disassembly of the next line or
9619 @chapter Examining Data
9621 @cindex printing data
9622 @cindex examining data
9625 The usual way to examine data in your program is with the @code{print}
9626 command (abbreviated @code{p}), or its synonym @code{inspect}. It
9627 evaluates and prints the value of an expression of the language your
9628 program is written in (@pxref{Languages, ,Using @value{GDBN} with
9629 Different Languages}). It may also print the expression using a
9630 Python-based pretty-printer (@pxref{Pretty Printing}).
9633 @item print [[@var{options}] --] @var{expr}
9634 @itemx print [[@var{options}] --] /@var{f} @var{expr}
9635 @var{expr} is an expression (in the source language). By default the
9636 value of @var{expr} is printed in a format appropriate to its data type;
9637 you can choose a different format by specifying @samp{/@var{f}}, where
9638 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
9641 @anchor{print options}
9642 The @code{print} command supports a number of options that allow
9643 overriding relevant global print settings as set by @code{set print}
9647 @item -address [@code{on}|@code{off}]
9648 Set printing of addresses.
9649 Related setting: @ref{set print address}.
9651 @item -array [@code{on}|@code{off}]
9652 Pretty formatting of arrays.
9653 Related setting: @ref{set print array}.
9655 @item -array-indexes [@code{on}|@code{off}]
9656 Set printing of array indexes.
9657 Related setting: @ref{set print array-indexes}.
9659 @item -elements @var{number-of-elements}|@code{unlimited}
9660 Set limit on string chars or array elements to print. The value
9661 @code{unlimited} causes there to be no limit. Related setting:
9662 @ref{set print elements}.
9664 @item -max-depth @var{depth}|@code{unlimited}
9665 Set the threshold after which nested structures are replaced with
9666 ellipsis. Related setting: @ref{set print max-depth}.
9668 @item -null-stop [@code{on}|@code{off}]
9669 Set printing of char arrays to stop at first null char. Related
9670 setting: @ref{set print null-stop}.
9672 @item -object [@code{on}|@code{off}]
9673 Set printing C@t{++} virtual function tables. Related setting:
9674 @ref{set print object}.
9676 @item -pretty [@code{on}|@code{off}]
9677 Set pretty formatting of structures. Related setting: @ref{set print
9680 @item -raw-values [@code{on}|@code{off}]
9681 Set whether to print values in raw form, bypassing any
9682 pretty-printers for that value. Related setting: @ref{set print
9685 @item -repeats @var{number-of-repeats}|@code{unlimited}
9686 Set threshold for repeated print elements. @code{unlimited} causes
9687 all elements to be individually printed. Related setting: @ref{set
9690 @item -static-members [@code{on}|@code{off}]
9691 Set printing C@t{++} static members. Related setting: @ref{set print
9694 @item -symbol [@code{on}|@code{off}]
9695 Set printing of symbol names when printing pointers. Related setting:
9696 @ref{set print symbol}.
9698 @item -union [@code{on}|@code{off}]
9699 Set printing of unions interior to structures. Related setting:
9700 @ref{set print union}.
9702 @item -vtbl [@code{on}|@code{off}]
9703 Set printing of C++ virtual function tables. Related setting:
9704 @ref{set print vtbl}.
9707 Because the @code{print} command accepts arbitrary expressions which
9708 may look like options (including abbreviations), if you specify any
9709 command option, then you must use a double dash (@code{--}) to mark
9710 the end of option processing.
9712 For example, this prints the value of the @code{-p} expression:
9715 (@value{GDBP}) print -p
9718 While this repeats the last value in the value history (see below)
9719 with the @code{-pretty} option in effect:
9722 (@value{GDBP}) print -p --
9725 Here is an example including both on option and an expression:
9729 (@value{GDBP}) print -pretty -- *myptr
9741 @item print [@var{options}]
9742 @itemx print [@var{options}] /@var{f}
9743 @cindex reprint the last value
9744 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
9745 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
9746 conveniently inspect the same value in an alternative format.
9749 A more low-level way of examining data is with the @code{x} command.
9750 It examines data in memory at a specified address and prints it in a
9751 specified format. @xref{Memory, ,Examining Memory}.
9753 If you are interested in information about types, or about how the
9754 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
9755 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
9758 @cindex exploring hierarchical data structures
9760 Another way of examining values of expressions and type information is
9761 through the Python extension command @code{explore} (available only if
9762 the @value{GDBN} build is configured with @code{--with-python}). It
9763 offers an interactive way to start at the highest level (or, the most
9764 abstract level) of the data type of an expression (or, the data type
9765 itself) and explore all the way down to leaf scalar values/fields
9766 embedded in the higher level data types.
9769 @item explore @var{arg}
9770 @var{arg} is either an expression (in the source language), or a type
9771 visible in the current context of the program being debugged.
9774 The working of the @code{explore} command can be illustrated with an
9775 example. If a data type @code{struct ComplexStruct} is defined in your
9785 struct ComplexStruct
9787 struct SimpleStruct *ss_p;
9793 followed by variable declarations as
9796 struct SimpleStruct ss = @{ 10, 1.11 @};
9797 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
9801 then, the value of the variable @code{cs} can be explored using the
9802 @code{explore} command as follows.
9805 (@value{GDBP}) explore cs
9806 The value of `cs' is a struct/class of type `struct ComplexStruct' with
9807 the following fields:
9809 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
9810 arr = <Enter 1 to explore this field of type `int [10]'>
9812 Enter the field number of choice:
9816 Since the fields of @code{cs} are not scalar values, you are being
9817 prompted to chose the field you want to explore. Let's say you choose
9818 the field @code{ss_p} by entering @code{0}. Then, since this field is a
9819 pointer, you will be asked if it is pointing to a single value. From
9820 the declaration of @code{cs} above, it is indeed pointing to a single
9821 value, hence you enter @code{y}. If you enter @code{n}, then you will
9822 be asked if it were pointing to an array of values, in which case this
9823 field will be explored as if it were an array.
9826 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
9827 Continue exploring it as a pointer to a single value [y/n]: y
9828 The value of `*(cs.ss_p)' is a struct/class of type `struct
9829 SimpleStruct' with the following fields:
9831 i = 10 .. (Value of type `int')
9832 d = 1.1100000000000001 .. (Value of type `double')
9834 Press enter to return to parent value:
9838 If the field @code{arr} of @code{cs} was chosen for exploration by
9839 entering @code{1} earlier, then since it is as array, you will be
9840 prompted to enter the index of the element in the array that you want
9844 `cs.arr' is an array of `int'.
9845 Enter the index of the element you want to explore in `cs.arr': 5
9847 `(cs.arr)[5]' is a scalar value of type `int'.
9851 Press enter to return to parent value:
9854 In general, at any stage of exploration, you can go deeper towards the
9855 leaf values by responding to the prompts appropriately, or hit the
9856 return key to return to the enclosing data structure (the @i{higher}
9857 level data structure).
9859 Similar to exploring values, you can use the @code{explore} command to
9860 explore types. Instead of specifying a value (which is typically a
9861 variable name or an expression valid in the current context of the
9862 program being debugged), you specify a type name. If you consider the
9863 same example as above, your can explore the type
9864 @code{struct ComplexStruct} by passing the argument
9865 @code{struct ComplexStruct} to the @code{explore} command.
9868 (@value{GDBP}) explore struct ComplexStruct
9872 By responding to the prompts appropriately in the subsequent interactive
9873 session, you can explore the type @code{struct ComplexStruct} in a
9874 manner similar to how the value @code{cs} was explored in the above
9877 The @code{explore} command also has two sub-commands,
9878 @code{explore value} and @code{explore type}. The former sub-command is
9879 a way to explicitly specify that value exploration of the argument is
9880 being invoked, while the latter is a way to explicitly specify that type
9881 exploration of the argument is being invoked.
9884 @item explore value @var{expr}
9885 @cindex explore value
9886 This sub-command of @code{explore} explores the value of the
9887 expression @var{expr} (if @var{expr} is an expression valid in the
9888 current context of the program being debugged). The behavior of this
9889 command is identical to that of the behavior of the @code{explore}
9890 command being passed the argument @var{expr}.
9892 @item explore type @var{arg}
9893 @cindex explore type
9894 This sub-command of @code{explore} explores the type of @var{arg} (if
9895 @var{arg} is a type visible in the current context of program being
9896 debugged), or the type of the value/expression @var{arg} (if @var{arg}
9897 is an expression valid in the current context of the program being
9898 debugged). If @var{arg} is a type, then the behavior of this command is
9899 identical to that of the @code{explore} command being passed the
9900 argument @var{arg}. If @var{arg} is an expression, then the behavior of
9901 this command will be identical to that of the @code{explore} command
9902 being passed the type of @var{arg} as the argument.
9906 * Expressions:: Expressions
9907 * Ambiguous Expressions:: Ambiguous Expressions
9908 * Variables:: Program variables
9909 * Arrays:: Artificial arrays
9910 * Output Formats:: Output formats
9911 * Memory:: Examining memory
9912 * Auto Display:: Automatic display
9913 * Print Settings:: Print settings
9914 * Pretty Printing:: Python pretty printing
9915 * Value History:: Value history
9916 * Convenience Vars:: Convenience variables
9917 * Convenience Funs:: Convenience functions
9918 * Registers:: Registers
9919 * Floating Point Hardware:: Floating point hardware
9920 * Vector Unit:: Vector Unit
9921 * OS Information:: Auxiliary data provided by operating system
9922 * Memory Region Attributes:: Memory region attributes
9923 * Dump/Restore Files:: Copy between memory and a file
9924 * Core File Generation:: Cause a program dump its core
9925 * Character Sets:: Debugging programs that use a different
9926 character set than GDB does
9927 * Caching Target Data:: Data caching for targets
9928 * Searching Memory:: Searching memory for a sequence of bytes
9929 * Value Sizes:: Managing memory allocated for values
9933 @section Expressions
9936 @code{print} and many other @value{GDBN} commands accept an expression and
9937 compute its value. Any kind of constant, variable or operator defined
9938 by the programming language you are using is valid in an expression in
9939 @value{GDBN}. This includes conditional expressions, function calls,
9940 casts, and string constants. It also includes preprocessor macros, if
9941 you compiled your program to include this information; see
9944 @cindex arrays in expressions
9945 @value{GDBN} supports array constants in expressions input by
9946 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
9947 you can use the command @code{print @{1, 2, 3@}} to create an array
9948 of three integers. If you pass an array to a function or assign it
9949 to a program variable, @value{GDBN} copies the array to memory that
9950 is @code{malloc}ed in the target program.
9952 Because C is so widespread, most of the expressions shown in examples in
9953 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
9954 Languages}, for information on how to use expressions in other
9957 In this section, we discuss operators that you can use in @value{GDBN}
9958 expressions regardless of your programming language.
9960 @cindex casts, in expressions
9961 Casts are supported in all languages, not just in C, because it is so
9962 useful to cast a number into a pointer in order to examine a structure
9963 at that address in memory.
9964 @c FIXME: casts supported---Mod2 true?
9966 @value{GDBN} supports these operators, in addition to those common
9967 to programming languages:
9971 @samp{@@} is a binary operator for treating parts of memory as arrays.
9972 @xref{Arrays, ,Artificial Arrays}, for more information.
9975 @samp{::} allows you to specify a variable in terms of the file or
9976 function where it is defined. @xref{Variables, ,Program Variables}.
9978 @cindex @{@var{type}@}
9979 @cindex type casting memory
9980 @cindex memory, viewing as typed object
9981 @cindex casts, to view memory
9982 @item @{@var{type}@} @var{addr}
9983 Refers to an object of type @var{type} stored at address @var{addr} in
9984 memory. The address @var{addr} may be any expression whose value is
9985 an integer or pointer (but parentheses are required around binary
9986 operators, just as in a cast). This construct is allowed regardless
9987 of what kind of data is normally supposed to reside at @var{addr}.
9990 @node Ambiguous Expressions
9991 @section Ambiguous Expressions
9992 @cindex ambiguous expressions
9994 Expressions can sometimes contain some ambiguous elements. For instance,
9995 some programming languages (notably Ada, C@t{++} and Objective-C) permit
9996 a single function name to be defined several times, for application in
9997 different contexts. This is called @dfn{overloading}. Another example
9998 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
9999 templates and is typically instantiated several times, resulting in
10000 the same function name being defined in different contexts.
10002 In some cases and depending on the language, it is possible to adjust
10003 the expression to remove the ambiguity. For instance in C@t{++}, you
10004 can specify the signature of the function you want to break on, as in
10005 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
10006 qualified name of your function often makes the expression unambiguous
10009 When an ambiguity that needs to be resolved is detected, the debugger
10010 has the capability to display a menu of numbered choices for each
10011 possibility, and then waits for the selection with the prompt @samp{>}.
10012 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
10013 aborts the current command. If the command in which the expression was
10014 used allows more than one choice to be selected, the next option in the
10015 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
10018 For example, the following session excerpt shows an attempt to set a
10019 breakpoint at the overloaded symbol @code{String::after}.
10020 We choose three particular definitions of that function name:
10022 @c FIXME! This is likely to change to show arg type lists, at least
10025 (@value{GDBP}) b String::after
10028 [2] file:String.cc; line number:867
10029 [3] file:String.cc; line number:860
10030 [4] file:String.cc; line number:875
10031 [5] file:String.cc; line number:853
10032 [6] file:String.cc; line number:846
10033 [7] file:String.cc; line number:735
10035 Breakpoint 1 at 0xb26c: file String.cc, line 867.
10036 Breakpoint 2 at 0xb344: file String.cc, line 875.
10037 Breakpoint 3 at 0xafcc: file String.cc, line 846.
10038 Multiple breakpoints were set.
10039 Use the "delete" command to delete unwanted
10046 @kindex set multiple-symbols
10047 @item set multiple-symbols @var{mode}
10048 @cindex multiple-symbols menu
10050 This option allows you to adjust the debugger behavior when an expression
10053 By default, @var{mode} is set to @code{all}. If the command with which
10054 the expression is used allows more than one choice, then @value{GDBN}
10055 automatically selects all possible choices. For instance, inserting
10056 a breakpoint on a function using an ambiguous name results in a breakpoint
10057 inserted on each possible match. However, if a unique choice must be made,
10058 then @value{GDBN} uses the menu to help you disambiguate the expression.
10059 For instance, printing the address of an overloaded function will result
10060 in the use of the menu.
10062 When @var{mode} is set to @code{ask}, the debugger always uses the menu
10063 when an ambiguity is detected.
10065 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
10066 an error due to the ambiguity and the command is aborted.
10068 @kindex show multiple-symbols
10069 @item show multiple-symbols
10070 Show the current value of the @code{multiple-symbols} setting.
10074 @section Program Variables
10076 The most common kind of expression to use is the name of a variable
10079 Variables in expressions are understood in the selected stack frame
10080 (@pxref{Selection, ,Selecting a Frame}); they must be either:
10084 global (or file-static)
10091 visible according to the scope rules of the
10092 programming language from the point of execution in that frame
10095 @noindent This means that in the function
10110 you can examine and use the variable @code{a} whenever your program is
10111 executing within the function @code{foo}, but you can only use or
10112 examine the variable @code{b} while your program is executing inside
10113 the block where @code{b} is declared.
10115 @cindex variable name conflict
10116 There is an exception: you can refer to a variable or function whose
10117 scope is a single source file even if the current execution point is not
10118 in this file. But it is possible to have more than one such variable or
10119 function with the same name (in different source files). If that
10120 happens, referring to that name has unpredictable effects. If you wish,
10121 you can specify a static variable in a particular function or file by
10122 using the colon-colon (@code{::}) notation:
10124 @cindex colon-colon, context for variables/functions
10126 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
10127 @cindex @code{::}, context for variables/functions
10130 @var{file}::@var{variable}
10131 @var{function}::@var{variable}
10135 Here @var{file} or @var{function} is the name of the context for the
10136 static @var{variable}. In the case of file names, you can use quotes to
10137 make sure @value{GDBN} parses the file name as a single word---for example,
10138 to print a global value of @code{x} defined in @file{f2.c}:
10141 (@value{GDBP}) p 'f2.c'::x
10144 The @code{::} notation is normally used for referring to
10145 static variables, since you typically disambiguate uses of local variables
10146 in functions by selecting the appropriate frame and using the
10147 simple name of the variable. However, you may also use this notation
10148 to refer to local variables in frames enclosing the selected frame:
10157 process (a); /* Stop here */
10168 For example, if there is a breakpoint at the commented line,
10169 here is what you might see
10170 when the program stops after executing the call @code{bar(0)}:
10175 (@value{GDBP}) p bar::a
10177 (@value{GDBP}) up 2
10178 #2 0x080483d0 in foo (a=5) at foobar.c:12
10181 (@value{GDBP}) p bar::a
10185 @cindex C@t{++} scope resolution
10186 These uses of @samp{::} are very rarely in conflict with the very
10187 similar use of the same notation in C@t{++}. When they are in
10188 conflict, the C@t{++} meaning takes precedence; however, this can be
10189 overridden by quoting the file or function name with single quotes.
10191 For example, suppose the program is stopped in a method of a class
10192 that has a field named @code{includefile}, and there is also an
10193 include file named @file{includefile} that defines a variable,
10194 @code{some_global}.
10197 (@value{GDBP}) p includefile
10199 (@value{GDBP}) p includefile::some_global
10200 A syntax error in expression, near `'.
10201 (@value{GDBP}) p 'includefile'::some_global
10205 @cindex wrong values
10206 @cindex variable values, wrong
10207 @cindex function entry/exit, wrong values of variables
10208 @cindex optimized code, wrong values of variables
10210 @emph{Warning:} Occasionally, a local variable may appear to have the
10211 wrong value at certain points in a function---just after entry to a new
10212 scope, and just before exit.
10214 You may see this problem when you are stepping by machine instructions.
10215 This is because, on most machines, it takes more than one instruction to
10216 set up a stack frame (including local variable definitions); if you are
10217 stepping by machine instructions, variables may appear to have the wrong
10218 values until the stack frame is completely built. On exit, it usually
10219 also takes more than one machine instruction to destroy a stack frame;
10220 after you begin stepping through that group of instructions, local
10221 variable definitions may be gone.
10223 This may also happen when the compiler does significant optimizations.
10224 To be sure of always seeing accurate values, turn off all optimization
10227 @cindex ``No symbol "foo" in current context''
10228 Another possible effect of compiler optimizations is to optimize
10229 unused variables out of existence, or assign variables to registers (as
10230 opposed to memory addresses). Depending on the support for such cases
10231 offered by the debug info format used by the compiler, @value{GDBN}
10232 might not be able to display values for such local variables. If that
10233 happens, @value{GDBN} will print a message like this:
10236 No symbol "foo" in current context.
10239 To solve such problems, either recompile without optimizations, or use a
10240 different debug info format, if the compiler supports several such
10241 formats. @xref{Compilation}, for more information on choosing compiler
10242 options. @xref{C, ,C and C@t{++}}, for more information about debug
10243 info formats that are best suited to C@t{++} programs.
10245 If you ask to print an object whose contents are unknown to
10246 @value{GDBN}, e.g., because its data type is not completely specified
10247 by the debug information, @value{GDBN} will say @samp{<incomplete
10248 type>}. @xref{Symbols, incomplete type}, for more about this.
10250 @cindex no debug info variables
10251 If you try to examine or use the value of a (global) variable for
10252 which @value{GDBN} has no type information, e.g., because the program
10253 includes no debug information, @value{GDBN} displays an error message.
10254 @xref{Symbols, unknown type}, for more about unknown types. If you
10255 cast the variable to its declared type, @value{GDBN} gets the
10256 variable's value using the cast-to type as the variable's type. For
10257 example, in a C program:
10260 (@value{GDBP}) p var
10261 'var' has unknown type; cast it to its declared type
10262 (@value{GDBP}) p (float) var
10266 If you append @kbd{@@entry} string to a function parameter name you get its
10267 value at the time the function got called. If the value is not available an
10268 error message is printed. Entry values are available only with some compilers.
10269 Entry values are normally also printed at the function parameter list according
10270 to @ref{set print entry-values}.
10273 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
10275 (@value{GDBP}) next
10277 (@value{GDBP}) print i
10279 (@value{GDBP}) print i@@entry
10283 Strings are identified as arrays of @code{char} values without specified
10284 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
10285 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
10286 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
10287 defines literal string type @code{"char"} as @code{char} without a sign.
10292 signed char var1[] = "A";
10295 You get during debugging
10297 (@value{GDBP}) print var0
10299 (@value{GDBP}) print var1
10300 $2 = @{65 'A', 0 '\0'@}
10304 @section Artificial Arrays
10306 @cindex artificial array
10308 @kindex @@@r{, referencing memory as an array}
10309 It is often useful to print out several successive objects of the
10310 same type in memory; a section of an array, or an array of
10311 dynamically determined size for which only a pointer exists in the
10314 You can do this by referring to a contiguous span of memory as an
10315 @dfn{artificial array}, using the binary operator @samp{@@}. The left
10316 operand of @samp{@@} should be the first element of the desired array
10317 and be an individual object. The right operand should be the desired length
10318 of the array. The result is an array value whose elements are all of
10319 the type of the left argument. The first element is actually the left
10320 argument; the second element comes from bytes of memory immediately
10321 following those that hold the first element, and so on. Here is an
10322 example. If a program says
10325 int *array = (int *) malloc (len * sizeof (int));
10329 you can print the contents of @code{array} with
10335 The left operand of @samp{@@} must reside in memory. Array values made
10336 with @samp{@@} in this way behave just like other arrays in terms of
10337 subscripting, and are coerced to pointers when used in expressions.
10338 Artificial arrays most often appear in expressions via the value history
10339 (@pxref{Value History, ,Value History}), after printing one out.
10341 Another way to create an artificial array is to use a cast.
10342 This re-interprets a value as if it were an array.
10343 The value need not be in memory:
10345 (@value{GDBP}) p/x (short[2])0x12345678
10346 $1 = @{0x1234, 0x5678@}
10349 As a convenience, if you leave the array length out (as in
10350 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
10351 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
10353 (@value{GDBP}) p/x (short[])0x12345678
10354 $2 = @{0x1234, 0x5678@}
10357 Sometimes the artificial array mechanism is not quite enough; in
10358 moderately complex data structures, the elements of interest may not
10359 actually be adjacent---for example, if you are interested in the values
10360 of pointers in an array. One useful work-around in this situation is
10361 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
10362 Variables}) as a counter in an expression that prints the first
10363 interesting value, and then repeat that expression via @key{RET}. For
10364 instance, suppose you have an array @code{dtab} of pointers to
10365 structures, and you are interested in the values of a field @code{fv}
10366 in each structure. Here is an example of what you might type:
10376 @node Output Formats
10377 @section Output Formats
10379 @cindex formatted output
10380 @cindex output formats
10381 By default, @value{GDBN} prints a value according to its data type. Sometimes
10382 this is not what you want. For example, you might want to print a number
10383 in hex, or a pointer in decimal. Or you might want to view data in memory
10384 at a certain address as a character string or as an instruction. To do
10385 these things, specify an @dfn{output format} when you print a value.
10387 The simplest use of output formats is to say how to print a value
10388 already computed. This is done by starting the arguments of the
10389 @code{print} command with a slash and a format letter. The format
10390 letters supported are:
10394 Regard the bits of the value as an integer, and print the integer in
10398 Print as integer in signed decimal.
10401 Print as integer in unsigned decimal.
10404 Print as integer in octal.
10407 Print as integer in binary. The letter @samp{t} stands for ``two''.
10408 @footnote{@samp{b} cannot be used because these format letters are also
10409 used with the @code{x} command, where @samp{b} stands for ``byte'';
10410 see @ref{Memory,,Examining Memory}.}
10413 @cindex unknown address, locating
10414 @cindex locate address
10415 Print as an address, both absolute in hexadecimal and as an offset from
10416 the nearest preceding symbol. You can use this format used to discover
10417 where (in what function) an unknown address is located:
10420 (@value{GDBP}) p/a 0x54320
10421 $3 = 0x54320 <_initialize_vx+396>
10425 The command @code{info symbol 0x54320} yields similar results.
10426 @xref{Symbols, info symbol}.
10429 Regard as an integer and print it as a character constant. This
10430 prints both the numerical value and its character representation. The
10431 character representation is replaced with the octal escape @samp{\nnn}
10432 for characters outside the 7-bit @sc{ascii} range.
10434 Without this format, @value{GDBN} displays @code{char},
10435 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
10436 constants. Single-byte members of vectors are displayed as integer
10440 Regard the bits of the value as a floating point number and print
10441 using typical floating point syntax.
10444 @cindex printing strings
10445 @cindex printing byte arrays
10446 Regard as a string, if possible. With this format, pointers to single-byte
10447 data are displayed as null-terminated strings and arrays of single-byte data
10448 are displayed as fixed-length strings. Other values are displayed in their
10451 Without this format, @value{GDBN} displays pointers to and arrays of
10452 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
10453 strings. Single-byte members of a vector are displayed as an integer
10457 Like @samp{x} formatting, the value is treated as an integer and
10458 printed as hexadecimal, but leading zeros are printed to pad the value
10459 to the size of the integer type.
10462 @cindex raw printing
10463 Print using the @samp{raw} formatting. By default, @value{GDBN} will
10464 use a Python-based pretty-printer, if one is available (@pxref{Pretty
10465 Printing}). This typically results in a higher-level display of the
10466 value's contents. The @samp{r} format bypasses any Python
10467 pretty-printer which might exist.
10470 For example, to print the program counter in hex (@pxref{Registers}), type
10477 Note that no space is required before the slash; this is because command
10478 names in @value{GDBN} cannot contain a slash.
10480 To reprint the last value in the value history with a different format,
10481 you can use the @code{print} command with just a format and no
10482 expression. For example, @samp{p/x} reprints the last value in hex.
10485 @section Examining Memory
10487 You can use the command @code{x} (for ``examine'') to examine memory in
10488 any of several formats, independently of your program's data types.
10490 @cindex examining memory
10492 @kindex x @r{(examine memory)}
10493 @item x/@var{nfu} @var{addr}
10494 @itemx x @var{addr}
10496 Use the @code{x} command to examine memory.
10499 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
10500 much memory to display and how to format it; @var{addr} is an
10501 expression giving the address where you want to start displaying memory.
10502 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
10503 Several commands set convenient defaults for @var{addr}.
10506 @item @var{n}, the repeat count
10507 The repeat count is a decimal integer; the default is 1. It specifies
10508 how much memory (counting by units @var{u}) to display. If a negative
10509 number is specified, memory is examined backward from @var{addr}.
10510 @c This really is **decimal**; unaffected by 'set radix' as of GDB
10513 @item @var{f}, the display format
10514 The display format is one of the formats used by @code{print}
10515 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
10516 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
10517 The default is @samp{x} (hexadecimal) initially. The default changes
10518 each time you use either @code{x} or @code{print}.
10520 @item @var{u}, the unit size
10521 The unit size is any of
10527 Halfwords (two bytes).
10529 Words (four bytes). This is the initial default.
10531 Giant words (eight bytes).
10534 Each time you specify a unit size with @code{x}, that size becomes the
10535 default unit the next time you use @code{x}. For the @samp{i} format,
10536 the unit size is ignored and is normally not written. For the @samp{s} format,
10537 the unit size defaults to @samp{b}, unless it is explicitly given.
10538 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
10539 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
10540 Note that the results depend on the programming language of the
10541 current compilation unit. If the language is C, the @samp{s}
10542 modifier will use the UTF-16 encoding while @samp{w} will use
10543 UTF-32. The encoding is set by the programming language and cannot
10546 @item @var{addr}, starting display address
10547 @var{addr} is the address where you want @value{GDBN} to begin displaying
10548 memory. The expression need not have a pointer value (though it may);
10549 it is always interpreted as an integer address of a byte of memory.
10550 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
10551 @var{addr} is usually just after the last address examined---but several
10552 other commands also set the default address: @code{info breakpoints} (to
10553 the address of the last breakpoint listed), @code{info line} (to the
10554 starting address of a line), and @code{print} (if you use it to display
10555 a value from memory).
10558 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
10559 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
10560 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
10561 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
10562 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
10564 You can also specify a negative repeat count to examine memory backward
10565 from the given address. For example, @samp{x/-3uh 0x54320} prints three
10566 halfwords (@code{h}) at @code{0x54314}, @code{0x54328}, and @code{0x5431c}.
10568 Since the letters indicating unit sizes are all distinct from the
10569 letters specifying output formats, you do not have to remember whether
10570 unit size or format comes first; either order works. The output
10571 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
10572 (However, the count @var{n} must come first; @samp{wx4} does not work.)
10574 Even though the unit size @var{u} is ignored for the formats @samp{s}
10575 and @samp{i}, you might still want to use a count @var{n}; for example,
10576 @samp{3i} specifies that you want to see three machine instructions,
10577 including any operands. For convenience, especially when used with
10578 the @code{display} command, the @samp{i} format also prints branch delay
10579 slot instructions, if any, beyond the count specified, which immediately
10580 follow the last instruction that is within the count. The command
10581 @code{disassemble} gives an alternative way of inspecting machine
10582 instructions; see @ref{Machine Code,,Source and Machine Code}.
10584 If a negative repeat count is specified for the formats @samp{s} or @samp{i},
10585 the command displays null-terminated strings or instructions before the given
10586 address as many as the absolute value of the given number. For the @samp{i}
10587 format, we use line number information in the debug info to accurately locate
10588 instruction boundaries while disassembling backward. If line info is not
10589 available, the command stops examining memory with an error message.
10591 All the defaults for the arguments to @code{x} are designed to make it
10592 easy to continue scanning memory with minimal specifications each time
10593 you use @code{x}. For example, after you have inspected three machine
10594 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
10595 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
10596 the repeat count @var{n} is used again; the other arguments default as
10597 for successive uses of @code{x}.
10599 When examining machine instructions, the instruction at current program
10600 counter is shown with a @code{=>} marker. For example:
10603 (@value{GDBP}) x/5i $pc-6
10604 0x804837f <main+11>: mov %esp,%ebp
10605 0x8048381 <main+13>: push %ecx
10606 0x8048382 <main+14>: sub $0x4,%esp
10607 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
10608 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
10611 @cindex @code{$_}, @code{$__}, and value history
10612 The addresses and contents printed by the @code{x} command are not saved
10613 in the value history because there is often too much of them and they
10614 would get in the way. Instead, @value{GDBN} makes these values available for
10615 subsequent use in expressions as values of the convenience variables
10616 @code{$_} and @code{$__}. After an @code{x} command, the last address
10617 examined is available for use in expressions in the convenience variable
10618 @code{$_}. The contents of that address, as examined, are available in
10619 the convenience variable @code{$__}.
10621 If the @code{x} command has a repeat count, the address and contents saved
10622 are from the last memory unit printed; this is not the same as the last
10623 address printed if several units were printed on the last line of output.
10625 @anchor{addressable memory unit}
10626 @cindex addressable memory unit
10627 Most targets have an addressable memory unit size of 8 bits. This means
10628 that to each memory address are associated 8 bits of data. Some
10629 targets, however, have other addressable memory unit sizes.
10630 Within @value{GDBN} and this document, the term
10631 @dfn{addressable memory unit} (or @dfn{memory unit} for short) is used
10632 when explicitly referring to a chunk of data of that size. The word
10633 @dfn{byte} is used to refer to a chunk of data of 8 bits, regardless of
10634 the addressable memory unit size of the target. For most systems,
10635 addressable memory unit is a synonym of byte.
10637 @cindex remote memory comparison
10638 @cindex target memory comparison
10639 @cindex verify remote memory image
10640 @cindex verify target memory image
10641 When you are debugging a program running on a remote target machine
10642 (@pxref{Remote Debugging}), you may wish to verify the program's image
10643 in the remote machine's memory against the executable file you
10644 downloaded to the target. Or, on any target, you may want to check
10645 whether the program has corrupted its own read-only sections. The
10646 @code{compare-sections} command is provided for such situations.
10649 @kindex compare-sections
10650 @item compare-sections @r{[}@var{section-name}@r{|}@code{-r}@r{]}
10651 Compare the data of a loadable section @var{section-name} in the
10652 executable file of the program being debugged with the same section in
10653 the target machine's memory, and report any mismatches. With no
10654 arguments, compares all loadable sections. With an argument of
10655 @code{-r}, compares all loadable read-only sections.
10657 Note: for remote targets, this command can be accelerated if the
10658 target supports computing the CRC checksum of a block of memory
10659 (@pxref{qCRC packet}).
10663 @section Automatic Display
10664 @cindex automatic display
10665 @cindex display of expressions
10667 If you find that you want to print the value of an expression frequently
10668 (to see how it changes), you might want to add it to the @dfn{automatic
10669 display list} so that @value{GDBN} prints its value each time your program stops.
10670 Each expression added to the list is given a number to identify it;
10671 to remove an expression from the list, you specify that number.
10672 The automatic display looks like this:
10676 3: bar[5] = (struct hack *) 0x3804
10680 This display shows item numbers, expressions and their current values. As with
10681 displays you request manually using @code{x} or @code{print}, you can
10682 specify the output format you prefer; in fact, @code{display} decides
10683 whether to use @code{print} or @code{x} depending your format
10684 specification---it uses @code{x} if you specify either the @samp{i}
10685 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
10689 @item display @var{expr}
10690 Add the expression @var{expr} to the list of expressions to display
10691 each time your program stops. @xref{Expressions, ,Expressions}.
10693 @code{display} does not repeat if you press @key{RET} again after using it.
10695 @item display/@var{fmt} @var{expr}
10696 For @var{fmt} specifying only a display format and not a size or
10697 count, add the expression @var{expr} to the auto-display list but
10698 arrange to display it each time in the specified format @var{fmt}.
10699 @xref{Output Formats,,Output Formats}.
10701 @item display/@var{fmt} @var{addr}
10702 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
10703 number of units, add the expression @var{addr} as a memory address to
10704 be examined each time your program stops. Examining means in effect
10705 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
10708 For example, @samp{display/i $pc} can be helpful, to see the machine
10709 instruction about to be executed each time execution stops (@samp{$pc}
10710 is a common name for the program counter; @pxref{Registers, ,Registers}).
10713 @kindex delete display
10715 @item undisplay @var{dnums}@dots{}
10716 @itemx delete display @var{dnums}@dots{}
10717 Remove items from the list of expressions to display. Specify the
10718 numbers of the displays that you want affected with the command
10719 argument @var{dnums}. It can be a single display number, one of the
10720 numbers shown in the first field of the @samp{info display} display;
10721 or it could be a range of display numbers, as in @code{2-4}.
10723 @code{undisplay} does not repeat if you press @key{RET} after using it.
10724 (Otherwise you would just get the error @samp{No display number @dots{}}.)
10726 @kindex disable display
10727 @item disable display @var{dnums}@dots{}
10728 Disable the display of item numbers @var{dnums}. A disabled display
10729 item is not printed automatically, but is not forgotten. It may be
10730 enabled again later. Specify the numbers of the displays that you
10731 want affected with the command argument @var{dnums}. It can be a
10732 single display number, one of the numbers shown in the first field of
10733 the @samp{info display} display; or it could be a range of display
10734 numbers, as in @code{2-4}.
10736 @kindex enable display
10737 @item enable display @var{dnums}@dots{}
10738 Enable display of item numbers @var{dnums}. It becomes effective once
10739 again in auto display of its expression, until you specify otherwise.
10740 Specify the numbers of the displays that you want affected with the
10741 command argument @var{dnums}. It can be a single display number, one
10742 of the numbers shown in the first field of the @samp{info display}
10743 display; or it could be a range of display numbers, as in @code{2-4}.
10746 Display the current values of the expressions on the list, just as is
10747 done when your program stops.
10749 @kindex info display
10751 Print the list of expressions previously set up to display
10752 automatically, each one with its item number, but without showing the
10753 values. This includes disabled expressions, which are marked as such.
10754 It also includes expressions which would not be displayed right now
10755 because they refer to automatic variables not currently available.
10758 @cindex display disabled out of scope
10759 If a display expression refers to local variables, then it does not make
10760 sense outside the lexical context for which it was set up. Such an
10761 expression is disabled when execution enters a context where one of its
10762 variables is not defined. For example, if you give the command
10763 @code{display last_char} while inside a function with an argument
10764 @code{last_char}, @value{GDBN} displays this argument while your program
10765 continues to stop inside that function. When it stops elsewhere---where
10766 there is no variable @code{last_char}---the display is disabled
10767 automatically. The next time your program stops where @code{last_char}
10768 is meaningful, you can enable the display expression once again.
10770 @node Print Settings
10771 @section Print Settings
10773 @cindex format options
10774 @cindex print settings
10775 @value{GDBN} provides the following ways to control how arrays, structures,
10776 and symbols are printed.
10779 These settings are useful for debugging programs in any language:
10783 @anchor{set print address}
10784 @item set print address
10785 @itemx set print address on
10786 @cindex print/don't print memory addresses
10787 @value{GDBN} prints memory addresses showing the location of stack
10788 traces, structure values, pointer values, breakpoints, and so forth,
10789 even when it also displays the contents of those addresses. The default
10790 is @code{on}. For example, this is what a stack frame display looks like with
10791 @code{set print address on}:
10796 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
10798 530 if (lquote != def_lquote)
10802 @item set print address off
10803 Do not print addresses when displaying their contents. For example,
10804 this is the same stack frame displayed with @code{set print address off}:
10808 (@value{GDBP}) set print addr off
10810 #0 set_quotes (lq="<<", rq=">>") at input.c:530
10811 530 if (lquote != def_lquote)
10815 You can use @samp{set print address off} to eliminate all machine
10816 dependent displays from the @value{GDBN} interface. For example, with
10817 @code{print address off}, you should get the same text for backtraces on
10818 all machines---whether or not they involve pointer arguments.
10821 @item show print address
10822 Show whether or not addresses are to be printed.
10825 When @value{GDBN} prints a symbolic address, it normally prints the
10826 closest earlier symbol plus an offset. If that symbol does not uniquely
10827 identify the address (for example, it is a name whose scope is a single
10828 source file), you may need to clarify. One way to do this is with
10829 @code{info line}, for example @samp{info line *0x4537}. Alternately,
10830 you can set @value{GDBN} to print the source file and line number when
10831 it prints a symbolic address:
10834 @item set print symbol-filename on
10835 @cindex source file and line of a symbol
10836 @cindex symbol, source file and line
10837 Tell @value{GDBN} to print the source file name and line number of a
10838 symbol in the symbolic form of an address.
10840 @item set print symbol-filename off
10841 Do not print source file name and line number of a symbol. This is the
10844 @item show print symbol-filename
10845 Show whether or not @value{GDBN} will print the source file name and
10846 line number of a symbol in the symbolic form of an address.
10849 Another situation where it is helpful to show symbol filenames and line
10850 numbers is when disassembling code; @value{GDBN} shows you the line
10851 number and source file that corresponds to each instruction.
10853 Also, you may wish to see the symbolic form only if the address being
10854 printed is reasonably close to the closest earlier symbol:
10857 @item set print max-symbolic-offset @var{max-offset}
10858 @itemx set print max-symbolic-offset unlimited
10859 @cindex maximum value for offset of closest symbol
10860 Tell @value{GDBN} to only display the symbolic form of an address if the
10861 offset between the closest earlier symbol and the address is less than
10862 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
10863 to always print the symbolic form of an address if any symbol precedes
10864 it. Zero is equivalent to @code{unlimited}.
10866 @item show print max-symbolic-offset
10867 Ask how large the maximum offset is that @value{GDBN} prints in a
10871 @cindex wild pointer, interpreting
10872 @cindex pointer, finding referent
10873 If you have a pointer and you are not sure where it points, try
10874 @samp{set print symbol-filename on}. Then you can determine the name
10875 and source file location of the variable where it points, using
10876 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
10877 For example, here @value{GDBN} shows that a variable @code{ptt} points
10878 at another variable @code{t}, defined in @file{hi2.c}:
10881 (@value{GDBP}) set print symbol-filename on
10882 (@value{GDBP}) p/a ptt
10883 $4 = 0xe008 <t in hi2.c>
10887 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
10888 does not show the symbol name and filename of the referent, even with
10889 the appropriate @code{set print} options turned on.
10892 You can also enable @samp{/a}-like formatting all the time using
10893 @samp{set print symbol on}:
10895 @anchor{set print symbol}
10897 @item set print symbol on
10898 Tell @value{GDBN} to print the symbol corresponding to an address, if
10901 @item set print symbol off
10902 Tell @value{GDBN} not to print the symbol corresponding to an
10903 address. In this mode, @value{GDBN} will still print the symbol
10904 corresponding to pointers to functions. This is the default.
10906 @item show print symbol
10907 Show whether @value{GDBN} will display the symbol corresponding to an
10911 Other settings control how different kinds of objects are printed:
10914 @anchor{set print array}
10915 @item set print array
10916 @itemx set print array on
10917 @cindex pretty print arrays
10918 Pretty print arrays. This format is more convenient to read,
10919 but uses more space. The default is off.
10921 @item set print array off
10922 Return to compressed format for arrays.
10924 @item show print array
10925 Show whether compressed or pretty format is selected for displaying
10928 @cindex print array indexes
10929 @anchor{set print array-indexes}
10930 @item set print array-indexes
10931 @itemx set print array-indexes on
10932 Print the index of each element when displaying arrays. May be more
10933 convenient to locate a given element in the array or quickly find the
10934 index of a given element in that printed array. The default is off.
10936 @item set print array-indexes off
10937 Stop printing element indexes when displaying arrays.
10939 @item show print array-indexes
10940 Show whether the index of each element is printed when displaying
10943 @anchor{set print elements}
10944 @item set print elements @var{number-of-elements}
10945 @itemx set print elements unlimited
10946 @cindex number of array elements to print
10947 @cindex limit on number of printed array elements
10948 Set a limit on how many elements of an array @value{GDBN} will print.
10949 If @value{GDBN} is printing a large array, it stops printing after it has
10950 printed the number of elements set by the @code{set print elements} command.
10951 This limit also applies to the display of strings.
10952 When @value{GDBN} starts, this limit is set to 200.
10953 Setting @var{number-of-elements} to @code{unlimited} or zero means
10954 that the number of elements to print is unlimited.
10956 @item show print elements
10957 Display the number of elements of a large array that @value{GDBN} will print.
10958 If the number is 0, then the printing is unlimited.
10960 @anchor{set print frame-arguments}
10961 @item set print frame-arguments @var{value}
10962 @kindex set print frame-arguments
10963 @cindex printing frame argument values
10964 @cindex print all frame argument values
10965 @cindex print frame argument values for scalars only
10966 @cindex do not print frame arguments
10967 This command allows to control how the values of arguments are printed
10968 when the debugger prints a frame (@pxref{Frames}). The possible
10973 The values of all arguments are printed.
10976 Print the value of an argument only if it is a scalar. The value of more
10977 complex arguments such as arrays, structures, unions, etc, is replaced
10978 by @code{@dots{}}. This is the default. Here is an example where
10979 only scalar arguments are shown:
10982 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
10987 None of the argument values are printed. Instead, the value of each argument
10988 is replaced by @code{@dots{}}. In this case, the example above now becomes:
10991 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
10996 Only the presence of arguments is indicated by @code{@dots{}}.
10997 The @code{@dots{}} are not printed for function without any arguments.
10998 None of the argument names and values are printed.
10999 In this case, the example above now becomes:
11002 #1 0x08048361 in call_me (@dots{}) at frame-args.c:23
11007 By default, only scalar arguments are printed. This command can be used
11008 to configure the debugger to print the value of all arguments, regardless
11009 of their type. However, it is often advantageous to not print the value
11010 of more complex parameters. For instance, it reduces the amount of
11011 information printed in each frame, making the backtrace more readable.
11012 Also, it improves performance when displaying Ada frames, because
11013 the computation of large arguments can sometimes be CPU-intensive,
11014 especially in large applications. Setting @code{print frame-arguments}
11015 to @code{scalars} (the default), @code{none} or @code{presence} avoids
11016 this computation, thus speeding up the display of each Ada frame.
11018 @item show print frame-arguments
11019 Show how the value of arguments should be displayed when printing a frame.
11021 @anchor{set print raw-frame-arguments}
11022 @item set print raw-frame-arguments on
11023 Print frame arguments in raw, non pretty-printed, form.
11025 @item set print raw-frame-arguments off
11026 Print frame arguments in pretty-printed form, if there is a pretty-printer
11027 for the value (@pxref{Pretty Printing}),
11028 otherwise print the value in raw form.
11029 This is the default.
11031 @item show print raw-frame-arguments
11032 Show whether to print frame arguments in raw form.
11034 @anchor{set print entry-values}
11035 @item set print entry-values @var{value}
11036 @kindex set print entry-values
11037 Set printing of frame argument values at function entry. In some cases
11038 @value{GDBN} can determine the value of function argument which was passed by
11039 the function caller, even if the value was modified inside the called function
11040 and therefore is different. With optimized code, the current value could be
11041 unavailable, but the entry value may still be known.
11043 The default value is @code{default} (see below for its description). Older
11044 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
11045 this feature will behave in the @code{default} setting the same way as with the
11048 This functionality is currently supported only by DWARF 2 debugging format and
11049 the compiler has to produce @samp{DW_TAG_call_site} tags. With
11050 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
11053 The @var{value} parameter can be one of the following:
11057 Print only actual parameter values, never print values from function entry
11061 #0 different (val=6)
11062 #0 lost (val=<optimized out>)
11064 #0 invalid (val=<optimized out>)
11068 Print only parameter values from function entry point. The actual parameter
11069 values are never printed.
11071 #0 equal (val@@entry=5)
11072 #0 different (val@@entry=5)
11073 #0 lost (val@@entry=5)
11074 #0 born (val@@entry=<optimized out>)
11075 #0 invalid (val@@entry=<optimized out>)
11079 Print only parameter values from function entry point. If value from function
11080 entry point is not known while the actual value is known, print the actual
11081 value for such parameter.
11083 #0 equal (val@@entry=5)
11084 #0 different (val@@entry=5)
11085 #0 lost (val@@entry=5)
11087 #0 invalid (val@@entry=<optimized out>)
11091 Print actual parameter values. If actual parameter value is not known while
11092 value from function entry point is known, print the entry point value for such
11096 #0 different (val=6)
11097 #0 lost (val@@entry=5)
11099 #0 invalid (val=<optimized out>)
11103 Always print both the actual parameter value and its value from function entry
11104 point, even if values of one or both are not available due to compiler
11107 #0 equal (val=5, val@@entry=5)
11108 #0 different (val=6, val@@entry=5)
11109 #0 lost (val=<optimized out>, val@@entry=5)
11110 #0 born (val=10, val@@entry=<optimized out>)
11111 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
11115 Print the actual parameter value if it is known and also its value from
11116 function entry point if it is known. If neither is known, print for the actual
11117 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
11118 values are known and identical, print the shortened
11119 @code{param=param@@entry=VALUE} notation.
11121 #0 equal (val=val@@entry=5)
11122 #0 different (val=6, val@@entry=5)
11123 #0 lost (val@@entry=5)
11125 #0 invalid (val=<optimized out>)
11129 Always print the actual parameter value. Print also its value from function
11130 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
11131 if both values are known and identical, print the shortened
11132 @code{param=param@@entry=VALUE} notation.
11134 #0 equal (val=val@@entry=5)
11135 #0 different (val=6, val@@entry=5)
11136 #0 lost (val=<optimized out>, val@@entry=5)
11138 #0 invalid (val=<optimized out>)
11142 For analysis messages on possible failures of frame argument values at function
11143 entry resolution see @ref{set debug entry-values}.
11145 @item show print entry-values
11146 Show the method being used for printing of frame argument values at function
11149 @anchor{set print frame-info}
11150 @item set print frame-info @var{value}
11151 @kindex set print frame-info
11152 @cindex printing frame information
11153 @cindex frame information, printing
11154 This command allows to control the information printed when
11155 the debugger prints a frame. See @ref{Frames}, @ref{Backtrace},
11156 for a general explanation about frames and frame information.
11157 Note that some other settings (such as @code{set print frame-arguments}
11158 and @code{set print address}) are also influencing if and how some frame
11159 information is displayed. In particular, the frame program counter is never
11160 printed if @code{set print address} is off.
11162 The possible values for @code{set print frame-info} are:
11164 @item short-location
11165 Print the frame level, the program counter (if not at the
11166 beginning of the location source line), the function, the function
11169 Same as @code{short-location} but also print the source file and source line
11171 @item location-and-address
11172 Same as @code{location} but print the program counter even if located at the
11173 beginning of the location source line.
11175 Print the program counter (if not at the beginning of the location
11176 source line), the line number and the source line.
11177 @item source-and-location
11178 Print what @code{location} and @code{source-line} are printing.
11180 The information printed for a frame is decided automatically
11181 by the @value{GDBN} command that prints a frame.
11182 For example, @code{frame} prints the information printed by
11183 @code{source-and-location} while @code{stepi} will switch between
11184 @code{source-line} and @code{source-and-location} depending on the program
11186 The default value is @code{auto}.
11189 @anchor{set print repeats}
11190 @item set print repeats @var{number-of-repeats}
11191 @itemx set print repeats unlimited
11192 @cindex repeated array elements
11193 Set the threshold for suppressing display of repeated array
11194 elements. When the number of consecutive identical elements of an
11195 array exceeds the threshold, @value{GDBN} prints the string
11196 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
11197 identical repetitions, instead of displaying the identical elements
11198 themselves. Setting the threshold to @code{unlimited} or zero will
11199 cause all elements to be individually printed. The default threshold
11202 @item show print repeats
11203 Display the current threshold for printing repeated identical
11206 @anchor{set print max-depth}
11207 @item set print max-depth @var{depth}
11208 @item set print max-depth unlimited
11209 @cindex printing nested structures
11210 Set the threshold after which nested structures are replaced with
11211 ellipsis, this can make visualising deeply nested structures easier.
11213 For example, given this C code
11216 typedef struct s1 @{ int a; @} s1;
11217 typedef struct s2 @{ s1 b; @} s2;
11218 typedef struct s3 @{ s2 c; @} s3;
11219 typedef struct s4 @{ s3 d; @} s4;
11221 s4 var = @{ @{ @{ @{ 3 @} @} @} @};
11224 The following table shows how different values of @var{depth} will
11225 effect how @code{var} is printed by @value{GDBN}:
11227 @multitable @columnfractions .3 .7
11228 @headitem @var{depth} setting @tab Result of @samp{p var}
11230 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
11232 @tab @code{$1 = @{...@}}
11234 @tab @code{$1 = @{d = @{...@}@}}
11236 @tab @code{$1 = @{d = @{c = @{...@}@}@}}
11238 @tab @code{$1 = @{d = @{c = @{b = @{...@}@}@}@}}
11240 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
11243 To see the contents of structures that have been hidden the user can
11244 either increase the print max-depth, or they can print the elements of
11245 the structure that are visible, for example
11248 (@value{GDBP}) set print max-depth 2
11249 (@value{GDBP}) p var
11250 $1 = @{d = @{c = @{...@}@}@}
11251 (@value{GDBP}) p var.d
11252 $2 = @{c = @{b = @{...@}@}@}
11253 (@value{GDBP}) p var.d.c
11254 $3 = @{b = @{a = 3@}@}
11257 The pattern used to replace nested structures varies based on
11258 language, for most languages @code{@{...@}} is used, but Fortran uses
11261 @item show print max-depth
11262 Display the current threshold after which nested structures are
11263 replaces with ellipsis.
11265 @anchor{set print null-stop}
11266 @item set print null-stop
11267 @cindex @sc{null} elements in arrays
11268 Cause @value{GDBN} to stop printing the characters of an array when the first
11269 @sc{null} is encountered. This is useful when large arrays actually
11270 contain only short strings.
11271 The default is off.
11273 @item show print null-stop
11274 Show whether @value{GDBN} stops printing an array on the first
11275 @sc{null} character.
11277 @anchor{set print pretty}
11278 @item set print pretty on
11279 @cindex print structures in indented form
11280 @cindex indentation in structure display
11281 Cause @value{GDBN} to print structures in an indented format with one member
11282 per line, like this:
11297 @item set print pretty off
11298 Cause @value{GDBN} to print structures in a compact format, like this:
11302 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
11303 meat = 0x54 "Pork"@}
11308 This is the default format.
11310 @item show print pretty
11311 Show which format @value{GDBN} is using to print structures.
11313 @anchor{set print raw-values}
11314 @item set print raw-values on
11315 Print values in raw form, without applying the pretty
11316 printers for the value.
11318 @item set print raw-values off
11319 Print values in pretty-printed form, if there is a pretty-printer
11320 for the value (@pxref{Pretty Printing}),
11321 otherwise print the value in raw form.
11323 The default setting is ``off''.
11325 @item show print raw-values
11326 Show whether to print values in raw form.
11328 @item set print sevenbit-strings on
11329 @cindex eight-bit characters in strings
11330 @cindex octal escapes in strings
11331 Print using only seven-bit characters; if this option is set,
11332 @value{GDBN} displays any eight-bit characters (in strings or
11333 character values) using the notation @code{\}@var{nnn}. This setting is
11334 best if you are working in English (@sc{ascii}) and you use the
11335 high-order bit of characters as a marker or ``meta'' bit.
11337 @item set print sevenbit-strings off
11338 Print full eight-bit characters. This allows the use of more
11339 international character sets, and is the default.
11341 @item show print sevenbit-strings
11342 Show whether or not @value{GDBN} is printing only seven-bit characters.
11344 @anchor{set print union}
11345 @item set print union on
11346 @cindex unions in structures, printing
11347 Tell @value{GDBN} to print unions which are contained in structures
11348 and other unions. This is the default setting.
11350 @item set print union off
11351 Tell @value{GDBN} not to print unions which are contained in
11352 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
11355 @item show print union
11356 Ask @value{GDBN} whether or not it will print unions which are contained in
11357 structures and other unions.
11359 For example, given the declarations
11362 typedef enum @{Tree, Bug@} Species;
11363 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
11364 typedef enum @{Caterpillar, Cocoon, Butterfly@}
11375 struct thing foo = @{Tree, @{Acorn@}@};
11379 with @code{set print union on} in effect @samp{p foo} would print
11382 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
11386 and with @code{set print union off} in effect it would print
11389 $1 = @{it = Tree, form = @{...@}@}
11393 @code{set print union} affects programs written in C-like languages
11399 These settings are of interest when debugging C@t{++} programs:
11402 @cindex demangling C@t{++} names
11403 @item set print demangle
11404 @itemx set print demangle on
11405 Print C@t{++} names in their source form rather than in the encoded
11406 (``mangled'') form passed to the assembler and linker for type-safe
11407 linkage. The default is on.
11409 @item show print demangle
11410 Show whether C@t{++} names are printed in mangled or demangled form.
11412 @item set print asm-demangle
11413 @itemx set print asm-demangle on
11414 Print C@t{++} names in their source form rather than their mangled form, even
11415 in assembler code printouts such as instruction disassemblies.
11416 The default is off.
11418 @item show print asm-demangle
11419 Show whether C@t{++} names in assembly listings are printed in mangled
11422 @cindex C@t{++} symbol decoding style
11423 @cindex symbol decoding style, C@t{++}
11424 @kindex set demangle-style
11425 @item set demangle-style @var{style}
11426 Choose among several encoding schemes used by different compilers to represent
11427 C@t{++} names. If you omit @var{style}, you will see a list of possible
11428 formats. The default value is @var{auto}, which lets @value{GDBN} choose a
11429 decoding style by inspecting your program.
11431 @item show demangle-style
11432 Display the encoding style currently in use for decoding C@t{++} symbols.
11434 @anchor{set print object}
11435 @item set print object
11436 @itemx set print object on
11437 @cindex derived type of an object, printing
11438 @cindex display derived types
11439 When displaying a pointer to an object, identify the @emph{actual}
11440 (derived) type of the object rather than the @emph{declared} type, using
11441 the virtual function table. Note that the virtual function table is
11442 required---this feature can only work for objects that have run-time
11443 type identification; a single virtual method in the object's declared
11444 type is sufficient. Note that this setting is also taken into account when
11445 working with variable objects via MI (@pxref{GDB/MI}).
11447 @item set print object off
11448 Display only the declared type of objects, without reference to the
11449 virtual function table. This is the default setting.
11451 @item show print object
11452 Show whether actual, or declared, object types are displayed.
11454 @anchor{set print static-members}
11455 @item set print static-members
11456 @itemx set print static-members on
11457 @cindex static members of C@t{++} objects
11458 Print static members when displaying a C@t{++} object. The default is on.
11460 @item set print static-members off
11461 Do not print static members when displaying a C@t{++} object.
11463 @item show print static-members
11464 Show whether C@t{++} static members are printed or not.
11466 @item set print pascal_static-members
11467 @itemx set print pascal_static-members on
11468 @cindex static members of Pascal objects
11469 @cindex Pascal objects, static members display
11470 Print static members when displaying a Pascal object. The default is on.
11472 @item set print pascal_static-members off
11473 Do not print static members when displaying a Pascal object.
11475 @item show print pascal_static-members
11476 Show whether Pascal static members are printed or not.
11478 @c These don't work with HP ANSI C++ yet.
11479 @anchor{set print vtbl}
11480 @item set print vtbl
11481 @itemx set print vtbl on
11482 @cindex pretty print C@t{++} virtual function tables
11483 @cindex virtual functions (C@t{++}) display
11484 @cindex VTBL display
11485 Pretty print C@t{++} virtual function tables. The default is off.
11486 (The @code{vtbl} commands do not work on programs compiled with the HP
11487 ANSI C@t{++} compiler (@code{aCC}).)
11489 @item set print vtbl off
11490 Do not pretty print C@t{++} virtual function tables.
11492 @item show print vtbl
11493 Show whether C@t{++} virtual function tables are pretty printed, or not.
11496 @node Pretty Printing
11497 @section Pretty Printing
11499 @value{GDBN} provides a mechanism to allow pretty-printing of values using
11500 Python code. It greatly simplifies the display of complex objects. This
11501 mechanism works for both MI and the CLI.
11504 * Pretty-Printer Introduction:: Introduction to pretty-printers
11505 * Pretty-Printer Example:: An example pretty-printer
11506 * Pretty-Printer Commands:: Pretty-printer commands
11509 @node Pretty-Printer Introduction
11510 @subsection Pretty-Printer Introduction
11512 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
11513 registered for the value. If there is then @value{GDBN} invokes the
11514 pretty-printer to print the value. Otherwise the value is printed normally.
11516 Pretty-printers are normally named. This makes them easy to manage.
11517 The @samp{info pretty-printer} command will list all the installed
11518 pretty-printers with their names.
11519 If a pretty-printer can handle multiple data types, then its
11520 @dfn{subprinters} are the printers for the individual data types.
11521 Each such subprinter has its own name.
11522 The format of the name is @var{printer-name};@var{subprinter-name}.
11524 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
11525 Typically they are automatically loaded and registered when the corresponding
11526 debug information is loaded, thus making them available without having to
11527 do anything special.
11529 There are three places where a pretty-printer can be registered.
11533 Pretty-printers registered globally are available when debugging
11537 Pretty-printers registered with a program space are available only
11538 when debugging that program.
11539 @xref{Progspaces In Python}, for more details on program spaces in Python.
11542 Pretty-printers registered with an objfile are loaded and unloaded
11543 with the corresponding objfile (e.g., shared library).
11544 @xref{Objfiles In Python}, for more details on objfiles in Python.
11547 @xref{Selecting Pretty-Printers}, for further information on how
11548 pretty-printers are selected,
11550 @xref{Writing a Pretty-Printer}, for implementing pretty printers
11553 @node Pretty-Printer Example
11554 @subsection Pretty-Printer Example
11556 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
11559 (@value{GDBP}) print s
11561 static npos = 4294967295,
11563 <std::allocator<char>> = @{
11564 <__gnu_cxx::new_allocator<char>> = @{
11565 <No data fields>@}, <No data fields>
11567 members of std::basic_string<char, std::char_traits<char>,
11568 std::allocator<char> >::_Alloc_hider:
11569 _M_p = 0x804a014 "abcd"
11574 With a pretty-printer for @code{std::string} only the contents are printed:
11577 (@value{GDBP}) print s
11581 @node Pretty-Printer Commands
11582 @subsection Pretty-Printer Commands
11583 @cindex pretty-printer commands
11586 @kindex info pretty-printer
11587 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11588 Print the list of installed pretty-printers.
11589 This includes disabled pretty-printers, which are marked as such.
11591 @var{object-regexp} is a regular expression matching the objects
11592 whose pretty-printers to list.
11593 Objects can be @code{global}, the program space's file
11594 (@pxref{Progspaces In Python}),
11595 and the object files within that program space (@pxref{Objfiles In Python}).
11596 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
11597 looks up a printer from these three objects.
11599 @var{name-regexp} is a regular expression matching the name of the printers
11602 @kindex disable pretty-printer
11603 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11604 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11605 A disabled pretty-printer is not forgotten, it may be enabled again later.
11607 @kindex enable pretty-printer
11608 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11609 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11614 Suppose we have three pretty-printers installed: one from library1.so
11615 named @code{foo} that prints objects of type @code{foo}, and
11616 another from library2.so named @code{bar} that prints two types of objects,
11617 @code{bar1} and @code{bar2}.
11620 (@value{GDBP}) info pretty-printer
11627 (@value{GDBP}) info pretty-printer library2
11632 (@value{GDBP}) disable pretty-printer library1
11634 2 of 3 printers enabled
11635 (@value{GDBP}) info pretty-printer
11642 (@value{GDBP}) disable pretty-printer library2 bar;bar1
11644 1 of 3 printers enabled
11645 (@value{GDBP}) info pretty-printer library2
11652 (@value{GDBP}) disable pretty-printer library2 bar
11654 0 of 3 printers enabled
11655 (@value{GDBP}) info pretty-printer library2
11664 Note that for @code{bar} the entire printer can be disabled,
11665 as can each individual subprinter.
11667 Printing values and frame arguments is done by default using
11668 the enabled pretty printers.
11670 The print option @code{-raw-values} and @value{GDBN} setting
11671 @code{set print raw-values} (@pxref{set print raw-values}) can be
11672 used to print values without applying the enabled pretty printers.
11674 Similarly, the backtrace option @code{-raw-frame-arguments} and
11675 @value{GDBN} setting @code{set print raw-frame-arguments}
11676 (@pxref{set print raw-frame-arguments}) can be used to ignore the
11677 enabled pretty printers when printing frame argument values.
11679 @node Value History
11680 @section Value History
11682 @cindex value history
11683 @cindex history of values printed by @value{GDBN}
11684 Values printed by the @code{print} command are saved in the @value{GDBN}
11685 @dfn{value history}. This allows you to refer to them in other expressions.
11686 Values are kept until the symbol table is re-read or discarded
11687 (for example with the @code{file} or @code{symbol-file} commands).
11688 When the symbol table changes, the value history is discarded,
11689 since the values may contain pointers back to the types defined in the
11694 @cindex history number
11695 The values printed are given @dfn{history numbers} by which you can
11696 refer to them. These are successive integers starting with one.
11697 @code{print} shows you the history number assigned to a value by
11698 printing @samp{$@var{num} = } before the value; here @var{num} is the
11701 To refer to any previous value, use @samp{$} followed by the value's
11702 history number. The way @code{print} labels its output is designed to
11703 remind you of this. Just @code{$} refers to the most recent value in
11704 the history, and @code{$$} refers to the value before that.
11705 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
11706 is the value just prior to @code{$$}, @code{$$1} is equivalent to
11707 @code{$$}, and @code{$$0} is equivalent to @code{$}.
11709 For example, suppose you have just printed a pointer to a structure and
11710 want to see the contents of the structure. It suffices to type
11716 If you have a chain of structures where the component @code{next} points
11717 to the next one, you can print the contents of the next one with this:
11724 You can print successive links in the chain by repeating this
11725 command---which you can do by just typing @key{RET}.
11727 Note that the history records values, not expressions. If the value of
11728 @code{x} is 4 and you type these commands:
11736 then the value recorded in the value history by the @code{print} command
11737 remains 4 even though the value of @code{x} has changed.
11740 @kindex show values
11742 Print the last ten values in the value history, with their item numbers.
11743 This is like @samp{p@ $$9} repeated ten times, except that @code{show
11744 values} does not change the history.
11746 @item show values @var{n}
11747 Print ten history values centered on history item number @var{n}.
11749 @item show values +
11750 Print ten history values just after the values last printed. If no more
11751 values are available, @code{show values +} produces no display.
11754 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
11755 same effect as @samp{show values +}.
11757 @node Convenience Vars
11758 @section Convenience Variables
11760 @cindex convenience variables
11761 @cindex user-defined variables
11762 @value{GDBN} provides @dfn{convenience variables} that you can use within
11763 @value{GDBN} to hold on to a value and refer to it later. These variables
11764 exist entirely within @value{GDBN}; they are not part of your program, and
11765 setting a convenience variable has no direct effect on further execution
11766 of your program. That is why you can use them freely.
11768 Convenience variables are prefixed with @samp{$}. Any name preceded by
11769 @samp{$} can be used for a convenience variable, unless it is one of
11770 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
11771 (Value history references, in contrast, are @emph{numbers} preceded
11772 by @samp{$}. @xref{Value History, ,Value History}.)
11774 You can save a value in a convenience variable with an assignment
11775 expression, just as you would set a variable in your program.
11779 set $foo = *object_ptr
11783 would save in @code{$foo} the value contained in the object pointed to by
11786 Using a convenience variable for the first time creates it, but its
11787 value is @code{void} until you assign a new value. You can alter the
11788 value with another assignment at any time.
11790 Convenience variables have no fixed types. You can assign a convenience
11791 variable any type of value, including structures and arrays, even if
11792 that variable already has a value of a different type. The convenience
11793 variable, when used as an expression, has the type of its current value.
11796 @kindex show convenience
11797 @cindex show all user variables and functions
11798 @item show convenience
11799 Print a list of convenience variables used so far, and their values,
11800 as well as a list of the convenience functions.
11801 Abbreviated @code{show conv}.
11803 @kindex init-if-undefined
11804 @cindex convenience variables, initializing
11805 @item init-if-undefined $@var{variable} = @var{expression}
11806 Set a convenience variable if it has not already been set. This is useful
11807 for user-defined commands that keep some state. It is similar, in concept,
11808 to using local static variables with initializers in C (except that
11809 convenience variables are global). It can also be used to allow users to
11810 override default values used in a command script.
11812 If the variable is already defined then the expression is not evaluated so
11813 any side-effects do not occur.
11816 One of the ways to use a convenience variable is as a counter to be
11817 incremented or a pointer to be advanced. For example, to print
11818 a field from successive elements of an array of structures:
11822 print bar[$i++]->contents
11826 Repeat that command by typing @key{RET}.
11828 Some convenience variables are created automatically by @value{GDBN} and given
11829 values likely to be useful.
11832 @vindex $_@r{, convenience variable}
11834 The variable @code{$_} is automatically set by the @code{x} command to
11835 the last address examined (@pxref{Memory, ,Examining Memory}). Other
11836 commands which provide a default address for @code{x} to examine also
11837 set @code{$_} to that address; these commands include @code{info line}
11838 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
11839 except when set by the @code{x} command, in which case it is a pointer
11840 to the type of @code{$__}.
11842 @vindex $__@r{, convenience variable}
11844 The variable @code{$__} is automatically set by the @code{x} command
11845 to the value found in the last address examined. Its type is chosen
11846 to match the format in which the data was printed.
11849 @vindex $_exitcode@r{, convenience variable}
11850 When the program being debugged terminates normally, @value{GDBN}
11851 automatically sets this variable to the exit code of the program, and
11852 resets @code{$_exitsignal} to @code{void}.
11855 @vindex $_exitsignal@r{, convenience variable}
11856 When the program being debugged dies due to an uncaught signal,
11857 @value{GDBN} automatically sets this variable to that signal's number,
11858 and resets @code{$_exitcode} to @code{void}.
11860 To distinguish between whether the program being debugged has exited
11861 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
11862 @code{$_exitsignal} is not @code{void}), the convenience function
11863 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
11864 Functions}). For example, considering the following source code:
11867 #include <signal.h>
11870 main (int argc, char *argv[])
11877 A valid way of telling whether the program being debugged has exited
11878 or signalled would be:
11881 (@value{GDBP}) define has_exited_or_signalled
11882 Type commands for definition of ``has_exited_or_signalled''.
11883 End with a line saying just ``end''.
11884 >if $_isvoid ($_exitsignal)
11885 >echo The program has exited\n
11887 >echo The program has signalled\n
11893 Program terminated with signal SIGALRM, Alarm clock.
11894 The program no longer exists.
11895 (@value{GDBP}) has_exited_or_signalled
11896 The program has signalled
11899 As can be seen, @value{GDBN} correctly informs that the program being
11900 debugged has signalled, since it calls @code{raise} and raises a
11901 @code{SIGALRM} signal. If the program being debugged had not called
11902 @code{raise}, then @value{GDBN} would report a normal exit:
11905 (@value{GDBP}) has_exited_or_signalled
11906 The program has exited
11910 The variable @code{$_exception} is set to the exception object being
11911 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
11913 @item $_ada_exception
11914 The variable @code{$_ada_exception} is set to the address of the
11915 exception being caught or thrown at an Ada exception-related
11916 catchpoint. @xref{Set Catchpoints}.
11919 @itemx $_probe_arg0@dots{}$_probe_arg11
11920 Arguments to a static probe. @xref{Static Probe Points}.
11923 @vindex $_sdata@r{, inspect, convenience variable}
11924 The variable @code{$_sdata} contains extra collected static tracepoint
11925 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
11926 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
11927 if extra static tracepoint data has not been collected.
11930 @vindex $_siginfo@r{, convenience variable}
11931 The variable @code{$_siginfo} contains extra signal information
11932 (@pxref{extra signal information}). Note that @code{$_siginfo}
11933 could be empty, if the application has not yet received any signals.
11934 For example, it will be empty before you execute the @code{run} command.
11937 @vindex $_tlb@r{, convenience variable}
11938 The variable @code{$_tlb} is automatically set when debugging
11939 applications running on MS-Windows in native mode or connected to
11940 gdbserver that supports the @code{qGetTIBAddr} request.
11941 @xref{General Query Packets}.
11942 This variable contains the address of the thread information block.
11945 The number of the current inferior. @xref{Inferiors and
11946 Programs, ,Debugging Multiple Inferiors and Programs}.
11949 The thread number of the current thread. @xref{thread numbers}.
11952 The global number of the current thread. @xref{global thread number}.
11954 @item $_thread_systag
11955 The target system's thread identifier (@var{systag}) string of the
11956 current thread. @xref{target system thread identifier}.
11958 @item $_thread_name
11959 The thread name string of the current thread, or the empty string if
11960 no name has been assigned. @xref{thread name}.
11964 @vindex $_gdb_major@r{, convenience variable}
11965 @vindex $_gdb_minor@r{, convenience variable}
11966 The major and minor version numbers of the running @value{GDBN}.
11967 Development snapshots and pretest versions have their minor version
11968 incremented by one; thus, @value{GDBN} pretest 9.11.90 will produce
11969 the value 12 for @code{$_gdb_minor}. These variables allow you to
11970 write scripts that work with different versions of @value{GDBN}
11971 without errors caused by features unavailable in some of those
11974 @item $_shell_exitcode
11975 @itemx $_shell_exitsignal
11976 @vindex $_shell_exitcode@r{, convenience variable}
11977 @vindex $_shell_exitsignal@r{, convenience variable}
11978 @cindex shell command, exit code
11979 @cindex shell command, exit signal
11980 @cindex exit status of shell commands
11981 @value{GDBN} commands such as @code{shell} and @code{|} are launching
11982 shell commands. When a launched command terminates, @value{GDBN}
11983 automatically maintains the variables @code{$_shell_exitcode}
11984 and @code{$_shell_exitsignal} according to the exit status of the last
11985 launched command. These variables are set and used similarly to
11986 the variables @code{$_exitcode} and @code{$_exitsignal}.
11989 The per-inferior heterogeneous agent number of the current thread, or 0
11990 if not associated with a heterogeneous dispatch. @xref{Heterogeneous
11994 The global heterogeneous agent number of the current thread, or 0 if not
11995 associated with a heterogeneous dispatch. @xref{Heterogeneous
11999 The per-inferior heterogeneous queue number of the current thread, or 0
12000 if not associated with a heterogeneous dispatch. @xref{Heterogeneous
12004 The global heterogeneous queue number of the current thread, or 0 if not
12005 associated with a heterogeneous dispatch. @xref{Heterogeneous
12009 The per-inferior heterogeneous dispatch number of the current thread, or
12010 0 if not associated with a heterogeneous dispatch. @xref{Heterogeneous
12014 The global heterogeneous dispatch number of the current thread, or 0 if
12015 not associated with a heterogeneous dispatch. @xref{Heterogeneous
12019 The per-inferior heterogeneous lane number of the current
12020 heterogeneous lane of the current thread. @xref{Heterogeneous
12023 @c FIXME-implementors!! Should there be @code{$_lane_index},
12024 @c @code{$_lane_active} and @code{$_lane_count} convenience variables?
12025 @c Note that the lane count needs to take into account when a grid
12026 @c size is not a multiple of the work-group size (resulting in partial
12027 @c work-groups on the dimension edges of the grid), and the work-group
12028 @c size is not a multiple of the wavefront size.
12031 The global heterogeneous lane number of the current heterogeneous
12032 lane. @xref{Heterogeneous Debugging}.
12034 @item $_lane_systag
12035 The target system's heterogeneous lane identifier (@var{lane_systag})
12036 string of the current lane in the current thread. @xref{target system
12040 The heterogeneous lane name string of the current heterogeneous lane, or
12041 the empty string if no name has been assigned by the @samp{lane name}
12042 command. @xref{Heterogeneous Debugging}.
12044 @item $_dispatch_pos
12045 The heterogeneous dispatch position string of the current thread, or the
12046 empty string if not associated with a heterogeneous dispatch.
12047 @xref{Heterogeneous Debugging}.
12049 @item $_thread_workgroup_pos
12050 @itemx $_lane_workgroup_pos
12051 The heterogeneous work-group position string of the current thread or
12052 heterogeneous lane respectively, or the empty string if not associated
12053 with a heterogeneous dispatch. @xref{Heterogeneous Debugging}.
12057 @node Convenience Funs
12058 @section Convenience Functions
12060 @cindex convenience functions
12061 @value{GDBN} also supplies some @dfn{convenience functions}. These
12062 have a syntax similar to convenience variables. A convenience
12063 function can be used in an expression just like an ordinary function;
12064 however, a convenience function is implemented internally to
12067 These functions do not require @value{GDBN} to be configured with
12068 @code{Python} support, which means that they are always available.
12072 @item $_isvoid (@var{expr})
12073 @findex $_isvoid@r{, convenience function}
12074 Return one if the expression @var{expr} is @code{void}. Otherwise it
12077 A @code{void} expression is an expression where the type of the result
12078 is @code{void}. For example, you can examine a convenience variable
12079 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
12083 (@value{GDBP}) print $_exitcode
12085 (@value{GDBP}) print $_isvoid ($_exitcode)
12088 Starting program: ./a.out
12089 [Inferior 1 (process 29572) exited normally]
12090 (@value{GDBP}) print $_exitcode
12092 (@value{GDBP}) print $_isvoid ($_exitcode)
12096 In the example above, we used @code{$_isvoid} to check whether
12097 @code{$_exitcode} is @code{void} before and after the execution of the
12098 program being debugged. Before the execution there is no exit code to
12099 be examined, therefore @code{$_exitcode} is @code{void}. After the
12100 execution the program being debugged returned zero, therefore
12101 @code{$_exitcode} is zero, which means that it is not @code{void}
12104 The @code{void} expression can also be a call of a function from the
12105 program being debugged. For example, given the following function:
12114 The result of calling it inside @value{GDBN} is @code{void}:
12117 (@value{GDBP}) print foo ()
12119 (@value{GDBP}) print $_isvoid (foo ())
12121 (@value{GDBP}) set $v = foo ()
12122 (@value{GDBP}) print $v
12124 (@value{GDBP}) print $_isvoid ($v)
12128 @item $_gdb_setting_str (@var{setting})
12129 @findex $_gdb_setting_str@r{, convenience function}
12130 Return the value of the @value{GDBN} @var{setting} as a string.
12131 @var{setting} is any setting that can be used in a @code{set} or
12132 @code{show} command (@pxref{Controlling GDB}).
12135 (@value{GDBP}) show print frame-arguments
12136 Printing of non-scalar frame arguments is "scalars".
12137 (@value{GDBP}) p $_gdb_setting_str("print frame-arguments")
12139 (@value{GDBP}) p $_gdb_setting_str("height")
12144 @item $_gdb_setting (@var{setting})
12145 @findex $_gdb_setting@r{, convenience function}
12146 Return the value of the @value{GDBN} @var{setting}.
12147 The type of the returned value depends on the setting.
12149 The value type for boolean and auto boolean settings is @code{int}.
12150 The boolean values @code{off} and @code{on} are converted to
12151 the integer values @code{0} and @code{1}. The value @code{auto} is
12152 converted to the value @code{-1}.
12154 The value type for integer settings is either @code{unsigned int}
12155 or @code{int}, depending on the setting.
12157 Some integer settings accept an @code{unlimited} value.
12158 Depending on the setting, the @code{set} command also accepts
12159 the value @code{0} or the value @code{@minus{}1} as a synonym for
12161 For example, @code{set height unlimited} is equivalent to
12162 @code{set height 0}.
12164 Some other settings that accept the @code{unlimited} value
12165 use the value @code{0} to literally mean zero.
12166 For example, @code{set history size 0} indicates to not
12167 record any @value{GDBN} commands in the command history.
12168 For such settings, @code{@minus{}1} is the synonym
12169 for @code{unlimited}.
12171 See the documentation of the corresponding @code{set} command for
12172 the numerical value equivalent to @code{unlimited}.
12174 The @code{$_gdb_setting} function converts the unlimited value
12175 to a @code{0} or a @code{@minus{}1} value according to what the
12176 @code{set} command uses.
12180 (@value{GDBP}) p $_gdb_setting_str("height")
12182 (@value{GDBP}) p $_gdb_setting("height")
12184 (@value{GDBP}) set height unlimited
12185 (@value{GDBP}) p $_gdb_setting_str("height")
12187 (@value{GDBP}) p $_gdb_setting("height")
12191 (@value{GDBP}) p $_gdb_setting_str("history size")
12193 (@value{GDBP}) p $_gdb_setting("history size")
12195 (@value{GDBP}) p $_gdb_setting_str("disassemble-next-line")
12197 (@value{GDBP}) p $_gdb_setting("disassemble-next-line")
12203 Other setting types (enum, filename, optional filename, string, string noescape)
12204 are returned as string values.
12207 @item $_gdb_maint_setting_str (@var{setting})
12208 @findex $_gdb_maint_setting_str@r{, convenience function}
12209 Like the @code{$_gdb_setting_str} function, but works with
12210 @code{maintenance set} variables.
12212 @item $_gdb_maint_setting (@var{setting})
12213 @findex $_gdb_maint_setting@r{, convenience function}
12214 Like the @code{$_gdb_setting} function, but works with
12215 @code{maintenance set} variables.
12219 The following functions require @value{GDBN} to be configured with
12220 @code{Python} support.
12224 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
12225 @findex $_memeq@r{, convenience function}
12226 Returns one if the @var{length} bytes at the addresses given by
12227 @var{buf1} and @var{buf2} are equal.
12228 Otherwise it returns zero.
12230 @item $_regex(@var{str}, @var{regex})
12231 @findex $_regex@r{, convenience function}
12232 Returns one if the string @var{str} matches the regular expression
12233 @var{regex}. Otherwise it returns zero.
12234 The syntax of the regular expression is that specified by Python's
12235 regular expression support.
12237 @item $_streq(@var{str1}, @var{str2})
12238 @findex $_streq@r{, convenience function}
12239 Returns one if the strings @var{str1} and @var{str2} are equal.
12240 Otherwise it returns zero.
12242 @item $_strlen(@var{str})
12243 @findex $_strlen@r{, convenience function}
12244 Returns the length of string @var{str}.
12246 @item $_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
12247 @findex $_caller_is@r{, convenience function}
12248 Returns one if the calling function's name is equal to @var{name}.
12249 Otherwise it returns zero.
12251 If the optional argument @var{number_of_frames} is provided,
12252 it is the number of frames up in the stack to look.
12258 (@value{GDBP}) backtrace
12260 at testsuite/gdb.python/py-caller-is.c:21
12261 #1 0x00000000004005a0 in middle_func ()
12262 at testsuite/gdb.python/py-caller-is.c:27
12263 #2 0x00000000004005ab in top_func ()
12264 at testsuite/gdb.python/py-caller-is.c:33
12265 #3 0x00000000004005b6 in main ()
12266 at testsuite/gdb.python/py-caller-is.c:39
12267 (@value{GDBP}) print $_caller_is ("middle_func")
12269 (@value{GDBP}) print $_caller_is ("top_func", 2)
12273 @item $_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
12274 @findex $_caller_matches@r{, convenience function}
12275 Returns one if the calling function's name matches the regular expression
12276 @var{regexp}. Otherwise it returns zero.
12278 If the optional argument @var{number_of_frames} is provided,
12279 it is the number of frames up in the stack to look.
12282 @item $_any_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
12283 @findex $_any_caller_is@r{, convenience function}
12284 Returns one if any calling function's name is equal to @var{name}.
12285 Otherwise it returns zero.
12287 If the optional argument @var{number_of_frames} is provided,
12288 it is the number of frames up in the stack to look.
12291 This function differs from @code{$_caller_is} in that this function
12292 checks all stack frames from the immediate caller to the frame specified
12293 by @var{number_of_frames}, whereas @code{$_caller_is} only checks the
12294 frame specified by @var{number_of_frames}.
12296 @item $_any_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
12297 @findex $_any_caller_matches@r{, convenience function}
12298 Returns one if any calling function's name matches the regular expression
12299 @var{regexp}. Otherwise it returns zero.
12301 If the optional argument @var{number_of_frames} is provided,
12302 it is the number of frames up in the stack to look.
12305 This function differs from @code{$_caller_matches} in that this function
12306 checks all stack frames from the immediate caller to the frame specified
12307 by @var{number_of_frames}, whereas @code{$_caller_matches} only checks the
12308 frame specified by @var{number_of_frames}.
12310 @item $_as_string(@var{value})
12311 @findex $_as_string@r{, convenience function}
12312 Return the string representation of @var{value}.
12314 This function is useful to obtain the textual label (enumerator) of an
12315 enumeration value. For example, assuming the variable @var{node} is of
12316 an enumerated type:
12319 (@value{GDBP}) printf "Visiting node of type %s\n", $_as_string(node)
12320 Visiting node of type NODE_INTEGER
12323 @item $_cimag(@var{value})
12324 @itemx $_creal(@var{value})
12325 @findex $_cimag@r{, convenience function}
12326 @findex $_creal@r{, convenience function}
12327 Return the imaginary (@code{$_cimag}) or real (@code{$_creal}) part of
12328 the complex number @var{value}.
12330 The type of the imaginary or real part depends on the type of the
12331 complex number, e.g., using @code{$_cimag} on a @code{float complex}
12332 will return an imaginary part of type @code{float}.
12334 @item $_thread_find(@var{regex})
12335 @findex $_thread_find@r{, convenience function}
12336 Searches for threads whose name or @var{systag} matches the supplied
12337 regular expression. The syntax of the regular expression is that
12338 specified by Python's regular expression support.
12340 Returns a string that is the space separated list of per-inferior
12341 thread numbers of the found threads. If debugging multiple inferiors,
12342 the thread numbers are qualified with the inferior number. If no
12343 threads are found, the empty string is returned. The string can be
12344 used in commands that accept a thread ID list. @xref{thread ID
12347 @c FIXME-implementors!! Should this convenience function return a
12348 @c tuple rather than a string?
12350 For example, the following command lists all threads that are part of
12351 the heterogeneous work-group with dispatch position @samp{(1,2,3)}
12352 (@pxref{Heterogeneous Debugging}):
12355 (@value{GDBP}) info threads $_thread_find ("work-group(1,2,3)")
12358 @item $_thread_find_first_gid(@var{regex})
12359 @findex $_thread_find_first_gid@r{, convenience function}
12360 Similar to the @code{$_thread_find} convenience function, except it
12361 returns a number that is the global thread number of one of the
12362 threads found, or 0 if no threads were found. The number can be used
12363 in commands that accept a global thread number. @xref{global thread
12366 @c FIXME-implementors!! If @code{$_thread_find} returns a tuple then
12367 @c this convenience function may not be necessary as one can simply
12368 @c add @samp{[0]} to access the first element of a tuple.
12370 For example, the following command sets the current thread to one of
12371 the threads that are part of the heterogeneous work-group with
12372 dispatch position @samp{(1,2,3)} (@pxref{Heterogeneous Debugging}):
12375 (@value{GDBP}) thread -gid $_thread_find_first_gid ("work-group(1,2,3)")
12378 @item $_lane_find(@var{regex})
12379 @itemx $_lane_find_first_gid(@var{regex})
12380 Similar to @samp{$_thread_find} and @samp{$_thread_find_first_gid}
12381 except for heterogeneous lanes. @xref{Heterogeneous Debugging}.
12385 @value{GDBN} provides the ability to list and get help on
12386 convenience functions.
12389 @item help function
12390 @kindex help function
12391 @cindex show all convenience functions
12392 Print a list of all convenience functions.
12399 You can refer to machine register contents, in expressions, as variables
12400 with names starting with @samp{$}. The names of registers are different
12401 for each machine; use @code{info registers} to see the names used on
12405 @kindex info registers
12406 @item info registers
12407 Print the names and values of all registers except floating-point
12408 and vector registers (in the selected stack frame).
12410 @kindex info all-registers
12411 @cindex floating point registers
12412 @item info all-registers
12413 Print the names and values of all registers, including floating-point
12414 and vector registers (in the selected stack frame).
12416 @item info registers @var{reggroup} @dots{}
12417 Print the name and value of the registers in each of the specified
12418 @var{reggroup}s. The @var{reggroup} can be any of those returned by
12419 @code{maint print reggroups} (@pxref{Maintenance Commands}).
12421 @item info registers @var{regname} @dots{}
12422 Print the @dfn{relativized} value of each specified register @var{regname}.
12423 As discussed in detail below, register values are normally relative to
12424 the selected stack frame. The @var{regname} may be any register name valid on
12425 the machine you are using, with or without the initial @samp{$}.
12428 @anchor{standard registers}
12429 @cindex stack pointer register
12430 @cindex program counter register
12431 @cindex process status register
12432 @cindex frame pointer register
12433 @cindex standard registers
12434 @value{GDBN} has four ``standard'' register names that are available (in
12435 expressions) on most machines---whenever they do not conflict with an
12436 architecture's canonical mnemonics for registers. The register names
12437 @code{$pc} and @code{$sp} are used for the program counter register and
12438 the stack pointer. @code{$fp} is used for a register that contains a
12439 pointer to the current stack frame, and @code{$ps} is used for a
12440 register that contains the processor status. For example,
12441 you could print the program counter in hex with
12448 or print the instruction to be executed next with
12455 or add four to the stack pointer@footnote{This is a way of removing
12456 one word from the stack, on machines where stacks grow downward in
12457 memory (most machines, nowadays). This assumes that the innermost
12458 stack frame is selected; setting @code{$sp} is not allowed when other
12459 stack frames are selected. To pop entire frames off the stack,
12460 regardless of machine architecture, use @code{return};
12461 see @ref{Returning, ,Returning from a Function}.} with
12467 Whenever possible, these four standard register names are available on
12468 your machine even though the machine has different canonical mnemonics,
12469 so long as there is no conflict. The @code{info registers} command
12470 shows the canonical names. For example, on the SPARC, @code{info
12471 registers} displays the processor status register as @code{$psr} but you
12472 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
12473 is an alias for the @sc{eflags} register.
12475 @value{GDBN} always considers the contents of an ordinary register as an
12476 integer when the register is examined in this way. Some machines have
12477 special registers which can hold nothing but floating point; these
12478 registers are considered to have floating point values. There is no way
12479 to refer to the contents of an ordinary register as floating point value
12480 (although you can @emph{print} it as a floating point value with
12481 @samp{print/f $@var{regname}}).
12483 Some registers have distinct ``raw'' and ``virtual'' data formats. This
12484 means that the data format in which the register contents are saved by
12485 the operating system is not the same one that your program normally
12486 sees. For example, the registers of the 68881 floating point
12487 coprocessor are always saved in ``extended'' (raw) format, but all C
12488 programs expect to work with ``double'' (virtual) format. In such
12489 cases, @value{GDBN} normally works with the virtual format only (the format
12490 that makes sense for your program), but the @code{info registers} command
12491 prints the data in both formats.
12493 @cindex SSE registers (x86)
12494 @cindex MMX registers (x86)
12495 Some machines have special registers whose contents can be interpreted
12496 in several different ways. For example, modern x86-based machines
12497 have SSE and MMX registers that can hold several values packed
12498 together in several different formats. @value{GDBN} refers to such
12499 registers in @code{struct} notation:
12502 (@value{GDBP}) print $xmm1
12504 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
12505 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
12506 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
12507 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
12508 v4_int32 = @{0, 20657912, 11, 13@},
12509 v2_int64 = @{88725056443645952, 55834574859@},
12510 uint128 = 0x0000000d0000000b013b36f800000000
12515 To set values of such registers, you need to tell @value{GDBN} which
12516 view of the register you wish to change, as if you were assigning
12517 value to a @code{struct} member:
12520 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
12523 Normally, register values are relative to the selected stack frame
12524 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
12525 value that the register would contain if all stack frames farther in
12526 were exited and their saved registers restored. In order to see the
12527 true contents of hardware registers, you must select the innermost
12528 frame (with @samp{frame 0}).
12530 @cindex caller-saved registers
12531 @cindex call-clobbered registers
12532 @cindex volatile registers
12533 @cindex <not saved> values
12534 Usually ABIs reserve some registers as not needed to be saved by the
12535 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
12536 registers). It may therefore not be possible for @value{GDBN} to know
12537 the value a register had before the call (in other words, in the outer
12538 frame), if the register value has since been changed by the callee.
12539 @value{GDBN} tries to deduce where the inner frame saved
12540 (``callee-saved'') registers, from the debug info, unwind info, or the
12541 machine code generated by your compiler. If some register is not
12542 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
12543 its own knowledge of the ABI, or because the debug/unwind info
12544 explicitly says the register's value is undefined), @value{GDBN}
12545 displays @w{@samp{<not saved>}} as the register's value. With targets
12546 that @value{GDBN} has no knowledge of the register saving convention,
12547 if a register was not saved by the callee, then its value and location
12548 in the outer frame are assumed to be the same of the inner frame.
12549 This is usually harmless, because if the register is call-clobbered,
12550 the caller either does not care what is in the register after the
12551 call, or has code to restore the value that it does care about. Note,
12552 however, that if you change such a register in the outer frame, you
12553 may also be affecting the inner frame. Also, the more ``outer'' the
12554 frame is you're looking at, the more likely a call-clobbered
12555 register's value is to be wrong, in the sense that it doesn't actually
12556 represent the value the register had just before the call.
12558 @node Floating Point Hardware
12559 @section Floating Point Hardware
12560 @cindex floating point
12562 Depending on the configuration, @value{GDBN} may be able to give
12563 you more information about the status of the floating point hardware.
12568 Display hardware-dependent information about the floating
12569 point unit. The exact contents and layout vary depending on the
12570 floating point chip. Currently, @samp{info float} is supported on
12571 the ARM and x86 machines.
12575 @section Vector Unit
12576 @cindex vector unit
12578 Depending on the configuration, @value{GDBN} may be able to give you
12579 more information about the status of the vector unit.
12582 @kindex info vector
12584 Display information about the vector unit. The exact contents and
12585 layout vary depending on the hardware.
12588 @node OS Information
12589 @section Operating System Auxiliary Information
12590 @cindex OS information
12592 @value{GDBN} provides interfaces to useful OS facilities that can help
12593 you debug your program.
12595 @cindex auxiliary vector
12596 @cindex vector, auxiliary
12597 Some operating systems supply an @dfn{auxiliary vector} to programs at
12598 startup. This is akin to the arguments and environment that you
12599 specify for a program, but contains a system-dependent variety of
12600 binary values that tell system libraries important details about the
12601 hardware, operating system, and process. Each value's purpose is
12602 identified by an integer tag; the meanings are well-known but system-specific.
12603 Depending on the configuration and operating system facilities,
12604 @value{GDBN} may be able to show you this information. For remote
12605 targets, this functionality may further depend on the remote stub's
12606 support of the @samp{qXfer:auxv:read} packet, see
12607 @ref{qXfer auxiliary vector read}.
12612 Display the auxiliary vector of the inferior, which can be either a
12613 live process or a core dump file. @value{GDBN} prints each tag value
12614 numerically, and also shows names and text descriptions for recognized
12615 tags. Some values in the vector are numbers, some bit masks, and some
12616 pointers to strings or other data. @value{GDBN} displays each value in the
12617 most appropriate form for a recognized tag, and in hexadecimal for
12618 an unrecognized tag.
12621 On some targets, @value{GDBN} can access operating system-specific
12622 information and show it to you. The types of information available
12623 will differ depending on the type of operating system running on the
12624 target. The mechanism used to fetch the data is described in
12625 @ref{Operating System Information}. For remote targets, this
12626 functionality depends on the remote stub's support of the
12627 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
12631 @item info os @var{infotype}
12633 Display OS information of the requested type.
12635 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
12637 @anchor{linux info os infotypes}
12639 @kindex info os cpus
12641 Display the list of all CPUs/cores. For each CPU/core, @value{GDBN} prints
12642 the available fields from /proc/cpuinfo. For each supported architecture
12643 different fields are available. Two common entries are processor which gives
12644 CPU number and bogomips; a system constant that is calculated during
12645 kernel initialization.
12647 @kindex info os files
12649 Display the list of open file descriptors on the target. For each
12650 file descriptor, @value{GDBN} prints the identifier of the process
12651 owning the descriptor, the command of the owning process, the value
12652 of the descriptor, and the target of the descriptor.
12654 @kindex info os modules
12656 Display the list of all loaded kernel modules on the target. For each
12657 module, @value{GDBN} prints the module name, the size of the module in
12658 bytes, the number of times the module is used, the dependencies of the
12659 module, the status of the module, and the address of the loaded module
12662 @kindex info os msg
12664 Display the list of all System V message queues on the target. For each
12665 message queue, @value{GDBN} prints the message queue key, the message
12666 queue identifier, the access permissions, the current number of bytes
12667 on the queue, the current number of messages on the queue, the processes
12668 that last sent and received a message on the queue, the user and group
12669 of the owner and creator of the message queue, the times at which a
12670 message was last sent and received on the queue, and the time at which
12671 the message queue was last changed.
12673 @kindex info os processes
12675 Display the list of processes on the target. For each process,
12676 @value{GDBN} prints the process identifier, the name of the user, the
12677 command corresponding to the process, and the list of processor cores
12678 that the process is currently running on. (To understand what these
12679 properties mean, for this and the following info types, please consult
12680 the general @sc{gnu}/Linux documentation.)
12682 @kindex info os procgroups
12684 Display the list of process groups on the target. For each process,
12685 @value{GDBN} prints the identifier of the process group that it belongs
12686 to, the command corresponding to the process group leader, the process
12687 identifier, and the command line of the process. The list is sorted
12688 first by the process group identifier, then by the process identifier,
12689 so that processes belonging to the same process group are grouped together
12690 and the process group leader is listed first.
12692 @kindex info os semaphores
12694 Display the list of all System V semaphore sets on the target. For each
12695 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
12696 set identifier, the access permissions, the number of semaphores in the
12697 set, the user and group of the owner and creator of the semaphore set,
12698 and the times at which the semaphore set was operated upon and changed.
12700 @kindex info os shm
12702 Display the list of all System V shared-memory regions on the target.
12703 For each shared-memory region, @value{GDBN} prints the region key,
12704 the shared-memory identifier, the access permissions, the size of the
12705 region, the process that created the region, the process that last
12706 attached to or detached from the region, the current number of live
12707 attaches to the region, and the times at which the region was last
12708 attached to, detach from, and changed.
12710 @kindex info os sockets
12712 Display the list of Internet-domain sockets on the target. For each
12713 socket, @value{GDBN} prints the address and port of the local and
12714 remote endpoints, the current state of the connection, the creator of
12715 the socket, the IP address family of the socket, and the type of the
12718 @kindex info os threads
12720 Display the list of threads running on the target. For each thread,
12721 @value{GDBN} prints the identifier of the process that the thread
12722 belongs to, the command of the process, the thread identifier, and the
12723 processor core that it is currently running on. The main thread of a
12724 process is not listed.
12728 If @var{infotype} is omitted, then list the possible values for
12729 @var{infotype} and the kind of OS information available for each
12730 @var{infotype}. If the target does not return a list of possible
12731 types, this command will report an error.
12734 @node Memory Region Attributes
12735 @section Memory Region Attributes
12736 @cindex memory region attributes
12738 @dfn{Memory region attributes} allow you to describe special handling
12739 required by regions of your target's memory. @value{GDBN} uses
12740 attributes to determine whether to allow certain types of memory
12741 accesses; whether to use specific width accesses; and whether to cache
12742 target memory. By default the description of memory regions is
12743 fetched from the target (if the current target supports this), but the
12744 user can override the fetched regions.
12746 Defined memory regions can be individually enabled and disabled. When a
12747 memory region is disabled, @value{GDBN} uses the default attributes when
12748 accessing memory in that region. Similarly, if no memory regions have
12749 been defined, @value{GDBN} uses the default attributes when accessing
12752 When a memory region is defined, it is given a number to identify it;
12753 to enable, disable, or remove a memory region, you specify that number.
12757 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
12758 Define a memory region bounded by @var{lower} and @var{upper} with
12759 attributes @var{attributes}@dots{}, and add it to the list of regions
12760 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
12761 case: it is treated as the target's maximum memory address.
12762 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
12765 Discard any user changes to the memory regions and use target-supplied
12766 regions, if available, or no regions if the target does not support.
12769 @item delete mem @var{nums}@dots{}
12770 Remove memory regions @var{nums}@dots{} from the list of regions
12771 monitored by @value{GDBN}.
12773 @kindex disable mem
12774 @item disable mem @var{nums}@dots{}
12775 Disable monitoring of memory regions @var{nums}@dots{}.
12776 A disabled memory region is not forgotten.
12777 It may be enabled again later.
12780 @item enable mem @var{nums}@dots{}
12781 Enable monitoring of memory regions @var{nums}@dots{}.
12785 Print a table of all defined memory regions, with the following columns
12789 @item Memory Region Number
12790 @item Enabled or Disabled.
12791 Enabled memory regions are marked with @samp{y}.
12792 Disabled memory regions are marked with @samp{n}.
12795 The address defining the inclusive lower bound of the memory region.
12798 The address defining the exclusive upper bound of the memory region.
12801 The list of attributes set for this memory region.
12806 @subsection Attributes
12808 @subsubsection Memory Access Mode
12809 The access mode attributes set whether @value{GDBN} may make read or
12810 write accesses to a memory region.
12812 While these attributes prevent @value{GDBN} from performing invalid
12813 memory accesses, they do nothing to prevent the target system, I/O DMA,
12814 etc.@: from accessing memory.
12818 Memory is read only.
12820 Memory is write only.
12822 Memory is read/write. This is the default.
12825 @subsubsection Memory Access Size
12826 The access size attribute tells @value{GDBN} to use specific sized
12827 accesses in the memory region. Often memory mapped device registers
12828 require specific sized accesses. If no access size attribute is
12829 specified, @value{GDBN} may use accesses of any size.
12833 Use 8 bit memory accesses.
12835 Use 16 bit memory accesses.
12837 Use 32 bit memory accesses.
12839 Use 64 bit memory accesses.
12842 @c @subsubsection Hardware/Software Breakpoints
12843 @c The hardware/software breakpoint attributes set whether @value{GDBN}
12844 @c will use hardware or software breakpoints for the internal breakpoints
12845 @c used by the step, next, finish, until, etc. commands.
12849 @c Always use hardware breakpoints
12850 @c @item swbreak (default)
12853 @subsubsection Data Cache
12854 The data cache attributes set whether @value{GDBN} will cache target
12855 memory. While this generally improves performance by reducing debug
12856 protocol overhead, it can lead to incorrect results because @value{GDBN}
12857 does not know about volatile variables or memory mapped device
12862 Enable @value{GDBN} to cache target memory.
12864 Disable @value{GDBN} from caching target memory. This is the default.
12867 @subsection Memory Access Checking
12868 @value{GDBN} can be instructed to refuse accesses to memory that is
12869 not explicitly described. This can be useful if accessing such
12870 regions has undesired effects for a specific target, or to provide
12871 better error checking. The following commands control this behaviour.
12874 @kindex set mem inaccessible-by-default
12875 @item set mem inaccessible-by-default [on|off]
12876 If @code{on} is specified, make @value{GDBN} treat memory not
12877 explicitly described by the memory ranges as non-existent and refuse accesses
12878 to such memory. The checks are only performed if there's at least one
12879 memory range defined. If @code{off} is specified, make @value{GDBN}
12880 treat the memory not explicitly described by the memory ranges as RAM.
12881 The default value is @code{on}.
12882 @kindex show mem inaccessible-by-default
12883 @item show mem inaccessible-by-default
12884 Show the current handling of accesses to unknown memory.
12888 @c @subsubsection Memory Write Verification
12889 @c The memory write verification attributes set whether @value{GDBN}
12890 @c will re-reads data after each write to verify the write was successful.
12894 @c @item noverify (default)
12897 @node Dump/Restore Files
12898 @section Copy Between Memory and a File
12899 @cindex dump/restore files
12900 @cindex append data to a file
12901 @cindex dump data to a file
12902 @cindex restore data from a file
12904 You can use the commands @code{dump}, @code{append}, and
12905 @code{restore} to copy data between target memory and a file. The
12906 @code{dump} and @code{append} commands write data to a file, and the
12907 @code{restore} command reads data from a file back into the inferior's
12908 memory. Files may be in binary, Motorola S-record, Intel hex,
12909 Tektronix Hex, or Verilog Hex format; however, @value{GDBN} can only
12910 append to binary files, and cannot read from Verilog Hex files.
12915 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12916 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
12917 Dump the contents of memory from @var{start_addr} to @var{end_addr},
12918 or the value of @var{expr}, to @var{filename} in the given format.
12920 The @var{format} parameter may be any one of:
12927 Motorola S-record format.
12929 Tektronix Hex format.
12931 Verilog Hex format.
12934 @value{GDBN} uses the same definitions of these formats as the
12935 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
12936 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
12940 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12941 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
12942 Append the contents of memory from @var{start_addr} to @var{end_addr},
12943 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
12944 (@value{GDBN} can only append data to files in raw binary form.)
12947 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
12948 Restore the contents of file @var{filename} into memory. The
12949 @code{restore} command can automatically recognize any known @sc{bfd}
12950 file format, except for raw binary. To restore a raw binary file you
12951 must specify the optional keyword @code{binary} after the filename.
12953 If @var{bias} is non-zero, its value will be added to the addresses
12954 contained in the file. Binary files always start at address zero, so
12955 they will be restored at address @var{bias}. Other bfd files have
12956 a built-in location; they will be restored at offset @var{bias}
12957 from that location.
12959 If @var{start} and/or @var{end} are non-zero, then only data between
12960 file offset @var{start} and file offset @var{end} will be restored.
12961 These offsets are relative to the addresses in the file, before
12962 the @var{bias} argument is applied.
12966 @node Core File Generation
12967 @section How to Produce a Core File from Your Program
12968 @cindex dump core from inferior
12970 A @dfn{core file} or @dfn{core dump} is a file that records the memory
12971 image of a running process and its process status (register values
12972 etc.). Its primary use is post-mortem debugging of a program that
12973 crashed while it ran outside a debugger. A program that crashes
12974 automatically produces a core file, unless this feature is disabled by
12975 the user. @xref{Files}, for information on invoking @value{GDBN} in
12976 the post-mortem debugging mode.
12978 Occasionally, you may wish to produce a core file of the program you
12979 are debugging in order to preserve a snapshot of its state.
12980 @value{GDBN} has a special command for that.
12984 @kindex generate-core-file
12985 @item generate-core-file [@var{file}]
12986 @itemx gcore [@var{file}]
12987 Produce a core dump of the inferior process. The optional argument
12988 @var{file} specifies the file name where to put the core dump. If not
12989 specified, the file name defaults to @file{core.@var{pid}}, where
12990 @var{pid} is the inferior process ID.
12992 Note that this command is implemented only for some systems (as of
12993 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
12995 On @sc{gnu}/Linux, this command can take into account the value of the
12996 file @file{/proc/@var{pid}/coredump_filter} when generating the core
12997 dump (@pxref{set use-coredump-filter}), and by default honors the
12998 @code{VM_DONTDUMP} flag for mappings where it is present in the file
12999 @file{/proc/@var{pid}/smaps} (@pxref{set dump-excluded-mappings}).
13001 @kindex set use-coredump-filter
13002 @anchor{set use-coredump-filter}
13003 @item set use-coredump-filter on
13004 @itemx set use-coredump-filter off
13005 Enable or disable the use of the file
13006 @file{/proc/@var{pid}/coredump_filter} when generating core dump
13007 files. This file is used by the Linux kernel to decide what types of
13008 memory mappings will be dumped or ignored when generating a core dump
13009 file. @var{pid} is the process ID of a currently running process.
13011 To make use of this feature, you have to write in the
13012 @file{/proc/@var{pid}/coredump_filter} file a value, in hexadecimal,
13013 which is a bit mask representing the memory mapping types. If a bit
13014 is set in the bit mask, then the memory mappings of the corresponding
13015 types will be dumped; otherwise, they will be ignored. This
13016 configuration is inherited by child processes. For more information
13017 about the bits that can be set in the
13018 @file{/proc/@var{pid}/coredump_filter} file, please refer to the
13019 manpage of @code{core(5)}.
13021 By default, this option is @code{on}. If this option is turned
13022 @code{off}, @value{GDBN} does not read the @file{coredump_filter} file
13023 and instead uses the same default value as the Linux kernel in order
13024 to decide which pages will be dumped in the core dump file. This
13025 value is currently @code{0x33}, which means that bits @code{0}
13026 (anonymous private mappings), @code{1} (anonymous shared mappings),
13027 @code{4} (ELF headers) and @code{5} (private huge pages) are active.
13028 This will cause these memory mappings to be dumped automatically.
13030 @kindex set dump-excluded-mappings
13031 @anchor{set dump-excluded-mappings}
13032 @item set dump-excluded-mappings on
13033 @itemx set dump-excluded-mappings off
13034 If @code{on} is specified, @value{GDBN} will dump memory mappings
13035 marked with the @code{VM_DONTDUMP} flag. This flag is represented in
13036 the file @file{/proc/@var{pid}/smaps} with the acronym @code{dd}.
13038 The default value is @code{off}.
13041 @node Character Sets
13042 @section Character Sets
13043 @cindex character sets
13045 @cindex translating between character sets
13046 @cindex host character set
13047 @cindex target character set
13049 If the program you are debugging uses a different character set to
13050 represent characters and strings than the one @value{GDBN} uses itself,
13051 @value{GDBN} can automatically translate between the character sets for
13052 you. The character set @value{GDBN} uses we call the @dfn{host
13053 character set}; the one the inferior program uses we call the
13054 @dfn{target character set}.
13056 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
13057 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
13058 remote protocol (@pxref{Remote Debugging}) to debug a program
13059 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
13060 then the host character set is Latin-1, and the target character set is
13061 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
13062 target-charset EBCDIC-US}, then @value{GDBN} translates between
13063 @sc{ebcdic} and Latin 1 as you print character or string values, or use
13064 character and string literals in expressions.
13066 @value{GDBN} has no way to automatically recognize which character set
13067 the inferior program uses; you must tell it, using the @code{set
13068 target-charset} command, described below.
13070 Here are the commands for controlling @value{GDBN}'s character set
13074 @item set target-charset @var{charset}
13075 @kindex set target-charset
13076 Set the current target character set to @var{charset}. To display the
13077 list of supported target character sets, type
13078 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
13080 @item set host-charset @var{charset}
13081 @kindex set host-charset
13082 Set the current host character set to @var{charset}.
13084 By default, @value{GDBN} uses a host character set appropriate to the
13085 system it is running on; you can override that default using the
13086 @code{set host-charset} command. On some systems, @value{GDBN} cannot
13087 automatically determine the appropriate host character set. In this
13088 case, @value{GDBN} uses @samp{UTF-8}.
13090 @value{GDBN} can only use certain character sets as its host character
13091 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
13092 @value{GDBN} will list the host character sets it supports.
13094 @item set charset @var{charset}
13095 @kindex set charset
13096 Set the current host and target character sets to @var{charset}. As
13097 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
13098 @value{GDBN} will list the names of the character sets that can be used
13099 for both host and target.
13102 @kindex show charset
13103 Show the names of the current host and target character sets.
13105 @item show host-charset
13106 @kindex show host-charset
13107 Show the name of the current host character set.
13109 @item show target-charset
13110 @kindex show target-charset
13111 Show the name of the current target character set.
13113 @item set target-wide-charset @var{charset}
13114 @kindex set target-wide-charset
13115 Set the current target's wide character set to @var{charset}. This is
13116 the character set used by the target's @code{wchar_t} type. To
13117 display the list of supported wide character sets, type
13118 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
13120 @item show target-wide-charset
13121 @kindex show target-wide-charset
13122 Show the name of the current target's wide character set.
13125 Here is an example of @value{GDBN}'s character set support in action.
13126 Assume that the following source code has been placed in the file
13127 @file{charset-test.c}:
13133 = @{72, 101, 108, 108, 111, 44, 32, 119,
13134 111, 114, 108, 100, 33, 10, 0@};
13135 char ibm1047_hello[]
13136 = @{200, 133, 147, 147, 150, 107, 64, 166,
13137 150, 153, 147, 132, 90, 37, 0@};
13141 printf ("Hello, world!\n");
13145 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
13146 containing the string @samp{Hello, world!} followed by a newline,
13147 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
13149 We compile the program, and invoke the debugger on it:
13152 $ gcc -g charset-test.c -o charset-test
13153 $ gdb -nw charset-test
13154 GNU gdb 2001-12-19-cvs
13155 Copyright 2001 Free Software Foundation, Inc.
13160 We can use the @code{show charset} command to see what character sets
13161 @value{GDBN} is currently using to interpret and display characters and
13165 (@value{GDBP}) show charset
13166 The current host and target character set is `ISO-8859-1'.
13170 For the sake of printing this manual, let's use @sc{ascii} as our
13171 initial character set:
13173 (@value{GDBP}) set charset ASCII
13174 (@value{GDBP}) show charset
13175 The current host and target character set is `ASCII'.
13179 Let's assume that @sc{ascii} is indeed the correct character set for our
13180 host system --- in other words, let's assume that if @value{GDBN} prints
13181 characters using the @sc{ascii} character set, our terminal will display
13182 them properly. Since our current target character set is also
13183 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
13186 (@value{GDBP}) print ascii_hello
13187 $1 = 0x401698 "Hello, world!\n"
13188 (@value{GDBP}) print ascii_hello[0]
13193 @value{GDBN} uses the target character set for character and string
13194 literals you use in expressions:
13197 (@value{GDBP}) print '+'
13202 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
13205 @value{GDBN} relies on the user to tell it which character set the
13206 target program uses. If we print @code{ibm1047_hello} while our target
13207 character set is still @sc{ascii}, we get jibberish:
13210 (@value{GDBP}) print ibm1047_hello
13211 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
13212 (@value{GDBP}) print ibm1047_hello[0]
13217 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
13218 @value{GDBN} tells us the character sets it supports:
13221 (@value{GDBP}) set target-charset
13222 ASCII EBCDIC-US IBM1047 ISO-8859-1
13223 (@value{GDBP}) set target-charset
13226 We can select @sc{ibm1047} as our target character set, and examine the
13227 program's strings again. Now the @sc{ascii} string is wrong, but
13228 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
13229 target character set, @sc{ibm1047}, to the host character set,
13230 @sc{ascii}, and they display correctly:
13233 (@value{GDBP}) set target-charset IBM1047
13234 (@value{GDBP}) show charset
13235 The current host character set is `ASCII'.
13236 The current target character set is `IBM1047'.
13237 (@value{GDBP}) print ascii_hello
13238 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
13239 (@value{GDBP}) print ascii_hello[0]
13241 (@value{GDBP}) print ibm1047_hello
13242 $8 = 0x4016a8 "Hello, world!\n"
13243 (@value{GDBP}) print ibm1047_hello[0]
13248 As above, @value{GDBN} uses the target character set for character and
13249 string literals you use in expressions:
13252 (@value{GDBP}) print '+'
13257 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
13260 @node Caching Target Data
13261 @section Caching Data of Targets
13262 @cindex caching data of targets
13264 @value{GDBN} caches data exchanged between the debugger and a target.
13265 Each cache is associated with the address space of the inferior.
13266 @xref{Inferiors and Programs}, about inferior and address space.
13267 Such caching generally improves performance in remote debugging
13268 (@pxref{Remote Debugging}), because it reduces the overhead of the
13269 remote protocol by bundling memory reads and writes into large chunks.
13270 Unfortunately, simply caching everything would lead to incorrect results,
13271 since @value{GDBN} does not necessarily know anything about volatile
13272 values, memory-mapped I/O addresses, etc. Furthermore, in non-stop mode
13273 (@pxref{Non-Stop Mode}) memory can be changed @emph{while} a gdb command
13275 Therefore, by default, @value{GDBN} only caches data
13276 known to be on the stack@footnote{In non-stop mode, it is moderately
13277 rare for a running thread to modify the stack of a stopped thread
13278 in a way that would interfere with a backtrace, and caching of
13279 stack reads provides a significant speed up of remote backtraces.} or
13280 in the code segment.
13281 Other regions of memory can be explicitly marked as
13282 cacheable; @pxref{Memory Region Attributes}.
13285 @kindex set remotecache
13286 @item set remotecache on
13287 @itemx set remotecache off
13288 This option no longer does anything; it exists for compatibility
13291 @kindex show remotecache
13292 @item show remotecache
13293 Show the current state of the obsolete remotecache flag.
13295 @kindex set stack-cache
13296 @item set stack-cache on
13297 @itemx set stack-cache off
13298 Enable or disable caching of stack accesses. When @code{on}, use
13299 caching. By default, this option is @code{on}.
13301 @kindex show stack-cache
13302 @item show stack-cache
13303 Show the current state of data caching for memory accesses.
13305 @kindex set code-cache
13306 @item set code-cache on
13307 @itemx set code-cache off
13308 Enable or disable caching of code segment accesses. When @code{on},
13309 use caching. By default, this option is @code{on}. This improves
13310 performance of disassembly in remote debugging.
13312 @kindex show code-cache
13313 @item show code-cache
13314 Show the current state of target memory cache for code segment
13317 @kindex info dcache
13318 @item info dcache @r{[}line@r{]}
13319 Print the information about the performance of data cache of the
13320 current inferior's address space. The information displayed
13321 includes the dcache width and depth, and for each cache line, its
13322 number, address, and how many times it was referenced. This
13323 command is useful for debugging the data cache operation.
13325 If a line number is specified, the contents of that line will be
13328 @item set dcache size @var{size}
13329 @cindex dcache size
13330 @kindex set dcache size
13331 Set maximum number of entries in dcache (dcache depth above).
13333 @item set dcache line-size @var{line-size}
13334 @cindex dcache line-size
13335 @kindex set dcache line-size
13336 Set number of bytes each dcache entry caches (dcache width above).
13337 Must be a power of 2.
13339 @item show dcache size
13340 @kindex show dcache size
13341 Show maximum number of dcache entries. @xref{Caching Target Data, info dcache}.
13343 @item show dcache line-size
13344 @kindex show dcache line-size
13345 Show default size of dcache lines.
13349 @node Searching Memory
13350 @section Search Memory
13351 @cindex searching memory
13353 Memory can be searched for a particular sequence of bytes with the
13354 @code{find} command.
13358 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
13359 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
13360 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
13361 etc. The search begins at address @var{start_addr} and continues for either
13362 @var{len} bytes or through to @var{end_addr} inclusive.
13365 @var{s} and @var{n} are optional parameters.
13366 They may be specified in either order, apart or together.
13369 @item @var{s}, search query size
13370 The size of each search query value.
13376 halfwords (two bytes)
13380 giant words (eight bytes)
13383 All values are interpreted in the current language.
13384 This means, for example, that if the current source language is C/C@t{++}
13385 then searching for the string ``hello'' includes the trailing '\0'.
13386 The null terminator can be removed from searching by using casts,
13387 e.g.: @samp{@{char[5]@}"hello"}.
13389 If the value size is not specified, it is taken from the
13390 value's type in the current language.
13391 This is useful when one wants to specify the search
13392 pattern as a mixture of types.
13393 Note that this means, for example, that in the case of C-like languages
13394 a search for an untyped 0x42 will search for @samp{(int) 0x42}
13395 which is typically four bytes.
13397 @item @var{n}, maximum number of finds
13398 The maximum number of matches to print. The default is to print all finds.
13401 You can use strings as search values. Quote them with double-quotes
13403 The string value is copied into the search pattern byte by byte,
13404 regardless of the endianness of the target and the size specification.
13406 The address of each match found is printed as well as a count of the
13407 number of matches found.
13409 The address of the last value found is stored in convenience variable
13411 A count of the number of matches is stored in @samp{$numfound}.
13413 For example, if stopped at the @code{printf} in this function:
13419 static char hello[] = "hello-hello";
13420 static struct @{ char c; short s; int i; @}
13421 __attribute__ ((packed)) mixed
13422 = @{ 'c', 0x1234, 0x87654321 @};
13423 printf ("%s\n", hello);
13428 you get during debugging:
13431 (@value{GDBP}) find &hello[0], +sizeof(hello), "hello"
13432 0x804956d <hello.1620+6>
13434 (@value{GDBP}) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
13435 0x8049567 <hello.1620>
13436 0x804956d <hello.1620+6>
13438 (@value{GDBP}) find &hello[0], +sizeof(hello), @{char[5]@}"hello"
13439 0x8049567 <hello.1620>
13440 0x804956d <hello.1620+6>
13442 (@value{GDBP}) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
13443 0x8049567 <hello.1620>
13445 (@value{GDBP}) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
13446 0x8049560 <mixed.1625>
13448 (@value{GDBP}) print $numfound
13450 (@value{GDBP}) print $_
13451 $2 = (void *) 0x8049560
13455 @section Value Sizes
13457 Whenever @value{GDBN} prints a value memory will be allocated within
13458 @value{GDBN} to hold the contents of the value. It is possible in
13459 some languages with dynamic typing systems, that an invalid program
13460 may indicate a value that is incorrectly large, this in turn may cause
13461 @value{GDBN} to try and allocate an overly large amount of memory.
13464 @kindex set max-value-size
13465 @item set max-value-size @var{bytes}
13466 @itemx set max-value-size unlimited
13467 Set the maximum size of memory that @value{GDBN} will allocate for the
13468 contents of a value to @var{bytes}, trying to display a value that
13469 requires more memory than that will result in an error.
13471 Setting this variable does not effect values that have already been
13472 allocated within @value{GDBN}, only future allocations.
13474 There's a minimum size that @code{max-value-size} can be set to in
13475 order that @value{GDBN} can still operate correctly, this minimum is
13476 currently 16 bytes.
13478 The limit applies to the results of some subexpressions as well as to
13479 complete expressions. For example, an expression denoting a simple
13480 integer component, such as @code{x.y.z}, may fail if the size of
13481 @var{x.y} is dynamic and exceeds @var{bytes}. On the other hand,
13482 @value{GDBN} is sometimes clever; the expression @code{A[i]}, where
13483 @var{A} is an array variable with non-constant size, will generally
13484 succeed regardless of the bounds on @var{A}, as long as the component
13485 size is less than @var{bytes}.
13487 The default value of @code{max-value-size} is currently 64k.
13489 @kindex show max-value-size
13490 @item show max-value-size
13491 Show the maximum size of memory, in bytes, that @value{GDBN} will
13492 allocate for the contents of a value.
13495 @node Optimized Code
13496 @chapter Debugging Optimized Code
13497 @cindex optimized code, debugging
13498 @cindex debugging optimized code
13500 Almost all compilers support optimization. With optimization
13501 disabled, the compiler generates assembly code that corresponds
13502 directly to your source code, in a simplistic way. As the compiler
13503 applies more powerful optimizations, the generated assembly code
13504 diverges from your original source code. With help from debugging
13505 information generated by the compiler, @value{GDBN} can map from
13506 the running program back to constructs from your original source.
13508 @value{GDBN} is more accurate with optimization disabled. If you
13509 can recompile without optimization, it is easier to follow the
13510 progress of your program during debugging. But, there are many cases
13511 where you may need to debug an optimized version.
13513 When you debug a program compiled with @samp{-g -O}, remember that the
13514 optimizer has rearranged your code; the debugger shows you what is
13515 really there. Do not be too surprised when the execution path does not
13516 exactly match your source file! An extreme example: if you define a
13517 variable, but never use it, @value{GDBN} never sees that
13518 variable---because the compiler optimizes it out of existence.
13520 Some things do not work as well with @samp{-g -O} as with just
13521 @samp{-g}, particularly on machines with instruction scheduling. If in
13522 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
13523 please report it to us as a bug (including a test case!).
13524 @xref{Variables}, for more information about debugging optimized code.
13527 * Inline Functions:: How @value{GDBN} presents inlining
13528 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
13531 @node Inline Functions
13532 @section Inline Functions
13533 @cindex inline functions, debugging
13535 @dfn{Inlining} is an optimization that inserts a copy of the function
13536 body directly at each call site, instead of jumping to a shared
13537 routine. @value{GDBN} displays inlined functions just like
13538 non-inlined functions. They appear in backtraces. You can view their
13539 arguments and local variables, step into them with @code{step}, skip
13540 them with @code{next}, and escape from them with @code{finish}.
13541 You can check whether a function was inlined by using the
13542 @code{info frame} command.
13544 For @value{GDBN} to support inlined functions, the compiler must
13545 record information about inlining in the debug information ---
13546 @value{NGCC} using the @sc{dwarf 2} format does this, and several
13547 other compilers do also. @value{GDBN} only supports inlined functions
13548 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
13549 do not emit two required attributes (@samp{DW_AT_call_file} and
13550 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
13551 function calls with earlier versions of @value{NGCC}. It instead
13552 displays the arguments and local variables of inlined functions as
13553 local variables in the caller.
13555 The body of an inlined function is directly included at its call site;
13556 unlike a non-inlined function, there are no instructions devoted to
13557 the call. @value{GDBN} still pretends that the call site and the
13558 start of the inlined function are different instructions. Stepping to
13559 the call site shows the call site, and then stepping again shows
13560 the first line of the inlined function, even though no additional
13561 instructions are executed.
13563 This makes source-level debugging much clearer; you can see both the
13564 context of the call and then the effect of the call. Only stepping by
13565 a single instruction using @code{stepi} or @code{nexti} does not do
13566 this; single instruction steps always show the inlined body.
13568 There are some ways that @value{GDBN} does not pretend that inlined
13569 function calls are the same as normal calls:
13573 Setting breakpoints at the call site of an inlined function may not
13574 work, because the call site does not contain any code. @value{GDBN}
13575 may incorrectly move the breakpoint to the next line of the enclosing
13576 function, after the call. This limitation will be removed in a future
13577 version of @value{GDBN}; until then, set a breakpoint on an earlier line
13578 or inside the inlined function instead.
13581 @value{GDBN} cannot locate the return value of inlined calls after
13582 using the @code{finish} command. This is a limitation of compiler-generated
13583 debugging information; after @code{finish}, you can step to the next line
13584 and print a variable where your program stored the return value.
13588 @node Tail Call Frames
13589 @section Tail Call Frames
13590 @cindex tail call frames, debugging
13592 Function @code{B} can call function @code{C} in its very last statement. In
13593 unoptimized compilation the call of @code{C} is immediately followed by return
13594 instruction at the end of @code{B} code. Optimizing compiler may replace the
13595 call and return in function @code{B} into one jump to function @code{C}
13596 instead. Such use of a jump instruction is called @dfn{tail call}.
13598 During execution of function @code{C}, there will be no indication in the
13599 function call stack frames that it was tail-called from @code{B}. If function
13600 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
13601 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
13602 some cases @value{GDBN} can determine that @code{C} was tail-called from
13603 @code{B}, and it will then create fictitious call frame for that, with the
13604 return address set up as if @code{B} called @code{C} normally.
13606 This functionality is currently supported only by DWARF 2 debugging format and
13607 the compiler has to produce @samp{DW_TAG_call_site} tags. With
13608 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
13611 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
13612 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
13615 (@value{GDBP}) x/i $pc - 2
13616 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
13617 (@value{GDBP}) info frame
13618 Stack level 1, frame at 0x7fffffffda30:
13619 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
13620 tail call frame, caller of frame at 0x7fffffffda30
13621 source language c++.
13622 Arglist at unknown address.
13623 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
13626 The detection of all the possible code path executions can find them ambiguous.
13627 There is no execution history stored (possible @ref{Reverse Execution} is never
13628 used for this purpose) and the last known caller could have reached the known
13629 callee by multiple different jump sequences. In such case @value{GDBN} still
13630 tries to show at least all the unambiguous top tail callers and all the
13631 unambiguous bottom tail calees, if any.
13634 @anchor{set debug entry-values}
13635 @item set debug entry-values
13636 @kindex set debug entry-values
13637 When set to on, enables printing of analysis messages for both frame argument
13638 values at function entry and tail calls. It will show all the possible valid
13639 tail calls code paths it has considered. It will also print the intersection
13640 of them with the final unambiguous (possibly partial or even empty) code path
13643 @item show debug entry-values
13644 @kindex show debug entry-values
13645 Show the current state of analysis messages printing for both frame argument
13646 values at function entry and tail calls.
13649 The analysis messages for tail calls can for example show why the virtual tail
13650 call frame for function @code{c} has not been recognized (due to the indirect
13651 reference by variable @code{x}):
13654 static void __attribute__((noinline, noclone)) c (void);
13655 void (*x) (void) = c;
13656 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
13657 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
13658 int main (void) @{ x (); return 0; @}
13660 Breakpoint 1, DW_OP_entry_value resolving cannot find
13661 DW_TAG_call_site 0x40039a in main
13663 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
13666 #1 0x000000000040039a in main () at t.c:5
13669 Another possibility is an ambiguous virtual tail call frames resolution:
13673 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
13674 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
13675 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
13676 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
13677 static void __attribute__((noinline, noclone)) b (void)
13678 @{ if (i) c (); else e (); @}
13679 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
13680 int main (void) @{ a (); return 0; @}
13682 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
13683 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
13684 tailcall: reduced: 0x4004d2(a) |
13687 #1 0x00000000004004d2 in a () at t.c:8
13688 #2 0x0000000000400395 in main () at t.c:9
13691 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
13692 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
13694 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
13695 @ifset HAVE_MAKEINFO_CLICK
13696 @set ARROW @click{}
13697 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
13698 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
13700 @ifclear HAVE_MAKEINFO_CLICK
13702 @set CALLSEQ1B @value{CALLSEQ1A}
13703 @set CALLSEQ2B @value{CALLSEQ2A}
13706 Frames #0 and #2 are real, #1 is a virtual tail call frame.
13707 The code can have possible execution paths @value{CALLSEQ1B} or
13708 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
13710 @code{initial:} state shows some random possible calling sequence @value{GDBN}
13711 has found. It then finds another possible calling sequence - that one is
13712 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
13713 printed as the @code{reduced:} calling sequence. That one could have many
13714 further @code{compare:} and @code{reduced:} statements as long as there remain
13715 any non-ambiguous sequence entries.
13717 For the frame of function @code{b} in both cases there are different possible
13718 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
13719 also ambiguous. The only non-ambiguous frame is the one for function @code{a},
13720 therefore this one is displayed to the user while the ambiguous frames are
13723 There can be also reasons why printing of frame argument values at function
13728 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
13729 static void __attribute__((noinline, noclone)) a (int i);
13730 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
13731 static void __attribute__((noinline, noclone)) a (int i)
13732 @{ if (i) b (i - 1); else c (0); @}
13733 int main (void) @{ a (5); return 0; @}
13736 #0 c (i=i@@entry=0) at t.c:2
13737 #1 0x0000000000400428 in a (DW_OP_entry_value resolving has found
13738 function "a" at 0x400420 can call itself via tail calls
13739 i=<optimized out>) at t.c:6
13740 #2 0x000000000040036e in main () at t.c:7
13743 @value{GDBN} cannot find out from the inferior state if and how many times did
13744 function @code{a} call itself (via function @code{b}) as these calls would be
13745 tail calls. Such tail calls would modify the @code{i} variable, therefore
13746 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
13747 prints @code{<optimized out>} instead.
13750 @chapter C Preprocessor Macros
13752 Some languages, such as C and C@t{++}, provide a way to define and invoke
13753 ``preprocessor macros'' which expand into strings of tokens.
13754 @value{GDBN} can evaluate expressions containing macro invocations, show
13755 the result of macro expansion, and show a macro's definition, including
13756 where it was defined.
13758 You may need to compile your program specially to provide @value{GDBN}
13759 with information about preprocessor macros. Most compilers do not
13760 include macros in their debugging information, even when you compile
13761 with the @option{-g} flag. @xref{Compilation}.
13763 A program may define a macro at one point, remove that definition later,
13764 and then provide a different definition after that. Thus, at different
13765 points in the program, a macro may have different definitions, or have
13766 no definition at all. If there is a current stack frame, @value{GDBN}
13767 uses the macros in scope at that frame's source code line. Otherwise,
13768 @value{GDBN} uses the macros in scope at the current listing location;
13771 Whenever @value{GDBN} evaluates an expression, it always expands any
13772 macro invocations present in the expression. @value{GDBN} also provides
13773 the following commands for working with macros explicitly.
13777 @kindex macro expand
13778 @cindex macro expansion, showing the results of preprocessor
13779 @cindex preprocessor macro expansion, showing the results of
13780 @cindex expanding preprocessor macros
13781 @item macro expand @var{expression}
13782 @itemx macro exp @var{expression}
13783 Show the results of expanding all preprocessor macro invocations in
13784 @var{expression}. Since @value{GDBN} simply expands macros, but does
13785 not parse the result, @var{expression} need not be a valid expression;
13786 it can be any string of tokens.
13789 @item macro expand-once @var{expression}
13790 @itemx macro exp1 @var{expression}
13791 @cindex expand macro once
13792 @i{(This command is not yet implemented.)} Show the results of
13793 expanding those preprocessor macro invocations that appear explicitly in
13794 @var{expression}. Macro invocations appearing in that expansion are
13795 left unchanged. This command allows you to see the effect of a
13796 particular macro more clearly, without being confused by further
13797 expansions. Since @value{GDBN} simply expands macros, but does not
13798 parse the result, @var{expression} need not be a valid expression; it
13799 can be any string of tokens.
13802 @cindex macro definition, showing
13803 @cindex definition of a macro, showing
13804 @cindex macros, from debug info
13805 @item info macro [-a|-all] [--] @var{macro}
13806 Show the current definition or all definitions of the named @var{macro},
13807 and describe the source location or compiler command-line where that
13808 definition was established. The optional double dash is to signify the end of
13809 argument processing and the beginning of @var{macro} for non C-like macros where
13810 the macro may begin with a hyphen.
13812 @kindex info macros
13813 @item info macros @var{location}
13814 Show all macro definitions that are in effect at the location specified
13815 by @var{location}, and describe the source location or compiler
13816 command-line where those definitions were established.
13818 @kindex macro define
13819 @cindex user-defined macros
13820 @cindex defining macros interactively
13821 @cindex macros, user-defined
13822 @item macro define @var{macro} @var{replacement-list}
13823 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
13824 Introduce a definition for a preprocessor macro named @var{macro},
13825 invocations of which are replaced by the tokens given in
13826 @var{replacement-list}. The first form of this command defines an
13827 ``object-like'' macro, which takes no arguments; the second form
13828 defines a ``function-like'' macro, which takes the arguments given in
13831 A definition introduced by this command is in scope in every
13832 expression evaluated in @value{GDBN}, until it is removed with the
13833 @code{macro undef} command, described below. The definition overrides
13834 all definitions for @var{macro} present in the program being debugged,
13835 as well as any previous user-supplied definition.
13837 @kindex macro undef
13838 @item macro undef @var{macro}
13839 Remove any user-supplied definition for the macro named @var{macro}.
13840 This command only affects definitions provided with the @code{macro
13841 define} command, described above; it cannot remove definitions present
13842 in the program being debugged.
13846 List all the macros defined using the @code{macro define} command.
13849 @cindex macros, example of debugging with
13850 Here is a transcript showing the above commands in action. First, we
13851 show our source files:
13856 #include "sample.h"
13859 #define ADD(x) (M + x)
13864 printf ("Hello, world!\n");
13866 printf ("We're so creative.\n");
13868 printf ("Goodbye, world!\n");
13875 Now, we compile the program using the @sc{gnu} C compiler,
13876 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
13877 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
13878 and @option{-gdwarf-4}; we recommend always choosing the most recent
13879 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
13880 includes information about preprocessor macros in the debugging
13884 $ gcc -gdwarf-2 -g3 sample.c -o sample
13888 Now, we start @value{GDBN} on our sample program:
13892 GNU gdb 2002-05-06-cvs
13893 Copyright 2002 Free Software Foundation, Inc.
13894 GDB is free software, @dots{}
13898 We can expand macros and examine their definitions, even when the
13899 program is not running. @value{GDBN} uses the current listing position
13900 to decide which macro definitions are in scope:
13903 (@value{GDBP}) list main
13906 5 #define ADD(x) (M + x)
13911 10 printf ("Hello, world!\n");
13913 12 printf ("We're so creative.\n");
13914 (@value{GDBP}) info macro ADD
13915 Defined at /home/jimb/gdb/macros/play/sample.c:5
13916 #define ADD(x) (M + x)
13917 (@value{GDBP}) info macro Q
13918 Defined at /home/jimb/gdb/macros/play/sample.h:1
13919 included at /home/jimb/gdb/macros/play/sample.c:2
13921 (@value{GDBP}) macro expand ADD(1)
13922 expands to: (42 + 1)
13923 (@value{GDBP}) macro expand-once ADD(1)
13924 expands to: once (M + 1)
13928 In the example above, note that @code{macro expand-once} expands only
13929 the macro invocation explicit in the original text --- the invocation of
13930 @code{ADD} --- but does not expand the invocation of the macro @code{M},
13931 which was introduced by @code{ADD}.
13933 Once the program is running, @value{GDBN} uses the macro definitions in
13934 force at the source line of the current stack frame:
13937 (@value{GDBP}) break main
13938 Breakpoint 1 at 0x8048370: file sample.c, line 10.
13940 Starting program: /home/jimb/gdb/macros/play/sample
13942 Breakpoint 1, main () at sample.c:10
13943 10 printf ("Hello, world!\n");
13947 At line 10, the definition of the macro @code{N} at line 9 is in force:
13950 (@value{GDBP}) info macro N
13951 Defined at /home/jimb/gdb/macros/play/sample.c:9
13953 (@value{GDBP}) macro expand N Q M
13954 expands to: 28 < 42
13955 (@value{GDBP}) print N Q M
13960 As we step over directives that remove @code{N}'s definition, and then
13961 give it a new definition, @value{GDBN} finds the definition (or lack
13962 thereof) in force at each point:
13965 (@value{GDBP}) next
13967 12 printf ("We're so creative.\n");
13968 (@value{GDBP}) info macro N
13969 The symbol `N' has no definition as a C/C++ preprocessor macro
13970 at /home/jimb/gdb/macros/play/sample.c:12
13971 (@value{GDBP}) next
13973 14 printf ("Goodbye, world!\n");
13974 (@value{GDBP}) info macro N
13975 Defined at /home/jimb/gdb/macros/play/sample.c:13
13977 (@value{GDBP}) macro expand N Q M
13978 expands to: 1729 < 42
13979 (@value{GDBP}) print N Q M
13984 In addition to source files, macros can be defined on the compilation command
13985 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
13986 such a way, @value{GDBN} displays the location of their definition as line zero
13987 of the source file submitted to the compiler.
13990 (@value{GDBP}) info macro __STDC__
13991 Defined at /home/jimb/gdb/macros/play/sample.c:0
13998 @chapter Tracepoints
13999 @c This chapter is based on the documentation written by Michael
14000 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
14002 @cindex tracepoints
14003 In some applications, it is not feasible for the debugger to interrupt
14004 the program's execution long enough for the developer to learn
14005 anything helpful about its behavior. If the program's correctness
14006 depends on its real-time behavior, delays introduced by a debugger
14007 might cause the program to change its behavior drastically, or perhaps
14008 fail, even when the code itself is correct. It is useful to be able
14009 to observe the program's behavior without interrupting it.
14011 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
14012 specify locations in the program, called @dfn{tracepoints}, and
14013 arbitrary expressions to evaluate when those tracepoints are reached.
14014 Later, using the @code{tfind} command, you can examine the values
14015 those expressions had when the program hit the tracepoints. The
14016 expressions may also denote objects in memory---structures or arrays,
14017 for example---whose values @value{GDBN} should record; while visiting
14018 a particular tracepoint, you may inspect those objects as if they were
14019 in memory at that moment. However, because @value{GDBN} records these
14020 values without interacting with you, it can do so quickly and
14021 unobtrusively, hopefully not disturbing the program's behavior.
14023 The tracepoint facility is currently available only for remote
14024 targets. @xref{Targets}. In addition, your remote target must know
14025 how to collect trace data. This functionality is implemented in the
14026 remote stub; however, none of the stubs distributed with @value{GDBN}
14027 support tracepoints as of this writing. The format of the remote
14028 packets used to implement tracepoints are described in @ref{Tracepoint
14031 It is also possible to get trace data from a file, in a manner reminiscent
14032 of corefiles; you specify the filename, and use @code{tfind} to search
14033 through the file. @xref{Trace Files}, for more details.
14035 This chapter describes the tracepoint commands and features.
14038 * Set Tracepoints::
14039 * Analyze Collected Data::
14040 * Tracepoint Variables::
14044 @node Set Tracepoints
14045 @section Commands to Set Tracepoints
14047 Before running such a @dfn{trace experiment}, an arbitrary number of
14048 tracepoints can be set. A tracepoint is actually a special type of
14049 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
14050 standard breakpoint commands. For instance, as with breakpoints,
14051 tracepoint numbers are successive integers starting from one, and many
14052 of the commands associated with tracepoints take the tracepoint number
14053 as their argument, to identify which tracepoint to work on.
14055 For each tracepoint, you can specify, in advance, some arbitrary set
14056 of data that you want the target to collect in the trace buffer when
14057 it hits that tracepoint. The collected data can include registers,
14058 local variables, or global data. Later, you can use @value{GDBN}
14059 commands to examine the values these data had at the time the
14060 tracepoint was hit.
14062 Tracepoints do not support every breakpoint feature. Ignore counts on
14063 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
14064 commands when they are hit. Tracepoints may not be thread-specific
14067 @cindex fast tracepoints
14068 Some targets may support @dfn{fast tracepoints}, which are inserted in
14069 a different way (such as with a jump instead of a trap), that is
14070 faster but possibly restricted in where they may be installed.
14072 @cindex static tracepoints
14073 @cindex markers, static tracepoints
14074 @cindex probing markers, static tracepoints
14075 Regular and fast tracepoints are dynamic tracing facilities, meaning
14076 that they can be used to insert tracepoints at (almost) any location
14077 in the target. Some targets may also support controlling @dfn{static
14078 tracepoints} from @value{GDBN}. With static tracing, a set of
14079 instrumentation points, also known as @dfn{markers}, are embedded in
14080 the target program, and can be activated or deactivated by name or
14081 address. These are usually placed at locations which facilitate
14082 investigating what the target is actually doing. @value{GDBN}'s
14083 support for static tracing includes being able to list instrumentation
14084 points, and attach them with @value{GDBN} defined high level
14085 tracepoints that expose the whole range of convenience of
14086 @value{GDBN}'s tracepoints support. Namely, support for collecting
14087 registers values and values of global or local (to the instrumentation
14088 point) variables; tracepoint conditions and trace state variables.
14089 The act of installing a @value{GDBN} static tracepoint on an
14090 instrumentation point, or marker, is referred to as @dfn{probing} a
14091 static tracepoint marker.
14093 @code{gdbserver} supports tracepoints on some target systems.
14094 @xref{Server,,Tracepoints support in @code{gdbserver}}.
14096 This section describes commands to set tracepoints and associated
14097 conditions and actions.
14100 * Create and Delete Tracepoints::
14101 * Enable and Disable Tracepoints::
14102 * Tracepoint Passcounts::
14103 * Tracepoint Conditions::
14104 * Trace State Variables::
14105 * Tracepoint Actions::
14106 * Listing Tracepoints::
14107 * Listing Static Tracepoint Markers::
14108 * Starting and Stopping Trace Experiments::
14109 * Tracepoint Restrictions::
14112 @node Create and Delete Tracepoints
14113 @subsection Create and Delete Tracepoints
14116 @cindex set tracepoint
14118 @item trace @var{location}
14119 The @code{trace} command is very similar to the @code{break} command.
14120 Its argument @var{location} can be any valid location.
14121 @xref{Specify Location}. The @code{trace} command defines a tracepoint,
14122 which is a point in the target program where the debugger will briefly stop,
14123 collect some data, and then allow the program to continue. Setting a tracepoint
14124 or changing its actions takes effect immediately if the remote stub
14125 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
14127 If remote stub doesn't support the @samp{InstallInTrace} feature, all
14128 these changes don't take effect until the next @code{tstart}
14129 command, and once a trace experiment is running, further changes will
14130 not have any effect until the next trace experiment starts. In addition,
14131 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
14132 address is not yet resolved. (This is similar to pending breakpoints.)
14133 Pending tracepoints are not downloaded to the target and not installed
14134 until they are resolved. The resolution of pending tracepoints requires
14135 @value{GDBN} support---when debugging with the remote target, and
14136 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
14137 tracing}), pending tracepoints can not be resolved (and downloaded to
14138 the remote stub) while @value{GDBN} is disconnected.
14140 Here are some examples of using the @code{trace} command:
14143 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
14145 (@value{GDBP}) @b{trace +2} // 2 lines forward
14147 (@value{GDBP}) @b{trace my_function} // first source line of function
14149 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
14151 (@value{GDBP}) @b{trace *0x2117c4} // an address
14155 You can abbreviate @code{trace} as @code{tr}.
14157 @item trace @var{location} if @var{cond}
14158 Set a tracepoint with condition @var{cond}; evaluate the expression
14159 @var{cond} each time the tracepoint is reached, and collect data only
14160 if the value is nonzero---that is, if @var{cond} evaluates as true.
14161 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
14162 information on tracepoint conditions.
14164 @item ftrace @var{location} [ if @var{cond} ]
14165 @cindex set fast tracepoint
14166 @cindex fast tracepoints, setting
14168 The @code{ftrace} command sets a fast tracepoint. For targets that
14169 support them, fast tracepoints will use a more efficient but possibly
14170 less general technique to trigger data collection, such as a jump
14171 instruction instead of a trap, or some sort of hardware support. It
14172 may not be possible to create a fast tracepoint at the desired
14173 location, in which case the command will exit with an explanatory
14176 @value{GDBN} handles arguments to @code{ftrace} exactly as for
14179 On 32-bit x86-architecture systems, fast tracepoints normally need to
14180 be placed at an instruction that is 5 bytes or longer, but can be
14181 placed at 4-byte instructions if the low 64K of memory of the target
14182 program is available to install trampolines. Some Unix-type systems,
14183 such as @sc{gnu}/Linux, exclude low addresses from the program's
14184 address space; but for instance with the Linux kernel it is possible
14185 to let @value{GDBN} use this area by doing a @command{sysctl} command
14186 to set the @code{mmap_min_addr} kernel parameter, as in
14189 sudo sysctl -w vm.mmap_min_addr=32768
14193 which sets the low address to 32K, which leaves plenty of room for
14194 trampolines. The minimum address should be set to a page boundary.
14196 @item strace @var{location} [ if @var{cond} ]
14197 @cindex set static tracepoint
14198 @cindex static tracepoints, setting
14199 @cindex probe static tracepoint marker
14201 The @code{strace} command sets a static tracepoint. For targets that
14202 support it, setting a static tracepoint probes a static
14203 instrumentation point, or marker, found at @var{location}. It may not
14204 be possible to set a static tracepoint at the desired location, in
14205 which case the command will exit with an explanatory message.
14207 @value{GDBN} handles arguments to @code{strace} exactly as for
14208 @code{trace}, with the addition that the user can also specify
14209 @code{-m @var{marker}} as @var{location}. This probes the marker
14210 identified by the @var{marker} string identifier. This identifier
14211 depends on the static tracepoint backend library your program is
14212 using. You can find all the marker identifiers in the @samp{ID} field
14213 of the @code{info static-tracepoint-markers} command output.
14214 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
14215 Markers}. For example, in the following small program using the UST
14221 trace_mark(ust, bar33, "str %s", "FOOBAZ");
14226 the marker id is composed of joining the first two arguments to the
14227 @code{trace_mark} call with a slash, which translates to:
14230 (@value{GDBP}) info static-tracepoint-markers
14231 Cnt Enb ID Address What
14232 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
14238 so you may probe the marker above with:
14241 (@value{GDBP}) strace -m ust/bar33
14244 Static tracepoints accept an extra collect action --- @code{collect
14245 $_sdata}. This collects arbitrary user data passed in the probe point
14246 call to the tracing library. In the UST example above, you'll see
14247 that the third argument to @code{trace_mark} is a printf-like format
14248 string. The user data is then the result of running that formatting
14249 string against the following arguments. Note that @code{info
14250 static-tracepoint-markers} command output lists that format string in
14251 the @samp{Data:} field.
14253 You can inspect this data when analyzing the trace buffer, by printing
14254 the $_sdata variable like any other variable available to
14255 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
14258 @cindex last tracepoint number
14259 @cindex recent tracepoint number
14260 @cindex tracepoint number
14261 The convenience variable @code{$tpnum} records the tracepoint number
14262 of the most recently set tracepoint.
14264 @kindex delete tracepoint
14265 @cindex tracepoint deletion
14266 @item delete tracepoint @r{[}@var{num}@r{]}
14267 Permanently delete one or more tracepoints. With no argument, the
14268 default is to delete all tracepoints. Note that the regular
14269 @code{delete} command can remove tracepoints also.
14274 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
14276 (@value{GDBP}) @b{delete trace} // remove all tracepoints
14280 You can abbreviate this command as @code{del tr}.
14283 @node Enable and Disable Tracepoints
14284 @subsection Enable and Disable Tracepoints
14286 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
14289 @kindex disable tracepoint
14290 @item disable tracepoint @r{[}@var{num}@r{]}
14291 Disable tracepoint @var{num}, or all tracepoints if no argument
14292 @var{num} is given. A disabled tracepoint will have no effect during
14293 a trace experiment, but it is not forgotten. You can re-enable
14294 a disabled tracepoint using the @code{enable tracepoint} command.
14295 If the command is issued during a trace experiment and the debug target
14296 has support for disabling tracepoints during a trace experiment, then the
14297 change will be effective immediately. Otherwise, it will be applied to the
14298 next trace experiment.
14300 @kindex enable tracepoint
14301 @item enable tracepoint @r{[}@var{num}@r{]}
14302 Enable tracepoint @var{num}, or all tracepoints. If this command is
14303 issued during a trace experiment and the debug target supports enabling
14304 tracepoints during a trace experiment, then the enabled tracepoints will
14305 become effective immediately. Otherwise, they will become effective the
14306 next time a trace experiment is run.
14309 @node Tracepoint Passcounts
14310 @subsection Tracepoint Passcounts
14314 @cindex tracepoint pass count
14315 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
14316 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
14317 automatically stop a trace experiment. If a tracepoint's passcount is
14318 @var{n}, then the trace experiment will be automatically stopped on
14319 the @var{n}'th time that tracepoint is hit. If the tracepoint number
14320 @var{num} is not specified, the @code{passcount} command sets the
14321 passcount of the most recently defined tracepoint. If no passcount is
14322 given, the trace experiment will run until stopped explicitly by the
14328 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
14329 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
14331 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
14332 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
14333 (@value{GDBP}) @b{trace foo}
14334 (@value{GDBP}) @b{pass 3}
14335 (@value{GDBP}) @b{trace bar}
14336 (@value{GDBP}) @b{pass 2}
14337 (@value{GDBP}) @b{trace baz}
14338 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
14339 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
14340 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
14341 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
14345 @node Tracepoint Conditions
14346 @subsection Tracepoint Conditions
14347 @cindex conditional tracepoints
14348 @cindex tracepoint conditions
14350 The simplest sort of tracepoint collects data every time your program
14351 reaches a specified place. You can also specify a @dfn{condition} for
14352 a tracepoint. A condition is just a Boolean expression in your
14353 programming language (@pxref{Expressions, ,Expressions}). A
14354 tracepoint with a condition evaluates the expression each time your
14355 program reaches it, and data collection happens only if the condition
14358 Tracepoint conditions can be specified when a tracepoint is set, by
14359 using @samp{if} in the arguments to the @code{trace} command.
14360 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
14361 also be set or changed at any time with the @code{condition} command,
14362 just as with breakpoints.
14364 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
14365 the conditional expression itself. Instead, @value{GDBN} encodes the
14366 expression into an agent expression (@pxref{Agent Expressions})
14367 suitable for execution on the target, independently of @value{GDBN}.
14368 Global variables become raw memory locations, locals become stack
14369 accesses, and so forth.
14371 For instance, suppose you have a function that is usually called
14372 frequently, but should not be called after an error has occurred. You
14373 could use the following tracepoint command to collect data about calls
14374 of that function that happen while the error code is propagating
14375 through the program; an unconditional tracepoint could end up
14376 collecting thousands of useless trace frames that you would have to
14380 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
14383 @node Trace State Variables
14384 @subsection Trace State Variables
14385 @cindex trace state variables
14387 A @dfn{trace state variable} is a special type of variable that is
14388 created and managed by target-side code. The syntax is the same as
14389 that for GDB's convenience variables (a string prefixed with ``$''),
14390 but they are stored on the target. They must be created explicitly,
14391 using a @code{tvariable} command. They are always 64-bit signed
14394 Trace state variables are remembered by @value{GDBN}, and downloaded
14395 to the target along with tracepoint information when the trace
14396 experiment starts. There are no intrinsic limits on the number of
14397 trace state variables, beyond memory limitations of the target.
14399 @cindex convenience variables, and trace state variables
14400 Although trace state variables are managed by the target, you can use
14401 them in print commands and expressions as if they were convenience
14402 variables; @value{GDBN} will get the current value from the target
14403 while the trace experiment is running. Trace state variables share
14404 the same namespace as other ``$'' variables, which means that you
14405 cannot have trace state variables with names like @code{$23} or
14406 @code{$pc}, nor can you have a trace state variable and a convenience
14407 variable with the same name.
14411 @item tvariable $@var{name} [ = @var{expression} ]
14413 The @code{tvariable} command creates a new trace state variable named
14414 @code{$@var{name}}, and optionally gives it an initial value of
14415 @var{expression}. The @var{expression} is evaluated when this command is
14416 entered; the result will be converted to an integer if possible,
14417 otherwise @value{GDBN} will report an error. A subsequent
14418 @code{tvariable} command specifying the same name does not create a
14419 variable, but instead assigns the supplied initial value to the
14420 existing variable of that name, overwriting any previous initial
14421 value. The default initial value is 0.
14423 @item info tvariables
14424 @kindex info tvariables
14425 List all the trace state variables along with their initial values.
14426 Their current values may also be displayed, if the trace experiment is
14429 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
14430 @kindex delete tvariable
14431 Delete the given trace state variables, or all of them if no arguments
14436 @node Tracepoint Actions
14437 @subsection Tracepoint Action Lists
14441 @cindex tracepoint actions
14442 @item actions @r{[}@var{num}@r{]}
14443 This command will prompt for a list of actions to be taken when the
14444 tracepoint is hit. If the tracepoint number @var{num} is not
14445 specified, this command sets the actions for the one that was most
14446 recently defined (so that you can define a tracepoint and then say
14447 @code{actions} without bothering about its number). You specify the
14448 actions themselves on the following lines, one action at a time, and
14449 terminate the actions list with a line containing just @code{end}. So
14450 far, the only defined actions are @code{collect}, @code{teval}, and
14451 @code{while-stepping}.
14453 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
14454 Commands, ,Breakpoint Command Lists}), except that only the defined
14455 actions are allowed; any other @value{GDBN} command is rejected.
14457 @cindex remove actions from a tracepoint
14458 To remove all actions from a tracepoint, type @samp{actions @var{num}}
14459 and follow it immediately with @samp{end}.
14462 (@value{GDBP}) @b{collect @var{data}} // collect some data
14464 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
14466 (@value{GDBP}) @b{end} // signals the end of actions.
14469 In the following example, the action list begins with @code{collect}
14470 commands indicating the things to be collected when the tracepoint is
14471 hit. Then, in order to single-step and collect additional data
14472 following the tracepoint, a @code{while-stepping} command is used,
14473 followed by the list of things to be collected after each step in a
14474 sequence of single steps. The @code{while-stepping} command is
14475 terminated by its own separate @code{end} command. Lastly, the action
14476 list is terminated by an @code{end} command.
14479 (@value{GDBP}) @b{trace foo}
14480 (@value{GDBP}) @b{actions}
14481 Enter actions for tracepoint 1, one per line:
14484 > while-stepping 12
14485 > collect $pc, arr[i]
14490 @kindex collect @r{(tracepoints)}
14491 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
14492 Collect values of the given expressions when the tracepoint is hit.
14493 This command accepts a comma-separated list of any valid expressions.
14494 In addition to global, static, or local variables, the following
14495 special arguments are supported:
14499 Collect all registers.
14502 Collect all function arguments.
14505 Collect all local variables.
14508 Collect the return address. This is helpful if you want to see more
14511 @emph{Note:} The return address location can not always be reliably
14512 determined up front, and the wrong address / registers may end up
14513 collected instead. On some architectures the reliability is higher
14514 for tracepoints at function entry, while on others it's the opposite.
14515 When this happens, backtracing will stop because the return address is
14516 found unavailable (unless another collect rule happened to match it).
14519 Collects the number of arguments from the static probe at which the
14520 tracepoint is located.
14521 @xref{Static Probe Points}.
14523 @item $_probe_arg@var{n}
14524 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
14525 from the static probe at which the tracepoint is located.
14526 @xref{Static Probe Points}.
14529 @vindex $_sdata@r{, collect}
14530 Collect static tracepoint marker specific data. Only available for
14531 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
14532 Lists}. On the UST static tracepoints library backend, an
14533 instrumentation point resembles a @code{printf} function call. The
14534 tracing library is able to collect user specified data formatted to a
14535 character string using the format provided by the programmer that
14536 instrumented the program. Other backends have similar mechanisms.
14537 Here's an example of a UST marker call:
14540 const char master_name[] = "$your_name";
14541 trace_mark(channel1, marker1, "hello %s", master_name)
14544 In this case, collecting @code{$_sdata} collects the string
14545 @samp{hello $yourname}. When analyzing the trace buffer, you can
14546 inspect @samp{$_sdata} like any other variable available to
14550 You can give several consecutive @code{collect} commands, each one
14551 with a single argument, or one @code{collect} command with several
14552 arguments separated by commas; the effect is the same.
14554 The optional @var{mods} changes the usual handling of the arguments.
14555 @code{s} requests that pointers to chars be handled as strings, in
14556 particular collecting the contents of the memory being pointed at, up
14557 to the first zero. The upper bound is by default the value of the
14558 @code{print elements} variable; if @code{s} is followed by a decimal
14559 number, that is the upper bound instead. So for instance
14560 @samp{collect/s25 mystr} collects as many as 25 characters at
14563 The command @code{info scope} (@pxref{Symbols, info scope}) is
14564 particularly useful for figuring out what data to collect.
14566 @kindex teval @r{(tracepoints)}
14567 @item teval @var{expr1}, @var{expr2}, @dots{}
14568 Evaluate the given expressions when the tracepoint is hit. This
14569 command accepts a comma-separated list of expressions. The results
14570 are discarded, so this is mainly useful for assigning values to trace
14571 state variables (@pxref{Trace State Variables}) without adding those
14572 values to the trace buffer, as would be the case if the @code{collect}
14575 @kindex while-stepping @r{(tracepoints)}
14576 @item while-stepping @var{n}
14577 Perform @var{n} single-step instruction traces after the tracepoint,
14578 collecting new data after each step. The @code{while-stepping}
14579 command is followed by the list of what to collect while stepping
14580 (followed by its own @code{end} command):
14583 > while-stepping 12
14584 > collect $regs, myglobal
14590 Note that @code{$pc} is not automatically collected by
14591 @code{while-stepping}; you need to explicitly collect that register if
14592 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
14595 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
14596 @kindex set default-collect
14597 @cindex default collection action
14598 This variable is a list of expressions to collect at each tracepoint
14599 hit. It is effectively an additional @code{collect} action prepended
14600 to every tracepoint action list. The expressions are parsed
14601 individually for each tracepoint, so for instance a variable named
14602 @code{xyz} may be interpreted as a global for one tracepoint, and a
14603 local for another, as appropriate to the tracepoint's location.
14605 @item show default-collect
14606 @kindex show default-collect
14607 Show the list of expressions that are collected by default at each
14612 @node Listing Tracepoints
14613 @subsection Listing Tracepoints
14616 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
14617 @kindex info tp @r{[}@var{n}@dots{}@r{]}
14618 @cindex information about tracepoints
14619 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
14620 Display information about the tracepoint @var{num}. If you don't
14621 specify a tracepoint number, displays information about all the
14622 tracepoints defined so far. The format is similar to that used for
14623 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
14624 command, simply restricting itself to tracepoints.
14626 A tracepoint's listing may include additional information specific to
14631 its passcount as given by the @code{passcount @var{n}} command
14634 the state about installed on target of each location
14638 (@value{GDBP}) @b{info trace}
14639 Num Type Disp Enb Address What
14640 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
14642 collect globfoo, $regs
14647 2 tracepoint keep y <MULTIPLE>
14649 2.1 y 0x0804859c in func4 at change-loc.h:35
14650 installed on target
14651 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
14652 installed on target
14653 2.3 y <PENDING> set_tracepoint
14654 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
14655 not installed on target
14660 This command can be abbreviated @code{info tp}.
14663 @node Listing Static Tracepoint Markers
14664 @subsection Listing Static Tracepoint Markers
14667 @kindex info static-tracepoint-markers
14668 @cindex information about static tracepoint markers
14669 @item info static-tracepoint-markers
14670 Display information about all static tracepoint markers defined in the
14673 For each marker, the following columns are printed:
14677 An incrementing counter, output to help readability. This is not a
14680 The marker ID, as reported by the target.
14681 @item Enabled or Disabled
14682 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
14683 that are not enabled.
14685 Where the marker is in your program, as a memory address.
14687 Where the marker is in the source for your program, as a file and line
14688 number. If the debug information included in the program does not
14689 allow @value{GDBN} to locate the source of the marker, this column
14690 will be left blank.
14694 In addition, the following information may be printed for each marker:
14698 User data passed to the tracing library by the marker call. In the
14699 UST backend, this is the format string passed as argument to the
14701 @item Static tracepoints probing the marker
14702 The list of static tracepoints attached to the marker.
14706 (@value{GDBP}) info static-tracepoint-markers
14707 Cnt ID Enb Address What
14708 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
14709 Data: number1 %d number2 %d
14710 Probed by static tracepoints: #2
14711 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
14717 @node Starting and Stopping Trace Experiments
14718 @subsection Starting and Stopping Trace Experiments
14721 @kindex tstart [ @var{notes} ]
14722 @cindex start a new trace experiment
14723 @cindex collected data discarded
14725 This command starts the trace experiment, and begins collecting data.
14726 It has the side effect of discarding all the data collected in the
14727 trace buffer during the previous trace experiment. If any arguments
14728 are supplied, they are taken as a note and stored with the trace
14729 experiment's state. The notes may be arbitrary text, and are
14730 especially useful with disconnected tracing in a multi-user context;
14731 the notes can explain what the trace is doing, supply user contact
14732 information, and so forth.
14734 @kindex tstop [ @var{notes} ]
14735 @cindex stop a running trace experiment
14737 This command stops the trace experiment. If any arguments are
14738 supplied, they are recorded with the experiment as a note. This is
14739 useful if you are stopping a trace started by someone else, for
14740 instance if the trace is interfering with the system's behavior and
14741 needs to be stopped quickly.
14743 @strong{Note}: a trace experiment and data collection may stop
14744 automatically if any tracepoint's passcount is reached
14745 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
14748 @cindex status of trace data collection
14749 @cindex trace experiment, status of
14751 This command displays the status of the current trace data
14755 Here is an example of the commands we described so far:
14758 (@value{GDBP}) @b{trace gdb_c_test}
14759 (@value{GDBP}) @b{actions}
14760 Enter actions for tracepoint #1, one per line.
14761 > collect $regs,$locals,$args
14762 > while-stepping 11
14766 (@value{GDBP}) @b{tstart}
14767 [time passes @dots{}]
14768 (@value{GDBP}) @b{tstop}
14771 @anchor{disconnected tracing}
14772 @cindex disconnected tracing
14773 You can choose to continue running the trace experiment even if
14774 @value{GDBN} disconnects from the target, voluntarily or
14775 involuntarily. For commands such as @code{detach}, the debugger will
14776 ask what you want to do with the trace. But for unexpected
14777 terminations (@value{GDBN} crash, network outage), it would be
14778 unfortunate to lose hard-won trace data, so the variable
14779 @code{disconnected-tracing} lets you decide whether the trace should
14780 continue running without @value{GDBN}.
14783 @item set disconnected-tracing on
14784 @itemx set disconnected-tracing off
14785 @kindex set disconnected-tracing
14786 Choose whether a tracing run should continue to run if @value{GDBN}
14787 has disconnected from the target. Note that @code{detach} or
14788 @code{quit} will ask you directly what to do about a running trace no
14789 matter what this variable's setting, so the variable is mainly useful
14790 for handling unexpected situations, such as loss of the network.
14792 @item show disconnected-tracing
14793 @kindex show disconnected-tracing
14794 Show the current choice for disconnected tracing.
14798 When you reconnect to the target, the trace experiment may or may not
14799 still be running; it might have filled the trace buffer in the
14800 meantime, or stopped for one of the other reasons. If it is running,
14801 it will continue after reconnection.
14803 Upon reconnection, the target will upload information about the
14804 tracepoints in effect. @value{GDBN} will then compare that
14805 information to the set of tracepoints currently defined, and attempt
14806 to match them up, allowing for the possibility that the numbers may
14807 have changed due to creation and deletion in the meantime. If one of
14808 the target's tracepoints does not match any in @value{GDBN}, the
14809 debugger will create a new tracepoint, so that you have a number with
14810 which to specify that tracepoint. This matching-up process is
14811 necessarily heuristic, and it may result in useless tracepoints being
14812 created; you may simply delete them if they are of no use.
14814 @cindex circular trace buffer
14815 If your target agent supports a @dfn{circular trace buffer}, then you
14816 can run a trace experiment indefinitely without filling the trace
14817 buffer; when space runs out, the agent deletes already-collected trace
14818 frames, oldest first, until there is enough room to continue
14819 collecting. This is especially useful if your tracepoints are being
14820 hit too often, and your trace gets terminated prematurely because the
14821 buffer is full. To ask for a circular trace buffer, simply set
14822 @samp{circular-trace-buffer} to on. You can set this at any time,
14823 including during tracing; if the agent can do it, it will change
14824 buffer handling on the fly, otherwise it will not take effect until
14828 @item set circular-trace-buffer on
14829 @itemx set circular-trace-buffer off
14830 @kindex set circular-trace-buffer
14831 Choose whether a tracing run should use a linear or circular buffer
14832 for trace data. A linear buffer will not lose any trace data, but may
14833 fill up prematurely, while a circular buffer will discard old trace
14834 data, but it will have always room for the latest tracepoint hits.
14836 @item show circular-trace-buffer
14837 @kindex show circular-trace-buffer
14838 Show the current choice for the trace buffer. Note that this may not
14839 match the agent's current buffer handling, nor is it guaranteed to
14840 match the setting that might have been in effect during a past run,
14841 for instance if you are looking at frames from a trace file.
14846 @item set trace-buffer-size @var{n}
14847 @itemx set trace-buffer-size unlimited
14848 @kindex set trace-buffer-size
14849 Request that the target use a trace buffer of @var{n} bytes. Not all
14850 targets will honor the request; they may have a compiled-in size for
14851 the trace buffer, or some other limitation. Set to a value of
14852 @code{unlimited} or @code{-1} to let the target use whatever size it
14853 likes. This is also the default.
14855 @item show trace-buffer-size
14856 @kindex show trace-buffer-size
14857 Show the current requested size for the trace buffer. Note that this
14858 will only match the actual size if the target supports size-setting,
14859 and was able to handle the requested size. For instance, if the
14860 target can only change buffer size between runs, this variable will
14861 not reflect the change until the next run starts. Use @code{tstatus}
14862 to get a report of the actual buffer size.
14866 @item set trace-user @var{text}
14867 @kindex set trace-user
14869 @item show trace-user
14870 @kindex show trace-user
14872 @item set trace-notes @var{text}
14873 @kindex set trace-notes
14874 Set the trace run's notes.
14876 @item show trace-notes
14877 @kindex show trace-notes
14878 Show the trace run's notes.
14880 @item set trace-stop-notes @var{text}
14881 @kindex set trace-stop-notes
14882 Set the trace run's stop notes. The handling of the note is as for
14883 @code{tstop} arguments; the set command is convenient way to fix a
14884 stop note that is mistaken or incomplete.
14886 @item show trace-stop-notes
14887 @kindex show trace-stop-notes
14888 Show the trace run's stop notes.
14892 @node Tracepoint Restrictions
14893 @subsection Tracepoint Restrictions
14895 @cindex tracepoint restrictions
14896 There are a number of restrictions on the use of tracepoints. As
14897 described above, tracepoint data gathering occurs on the target
14898 without interaction from @value{GDBN}. Thus the full capabilities of
14899 the debugger are not available during data gathering, and then at data
14900 examination time, you will be limited by only having what was
14901 collected. The following items describe some common problems, but it
14902 is not exhaustive, and you may run into additional difficulties not
14908 Tracepoint expressions are intended to gather objects (lvalues). Thus
14909 the full flexibility of GDB's expression evaluator is not available.
14910 You cannot call functions, cast objects to aggregate types, access
14911 convenience variables or modify values (except by assignment to trace
14912 state variables). Some language features may implicitly call
14913 functions (for instance Objective-C fields with accessors), and therefore
14914 cannot be collected either.
14917 Collection of local variables, either individually or in bulk with
14918 @code{$locals} or @code{$args}, during @code{while-stepping} may
14919 behave erratically. The stepping action may enter a new scope (for
14920 instance by stepping into a function), or the location of the variable
14921 may change (for instance it is loaded into a register). The
14922 tracepoint data recorded uses the location information for the
14923 variables that is correct for the tracepoint location. When the
14924 tracepoint is created, it is not possible, in general, to determine
14925 where the steps of a @code{while-stepping} sequence will advance the
14926 program---particularly if a conditional branch is stepped.
14929 Collection of an incompletely-initialized or partially-destroyed object
14930 may result in something that @value{GDBN} cannot display, or displays
14931 in a misleading way.
14934 When @value{GDBN} displays a pointer to character it automatically
14935 dereferences the pointer to also display characters of the string
14936 being pointed to. However, collecting the pointer during tracing does
14937 not automatically collect the string. You need to explicitly
14938 dereference the pointer and provide size information if you want to
14939 collect not only the pointer, but the memory pointed to. For example,
14940 @code{*ptr@@50} can be used to collect the 50 element array pointed to
14944 It is not possible to collect a complete stack backtrace at a
14945 tracepoint. Instead, you may collect the registers and a few hundred
14946 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
14947 (adjust to use the name of the actual stack pointer register on your
14948 target architecture, and the amount of stack you wish to capture).
14949 Then the @code{backtrace} command will show a partial backtrace when
14950 using a trace frame. The number of stack frames that can be examined
14951 depends on the sizes of the frames in the collected stack. Note that
14952 if you ask for a block so large that it goes past the bottom of the
14953 stack, the target agent may report an error trying to read from an
14957 If you do not collect registers at a tracepoint, @value{GDBN} can
14958 infer that the value of @code{$pc} must be the same as the address of
14959 the tracepoint and use that when you are looking at a trace frame
14960 for that tracepoint. However, this cannot work if the tracepoint has
14961 multiple locations (for instance if it was set in a function that was
14962 inlined), or if it has a @code{while-stepping} loop. In those cases
14963 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
14968 @node Analyze Collected Data
14969 @section Using the Collected Data
14971 After the tracepoint experiment ends, you use @value{GDBN} commands
14972 for examining the trace data. The basic idea is that each tracepoint
14973 collects a trace @dfn{snapshot} every time it is hit and another
14974 snapshot every time it single-steps. All these snapshots are
14975 consecutively numbered from zero and go into a buffer, and you can
14976 examine them later. The way you examine them is to @dfn{focus} on a
14977 specific trace snapshot. When the remote stub is focused on a trace
14978 snapshot, it will respond to all @value{GDBN} requests for memory and
14979 registers by reading from the buffer which belongs to that snapshot,
14980 rather than from @emph{real} memory or registers of the program being
14981 debugged. This means that @strong{all} @value{GDBN} commands
14982 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
14983 behave as if we were currently debugging the program state as it was
14984 when the tracepoint occurred. Any requests for data that are not in
14985 the buffer will fail.
14988 * tfind:: How to select a trace snapshot
14989 * tdump:: How to display all data for a snapshot
14990 * save tracepoints:: How to save tracepoints for a future run
14994 @subsection @code{tfind @var{n}}
14997 @cindex select trace snapshot
14998 @cindex find trace snapshot
14999 The basic command for selecting a trace snapshot from the buffer is
15000 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
15001 counting from zero. If no argument @var{n} is given, the next
15002 snapshot is selected.
15004 Here are the various forms of using the @code{tfind} command.
15008 Find the first snapshot in the buffer. This is a synonym for
15009 @code{tfind 0} (since 0 is the number of the first snapshot).
15012 Stop debugging trace snapshots, resume @emph{live} debugging.
15015 Same as @samp{tfind none}.
15018 No argument means find the next trace snapshot or find the first
15019 one if no trace snapshot is selected.
15022 Find the previous trace snapshot before the current one. This permits
15023 retracing earlier steps.
15025 @item tfind tracepoint @var{num}
15026 Find the next snapshot associated with tracepoint @var{num}. Search
15027 proceeds forward from the last examined trace snapshot. If no
15028 argument @var{num} is given, it means find the next snapshot collected
15029 for the same tracepoint as the current snapshot.
15031 @item tfind pc @var{addr}
15032 Find the next snapshot associated with the value @var{addr} of the
15033 program counter. Search proceeds forward from the last examined trace
15034 snapshot. If no argument @var{addr} is given, it means find the next
15035 snapshot with the same value of PC as the current snapshot.
15037 @item tfind outside @var{addr1}, @var{addr2}
15038 Find the next snapshot whose PC is outside the given range of
15039 addresses (exclusive).
15041 @item tfind range @var{addr1}, @var{addr2}
15042 Find the next snapshot whose PC is between @var{addr1} and
15043 @var{addr2} (inclusive).
15045 @item tfind line @r{[}@var{file}:@r{]}@var{n}
15046 Find the next snapshot associated with the source line @var{n}. If
15047 the optional argument @var{file} is given, refer to line @var{n} in
15048 that source file. Search proceeds forward from the last examined
15049 trace snapshot. If no argument @var{n} is given, it means find the
15050 next line other than the one currently being examined; thus saying
15051 @code{tfind line} repeatedly can appear to have the same effect as
15052 stepping from line to line in a @emph{live} debugging session.
15055 The default arguments for the @code{tfind} commands are specifically
15056 designed to make it easy to scan through the trace buffer. For
15057 instance, @code{tfind} with no argument selects the next trace
15058 snapshot, and @code{tfind -} with no argument selects the previous
15059 trace snapshot. So, by giving one @code{tfind} command, and then
15060 simply hitting @key{RET} repeatedly you can examine all the trace
15061 snapshots in order. Or, by saying @code{tfind -} and then hitting
15062 @key{RET} repeatedly you can examine the snapshots in reverse order.
15063 The @code{tfind line} command with no argument selects the snapshot
15064 for the next source line executed. The @code{tfind pc} command with
15065 no argument selects the next snapshot with the same program counter
15066 (PC) as the current frame. The @code{tfind tracepoint} command with
15067 no argument selects the next trace snapshot collected by the same
15068 tracepoint as the current one.
15070 In addition to letting you scan through the trace buffer manually,
15071 these commands make it easy to construct @value{GDBN} scripts that
15072 scan through the trace buffer and print out whatever collected data
15073 you are interested in. Thus, if we want to examine the PC, FP, and SP
15074 registers from each trace frame in the buffer, we can say this:
15077 (@value{GDBP}) @b{tfind start}
15078 (@value{GDBP}) @b{while ($trace_frame != -1)}
15079 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
15080 $trace_frame, $pc, $sp, $fp
15084 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
15085 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
15086 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
15087 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
15088 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
15089 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
15090 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
15091 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
15092 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
15093 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
15094 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
15097 Or, if we want to examine the variable @code{X} at each source line in
15101 (@value{GDBP}) @b{tfind start}
15102 (@value{GDBP}) @b{while ($trace_frame != -1)}
15103 > printf "Frame %d, X == %d\n", $trace_frame, X
15113 @subsection @code{tdump}
15115 @cindex dump all data collected at tracepoint
15116 @cindex tracepoint data, display
15118 This command takes no arguments. It prints all the data collected at
15119 the current trace snapshot.
15122 (@value{GDBP}) @b{trace 444}
15123 (@value{GDBP}) @b{actions}
15124 Enter actions for tracepoint #2, one per line:
15125 > collect $regs, $locals, $args, gdb_long_test
15128 (@value{GDBP}) @b{tstart}
15130 (@value{GDBP}) @b{tfind line 444}
15131 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
15133 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
15135 (@value{GDBP}) @b{tdump}
15136 Data collected at tracepoint 2, trace frame 1:
15137 d0 0xc4aa0085 -995491707
15141 d4 0x71aea3d 119204413
15144 d7 0x380035 3670069
15145 a0 0x19e24a 1696330
15146 a1 0x3000668 50333288
15148 a3 0x322000 3284992
15149 a4 0x3000698 50333336
15150 a5 0x1ad3cc 1758156
15151 fp 0x30bf3c 0x30bf3c
15152 sp 0x30bf34 0x30bf34
15154 pc 0x20b2c8 0x20b2c8
15158 p = 0x20e5b4 "gdb-test"
15165 gdb_long_test = 17 '\021'
15170 @code{tdump} works by scanning the tracepoint's current collection
15171 actions and printing the value of each expression listed. So
15172 @code{tdump} can fail, if after a run, you change the tracepoint's
15173 actions to mention variables that were not collected during the run.
15175 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
15176 uses the collected value of @code{$pc} to distinguish between trace
15177 frames that were collected at the tracepoint hit, and frames that were
15178 collected while stepping. This allows it to correctly choose whether
15179 to display the basic list of collections, or the collections from the
15180 body of the while-stepping loop. However, if @code{$pc} was not collected,
15181 then @code{tdump} will always attempt to dump using the basic collection
15182 list, and may fail if a while-stepping frame does not include all the
15183 same data that is collected at the tracepoint hit.
15184 @c This is getting pretty arcane, example would be good.
15186 @node save tracepoints
15187 @subsection @code{save tracepoints @var{filename}}
15188 @kindex save tracepoints
15189 @kindex save-tracepoints
15190 @cindex save tracepoints for future sessions
15192 This command saves all current tracepoint definitions together with
15193 their actions and passcounts, into a file @file{@var{filename}}
15194 suitable for use in a later debugging session. To read the saved
15195 tracepoint definitions, use the @code{source} command (@pxref{Command
15196 Files}). The @w{@code{save-tracepoints}} command is a deprecated
15197 alias for @w{@code{save tracepoints}}
15199 @node Tracepoint Variables
15200 @section Convenience Variables for Tracepoints
15201 @cindex tracepoint variables
15202 @cindex convenience variables for tracepoints
15205 @vindex $trace_frame
15206 @item (int) $trace_frame
15207 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
15208 snapshot is selected.
15210 @vindex $tracepoint
15211 @item (int) $tracepoint
15212 The tracepoint for the current trace snapshot.
15214 @vindex $trace_line
15215 @item (int) $trace_line
15216 The line number for the current trace snapshot.
15218 @vindex $trace_file
15219 @item (char []) $trace_file
15220 The source file for the current trace snapshot.
15222 @vindex $trace_func
15223 @item (char []) $trace_func
15224 The name of the function containing @code{$tracepoint}.
15227 Note: @code{$trace_file} is not suitable for use in @code{printf},
15228 use @code{output} instead.
15230 Here's a simple example of using these convenience variables for
15231 stepping through all the trace snapshots and printing some of their
15232 data. Note that these are not the same as trace state variables,
15233 which are managed by the target.
15236 (@value{GDBP}) @b{tfind start}
15238 (@value{GDBP}) @b{while $trace_frame != -1}
15239 > output $trace_file
15240 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
15246 @section Using Trace Files
15247 @cindex trace files
15249 In some situations, the target running a trace experiment may no
15250 longer be available; perhaps it crashed, or the hardware was needed
15251 for a different activity. To handle these cases, you can arrange to
15252 dump the trace data into a file, and later use that file as a source
15253 of trace data, via the @code{target tfile} command.
15258 @item tsave [ -r ] @var{filename}
15259 @itemx tsave [-ctf] @var{dirname}
15260 Save the trace data to @var{filename}. By default, this command
15261 assumes that @var{filename} refers to the host filesystem, so if
15262 necessary @value{GDBN} will copy raw trace data up from the target and
15263 then save it. If the target supports it, you can also supply the
15264 optional argument @code{-r} (``remote'') to direct the target to save
15265 the data directly into @var{filename} in its own filesystem, which may be
15266 more efficient if the trace buffer is very large. (Note, however, that
15267 @code{target tfile} can only read from files accessible to the host.)
15268 By default, this command will save trace frame in tfile format.
15269 You can supply the optional argument @code{-ctf} to save data in CTF
15270 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
15271 that can be shared by multiple debugging and tracing tools. Please go to
15272 @indicateurl{http://www.efficios.com/ctf} to get more information.
15274 @kindex target tfile
15278 @item target tfile @var{filename}
15279 @itemx target ctf @var{dirname}
15280 Use the file named @var{filename} or directory named @var{dirname} as
15281 a source of trace data. Commands that examine data work as they do with
15282 a live target, but it is not possible to run any new trace experiments.
15283 @code{tstatus} will report the state of the trace run at the moment
15284 the data was saved, as well as the current trace frame you are examining.
15285 Both @var{filename} and @var{dirname} must be on a filesystem accessible to
15289 (@value{GDBP}) target ctf ctf.ctf
15290 (@value{GDBP}) tfind
15291 Found trace frame 0, tracepoint 2
15292 39 ++a; /* set tracepoint 1 here */
15293 (@value{GDBP}) tdump
15294 Data collected at tracepoint 2, trace frame 0:
15298 c = @{"123", "456", "789", "123", "456", "789"@}
15299 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
15307 @chapter Debugging Programs That Use Overlays
15310 If your program is too large to fit completely in your target system's
15311 memory, you can sometimes use @dfn{overlays} to work around this
15312 problem. @value{GDBN} provides some support for debugging programs that
15316 * How Overlays Work:: A general explanation of overlays.
15317 * Overlay Commands:: Managing overlays in @value{GDBN}.
15318 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
15319 mapped by asking the inferior.
15320 * Overlay Sample Program:: A sample program using overlays.
15323 @node How Overlays Work
15324 @section How Overlays Work
15325 @cindex mapped overlays
15326 @cindex unmapped overlays
15327 @cindex load address, overlay's
15328 @cindex mapped address
15329 @cindex overlay area
15331 Suppose you have a computer whose instruction address space is only 64
15332 kilobytes long, but which has much more memory which can be accessed by
15333 other means: special instructions, segment registers, or memory
15334 management hardware, for example. Suppose further that you want to
15335 adapt a program which is larger than 64 kilobytes to run on this system.
15337 One solution is to identify modules of your program which are relatively
15338 independent, and need not call each other directly; call these modules
15339 @dfn{overlays}. Separate the overlays from the main program, and place
15340 their machine code in the larger memory. Place your main program in
15341 instruction memory, but leave at least enough space there to hold the
15342 largest overlay as well.
15344 Now, to call a function located in an overlay, you must first copy that
15345 overlay's machine code from the large memory into the space set aside
15346 for it in the instruction memory, and then jump to its entry point
15349 @c NB: In the below the mapped area's size is greater or equal to the
15350 @c size of all overlays. This is intentional to remind the developer
15351 @c that overlays don't necessarily need to be the same size.
15355 Data Instruction Larger
15356 Address Space Address Space Address Space
15357 +-----------+ +-----------+ +-----------+
15359 +-----------+ +-----------+ +-----------+<-- overlay 1
15360 | program | | main | .----| overlay 1 | load address
15361 | variables | | program | | +-----------+
15362 | and heap | | | | | |
15363 +-----------+ | | | +-----------+<-- overlay 2
15364 | | +-----------+ | | | load address
15365 +-----------+ | | | .-| overlay 2 |
15367 mapped --->+-----------+ | | +-----------+
15368 address | | | | | |
15369 | overlay | <-' | | |
15370 | area | <---' +-----------+<-- overlay 3
15371 | | <---. | | load address
15372 +-----------+ `--| overlay 3 |
15379 @anchor{A code overlay}A code overlay
15383 The diagram (@pxref{A code overlay}) shows a system with separate data
15384 and instruction address spaces. To map an overlay, the program copies
15385 its code from the larger address space to the instruction address space.
15386 Since the overlays shown here all use the same mapped address, only one
15387 may be mapped at a time. For a system with a single address space for
15388 data and instructions, the diagram would be similar, except that the
15389 program variables and heap would share an address space with the main
15390 program and the overlay area.
15392 An overlay loaded into instruction memory and ready for use is called a
15393 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
15394 instruction memory. An overlay not present (or only partially present)
15395 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
15396 is its address in the larger memory. The mapped address is also called
15397 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
15398 called the @dfn{load memory address}, or @dfn{LMA}.
15400 Unfortunately, overlays are not a completely transparent way to adapt a
15401 program to limited instruction memory. They introduce a new set of
15402 global constraints you must keep in mind as you design your program:
15407 Before calling or returning to a function in an overlay, your program
15408 must make sure that overlay is actually mapped. Otherwise, the call or
15409 return will transfer control to the right address, but in the wrong
15410 overlay, and your program will probably crash.
15413 If the process of mapping an overlay is expensive on your system, you
15414 will need to choose your overlays carefully to minimize their effect on
15415 your program's performance.
15418 The executable file you load onto your system must contain each
15419 overlay's instructions, appearing at the overlay's load address, not its
15420 mapped address. However, each overlay's instructions must be relocated
15421 and its symbols defined as if the overlay were at its mapped address.
15422 You can use GNU linker scripts to specify different load and relocation
15423 addresses for pieces of your program; see @ref{Overlay Description,,,
15424 ld.info, Using ld: the GNU linker}.
15427 The procedure for loading executable files onto your system must be able
15428 to load their contents into the larger address space as well as the
15429 instruction and data spaces.
15433 The overlay system described above is rather simple, and could be
15434 improved in many ways:
15439 If your system has suitable bank switch registers or memory management
15440 hardware, you could use those facilities to make an overlay's load area
15441 contents simply appear at their mapped address in instruction space.
15442 This would probably be faster than copying the overlay to its mapped
15443 area in the usual way.
15446 If your overlays are small enough, you could set aside more than one
15447 overlay area, and have more than one overlay mapped at a time.
15450 You can use overlays to manage data, as well as instructions. In
15451 general, data overlays are even less transparent to your design than
15452 code overlays: whereas code overlays only require care when you call or
15453 return to functions, data overlays require care every time you access
15454 the data. Also, if you change the contents of a data overlay, you
15455 must copy its contents back out to its load address before you can copy a
15456 different data overlay into the same mapped area.
15461 @node Overlay Commands
15462 @section Overlay Commands
15464 To use @value{GDBN}'s overlay support, each overlay in your program must
15465 correspond to a separate section of the executable file. The section's
15466 virtual memory address and load memory address must be the overlay's
15467 mapped and load addresses. Identifying overlays with sections allows
15468 @value{GDBN} to determine the appropriate address of a function or
15469 variable, depending on whether the overlay is mapped or not.
15471 @value{GDBN}'s overlay commands all start with the word @code{overlay};
15472 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
15477 Disable @value{GDBN}'s overlay support. When overlay support is
15478 disabled, @value{GDBN} assumes that all functions and variables are
15479 always present at their mapped addresses. By default, @value{GDBN}'s
15480 overlay support is disabled.
15482 @item overlay manual
15483 @cindex manual overlay debugging
15484 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
15485 relies on you to tell it which overlays are mapped, and which are not,
15486 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
15487 commands described below.
15489 @item overlay map-overlay @var{overlay}
15490 @itemx overlay map @var{overlay}
15491 @cindex map an overlay
15492 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
15493 be the name of the object file section containing the overlay. When an
15494 overlay is mapped, @value{GDBN} assumes it can find the overlay's
15495 functions and variables at their mapped addresses. @value{GDBN} assumes
15496 that any other overlays whose mapped ranges overlap that of
15497 @var{overlay} are now unmapped.
15499 @item overlay unmap-overlay @var{overlay}
15500 @itemx overlay unmap @var{overlay}
15501 @cindex unmap an overlay
15502 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
15503 must be the name of the object file section containing the overlay.
15504 When an overlay is unmapped, @value{GDBN} assumes it can find the
15505 overlay's functions and variables at their load addresses.
15508 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
15509 consults a data structure the overlay manager maintains in the inferior
15510 to see which overlays are mapped. For details, see @ref{Automatic
15511 Overlay Debugging}.
15513 @item overlay load-target
15514 @itemx overlay load
15515 @cindex reloading the overlay table
15516 Re-read the overlay table from the inferior. Normally, @value{GDBN}
15517 re-reads the table @value{GDBN} automatically each time the inferior
15518 stops, so this command should only be necessary if you have changed the
15519 overlay mapping yourself using @value{GDBN}. This command is only
15520 useful when using automatic overlay debugging.
15522 @item overlay list-overlays
15523 @itemx overlay list
15524 @cindex listing mapped overlays
15525 Display a list of the overlays currently mapped, along with their mapped
15526 addresses, load addresses, and sizes.
15530 Normally, when @value{GDBN} prints a code address, it includes the name
15531 of the function the address falls in:
15534 (@value{GDBP}) print main
15535 $3 = @{int ()@} 0x11a0 <main>
15538 When overlay debugging is enabled, @value{GDBN} recognizes code in
15539 unmapped overlays, and prints the names of unmapped functions with
15540 asterisks around them. For example, if @code{foo} is a function in an
15541 unmapped overlay, @value{GDBN} prints it this way:
15544 (@value{GDBP}) overlay list
15545 No sections are mapped.
15546 (@value{GDBP}) print foo
15547 $5 = @{int (int)@} 0x100000 <*foo*>
15550 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
15554 (@value{GDBP}) overlay list
15555 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
15556 mapped at 0x1016 - 0x104a
15557 (@value{GDBP}) print foo
15558 $6 = @{int (int)@} 0x1016 <foo>
15561 When overlay debugging is enabled, @value{GDBN} can find the correct
15562 address for functions and variables in an overlay, whether or not the
15563 overlay is mapped. This allows most @value{GDBN} commands, like
15564 @code{break} and @code{disassemble}, to work normally, even on unmapped
15565 code. However, @value{GDBN}'s breakpoint support has some limitations:
15569 @cindex breakpoints in overlays
15570 @cindex overlays, setting breakpoints in
15571 You can set breakpoints in functions in unmapped overlays, as long as
15572 @value{GDBN} can write to the overlay at its load address.
15574 @value{GDBN} can not set hardware or simulator-based breakpoints in
15575 unmapped overlays. However, if you set a breakpoint at the end of your
15576 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
15577 you are using manual overlay management), @value{GDBN} will re-set its
15578 breakpoints properly.
15582 @node Automatic Overlay Debugging
15583 @section Automatic Overlay Debugging
15584 @cindex automatic overlay debugging
15586 @value{GDBN} can automatically track which overlays are mapped and which
15587 are not, given some simple co-operation from the overlay manager in the
15588 inferior. If you enable automatic overlay debugging with the
15589 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
15590 looks in the inferior's memory for certain variables describing the
15591 current state of the overlays.
15593 Here are the variables your overlay manager must define to support
15594 @value{GDBN}'s automatic overlay debugging:
15598 @item @code{_ovly_table}:
15599 This variable must be an array of the following structures:
15604 /* The overlay's mapped address. */
15607 /* The size of the overlay, in bytes. */
15608 unsigned long size;
15610 /* The overlay's load address. */
15613 /* Non-zero if the overlay is currently mapped;
15615 unsigned long mapped;
15619 @item @code{_novlys}:
15620 This variable must be a four-byte signed integer, holding the total
15621 number of elements in @code{_ovly_table}.
15625 To decide whether a particular overlay is mapped or not, @value{GDBN}
15626 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
15627 @code{lma} members equal the VMA and LMA of the overlay's section in the
15628 executable file. When @value{GDBN} finds a matching entry, it consults
15629 the entry's @code{mapped} member to determine whether the overlay is
15632 In addition, your overlay manager may define a function called
15633 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
15634 will silently set a breakpoint there. If the overlay manager then
15635 calls this function whenever it has changed the overlay table, this
15636 will enable @value{GDBN} to accurately keep track of which overlays
15637 are in program memory, and update any breakpoints that may be set
15638 in overlays. This will allow breakpoints to work even if the
15639 overlays are kept in ROM or other non-writable memory while they
15640 are not being executed.
15642 @node Overlay Sample Program
15643 @section Overlay Sample Program
15644 @cindex overlay example program
15646 When linking a program which uses overlays, you must place the overlays
15647 at their load addresses, while relocating them to run at their mapped
15648 addresses. To do this, you must write a linker script (@pxref{Overlay
15649 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
15650 since linker scripts are specific to a particular host system, target
15651 architecture, and target memory layout, this manual cannot provide
15652 portable sample code demonstrating @value{GDBN}'s overlay support.
15654 However, the @value{GDBN} source distribution does contain an overlaid
15655 program, with linker scripts for a few systems, as part of its test
15656 suite. The program consists of the following files from
15657 @file{gdb/testsuite/gdb.base}:
15661 The main program file.
15663 A simple overlay manager, used by @file{overlays.c}.
15668 Overlay modules, loaded and used by @file{overlays.c}.
15671 Linker scripts for linking the test program on the @code{d10v-elf}
15672 and @code{m32r-elf} targets.
15675 You can build the test program using the @code{d10v-elf} GCC
15676 cross-compiler like this:
15679 $ d10v-elf-gcc -g -c overlays.c
15680 $ d10v-elf-gcc -g -c ovlymgr.c
15681 $ d10v-elf-gcc -g -c foo.c
15682 $ d10v-elf-gcc -g -c bar.c
15683 $ d10v-elf-gcc -g -c baz.c
15684 $ d10v-elf-gcc -g -c grbx.c
15685 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
15686 baz.o grbx.o -Wl,-Td10v.ld -o overlays
15689 The build process is identical for any other architecture, except that
15690 you must substitute the appropriate compiler and linker script for the
15691 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
15695 @chapter Using @value{GDBN} with Different Languages
15698 Although programming languages generally have common aspects, they are
15699 rarely expressed in the same manner. For instance, in ANSI C,
15700 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
15701 Modula-2, it is accomplished by @code{p^}. Values can also be
15702 represented (and displayed) differently. Hex numbers in C appear as
15703 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
15705 @cindex working language
15706 Language-specific information is built into @value{GDBN} for some languages,
15707 allowing you to express operations like the above in your program's
15708 native language, and allowing @value{GDBN} to output values in a manner
15709 consistent with the syntax of your program's native language. The
15710 language you use to build expressions is called the @dfn{working
15714 * Setting:: Switching between source languages
15715 * Show:: Displaying the language
15716 * Checks:: Type and range checks
15717 * Supported Languages:: Supported languages
15718 * Unsupported Languages:: Unsupported languages
15722 @section Switching Between Source Languages
15724 There are two ways to control the working language---either have @value{GDBN}
15725 set it automatically, or select it manually yourself. You can use the
15726 @code{set language} command for either purpose. On startup, @value{GDBN}
15727 defaults to setting the language automatically. The working language is
15728 used to determine how expressions you type are interpreted, how values
15731 In addition to the working language, every source file that
15732 @value{GDBN} knows about has its own working language. For some object
15733 file formats, the compiler might indicate which language a particular
15734 source file is in. However, most of the time @value{GDBN} infers the
15735 language from the name of the file. The language of a source file
15736 controls whether C@t{++} names are demangled---this way @code{backtrace} can
15737 show each frame appropriately for its own language. There is no way to
15738 set the language of a source file from within @value{GDBN}, but you can
15739 set the language associated with a filename extension. @xref{Show, ,
15740 Displaying the Language}.
15742 This is most commonly a problem when you use a program, such
15743 as @code{cfront} or @code{f2c}, that generates C but is written in
15744 another language. In that case, make the
15745 program use @code{#line} directives in its C output; that way
15746 @value{GDBN} will know the correct language of the source code of the original
15747 program, and will display that source code, not the generated C code.
15750 * Filenames:: Filename extensions and languages.
15751 * Manually:: Setting the working language manually
15752 * Automatically:: Having @value{GDBN} infer the source language
15756 @subsection List of Filename Extensions and Languages
15758 If a source file name ends in one of the following extensions, then
15759 @value{GDBN} infers that its language is the one indicated.
15777 C@t{++} source file
15783 Objective-C source file
15787 Fortran source file
15790 Modula-2 source file
15794 Assembler source file. This actually behaves almost like C, but
15795 @value{GDBN} does not skip over function prologues when stepping.
15798 In addition, you may set the language associated with a filename
15799 extension. @xref{Show, , Displaying the Language}.
15802 @subsection Setting the Working Language
15804 If you allow @value{GDBN} to set the language automatically,
15805 expressions are interpreted the same way in your debugging session and
15808 @kindex set language
15809 If you wish, you may set the language manually. To do this, issue the
15810 command @samp{set language @var{lang}}, where @var{lang} is the name of
15811 a language, such as
15812 @code{c} or @code{modula-2}.
15813 For a list of the supported languages, type @samp{set language}.
15815 Setting the language manually prevents @value{GDBN} from updating the working
15816 language automatically. This can lead to confusion if you try
15817 to debug a program when the working language is not the same as the
15818 source language, when an expression is acceptable to both
15819 languages---but means different things. For instance, if the current
15820 source file were written in C, and @value{GDBN} was parsing Modula-2, a
15828 might not have the effect you intended. In C, this means to add
15829 @code{b} and @code{c} and place the result in @code{a}. The result
15830 printed would be the value of @code{a}. In Modula-2, this means to compare
15831 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
15833 @node Automatically
15834 @subsection Having @value{GDBN} Infer the Source Language
15836 To have @value{GDBN} set the working language automatically, use
15837 @samp{set language local} or @samp{set language auto}. @value{GDBN}
15838 then infers the working language. That is, when your program stops in a
15839 frame (usually by encountering a breakpoint), @value{GDBN} sets the
15840 working language to the language recorded for the function in that
15841 frame. If the language for a frame is unknown (that is, if the function
15842 or block corresponding to the frame was defined in a source file that
15843 does not have a recognized extension), the current working language is
15844 not changed, and @value{GDBN} issues a warning.
15846 This may not seem necessary for most programs, which are written
15847 entirely in one source language. However, program modules and libraries
15848 written in one source language can be used by a main program written in
15849 a different source language. Using @samp{set language auto} in this
15850 case frees you from having to set the working language manually.
15853 @section Displaying the Language
15855 The following commands help you find out which language is the
15856 working language, and also what language source files were written in.
15859 @item show language
15860 @anchor{show language}
15861 @kindex show language
15862 Display the current working language. This is the
15863 language you can use with commands such as @code{print} to
15864 build and compute expressions that may involve variables in your program.
15867 @kindex info frame@r{, show the source language}
15868 Display the source language for this frame. This language becomes the
15869 working language if you use an identifier from this frame.
15870 @xref{Frame Info, ,Information about a Frame}, to identify the other
15871 information listed here.
15874 @kindex info source@r{, show the source language}
15875 Display the source language of this source file.
15876 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
15877 information listed here.
15880 In unusual circumstances, you may have source files with extensions
15881 not in the standard list. You can then set the extension associated
15882 with a language explicitly:
15885 @item set extension-language @var{ext} @var{language}
15886 @kindex set extension-language
15887 Tell @value{GDBN} that source files with extension @var{ext} are to be
15888 assumed as written in the source language @var{language}.
15890 @item info extensions
15891 @kindex info extensions
15892 List all the filename extensions and the associated languages.
15896 @section Type and Range Checking
15898 Some languages are designed to guard you against making seemingly common
15899 errors through a series of compile- and run-time checks. These include
15900 checking the type of arguments to functions and operators and making
15901 sure mathematical overflows are caught at run time. Checks such as
15902 these help to ensure a program's correctness once it has been compiled
15903 by eliminating type mismatches and providing active checks for range
15904 errors when your program is running.
15906 By default @value{GDBN} checks for these errors according to the
15907 rules of the current source language. Although @value{GDBN} does not check
15908 the statements in your program, it can check expressions entered directly
15909 into @value{GDBN} for evaluation via the @code{print} command, for example.
15912 * Type Checking:: An overview of type checking
15913 * Range Checking:: An overview of range checking
15916 @cindex type checking
15917 @cindex checks, type
15918 @node Type Checking
15919 @subsection An Overview of Type Checking
15921 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
15922 arguments to operators and functions have to be of the correct type,
15923 otherwise an error occurs. These checks prevent type mismatch
15924 errors from ever causing any run-time problems. For example,
15927 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
15929 (@value{GDBP}) print obj.my_method (0)
15932 (@value{GDBP}) print obj.my_method (0x1234)
15933 Cannot resolve method klass::my_method to any overloaded instance
15936 The second example fails because in C@t{++} the integer constant
15937 @samp{0x1234} is not type-compatible with the pointer parameter type.
15939 For the expressions you use in @value{GDBN} commands, you can tell
15940 @value{GDBN} to not enforce strict type checking or
15941 to treat any mismatches as errors and abandon the expression;
15942 When type checking is disabled, @value{GDBN} successfully evaluates
15943 expressions like the second example above.
15945 Even if type checking is off, there may be other reasons
15946 related to type that prevent @value{GDBN} from evaluating an expression.
15947 For instance, @value{GDBN} does not know how to add an @code{int} and
15948 a @code{struct foo}. These particular type errors have nothing to do
15949 with the language in use and usually arise from expressions which make
15950 little sense to evaluate anyway.
15952 @value{GDBN} provides some additional commands for controlling type checking:
15954 @kindex set check type
15955 @kindex show check type
15957 @item set check type on
15958 @itemx set check type off
15959 Set strict type checking on or off. If any type mismatches occur in
15960 evaluating an expression while type checking is on, @value{GDBN} prints a
15961 message and aborts evaluation of the expression.
15963 @item show check type
15964 Show the current setting of type checking and whether @value{GDBN}
15965 is enforcing strict type checking rules.
15968 @cindex range checking
15969 @cindex checks, range
15970 @node Range Checking
15971 @subsection An Overview of Range Checking
15973 In some languages (such as Modula-2), it is an error to exceed the
15974 bounds of a type; this is enforced with run-time checks. Such range
15975 checking is meant to ensure program correctness by making sure
15976 computations do not overflow, or indices on an array element access do
15977 not exceed the bounds of the array.
15979 For expressions you use in @value{GDBN} commands, you can tell
15980 @value{GDBN} to treat range errors in one of three ways: ignore them,
15981 always treat them as errors and abandon the expression, or issue
15982 warnings but evaluate the expression anyway.
15984 A range error can result from numerical overflow, from exceeding an
15985 array index bound, or when you type a constant that is not a member
15986 of any type. Some languages, however, do not treat overflows as an
15987 error. In many implementations of C, mathematical overflow causes the
15988 result to ``wrap around'' to lower values---for example, if @var{m} is
15989 the largest integer value, and @var{s} is the smallest, then
15992 @var{m} + 1 @result{} @var{s}
15995 This, too, is specific to individual languages, and in some cases
15996 specific to individual compilers or machines. @xref{Supported Languages, ,
15997 Supported Languages}, for further details on specific languages.
15999 @value{GDBN} provides some additional commands for controlling the range checker:
16001 @kindex set check range
16002 @kindex show check range
16004 @item set check range auto
16005 Set range checking on or off based on the current working language.
16006 @xref{Supported Languages, ,Supported Languages}, for the default settings for
16009 @item set check range on
16010 @itemx set check range off
16011 Set range checking on or off, overriding the default setting for the
16012 current working language. A warning is issued if the setting does not
16013 match the language default. If a range error occurs and range checking is on,
16014 then a message is printed and evaluation of the expression is aborted.
16016 @item set check range warn
16017 Output messages when the @value{GDBN} range checker detects a range error,
16018 but attempt to evaluate the expression anyway. Evaluating the
16019 expression may still be impossible for other reasons, such as accessing
16020 memory that the process does not own (a typical example from many Unix
16024 Show the current setting of the range checker, and whether or not it is
16025 being set automatically by @value{GDBN}.
16028 @node Supported Languages
16029 @section Supported Languages
16031 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran,
16032 OpenCL C, Pascal, Rust, assembly, Modula-2, and Ada.
16033 @c This is false ...
16034 Some @value{GDBN} features may be used in expressions regardless of the
16035 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
16036 and the @samp{@{type@}addr} construct (@pxref{Expressions,
16037 ,Expressions}) can be used with the constructs of any supported
16040 The following sections detail to what degree each source language is
16041 supported by @value{GDBN}. These sections are not meant to be language
16042 tutorials or references, but serve only as a reference guide to what the
16043 @value{GDBN} expression parser accepts, and what input and output
16044 formats should look like for different languages. There are many good
16045 books written on each of these languages; please look to these for a
16046 language reference or tutorial.
16049 * C:: C and C@t{++}
16052 * Objective-C:: Objective-C
16053 * OpenCL C:: OpenCL C
16054 * Fortran:: Fortran
16057 * Modula-2:: Modula-2
16063 @subsection C and C@t{++}
16065 @cindex C and C@t{++}
16066 @cindex expressions in C or C@t{++}
16068 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
16069 to both languages. Whenever this is the case, we discuss those languages
16073 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
16074 @cindex @sc{gnu} C@t{++}
16075 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
16076 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
16077 effectively, you must compile your C@t{++} programs with a supported
16078 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
16079 compiler (@code{aCC}).
16082 * C Operators:: C and C@t{++} operators
16083 * C Constants:: C and C@t{++} constants
16084 * C Plus Plus Expressions:: C@t{++} expressions
16085 * C Defaults:: Default settings for C and C@t{++}
16086 * C Checks:: C and C@t{++} type and range checks
16087 * Debugging C:: @value{GDBN} and C
16088 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
16089 * Decimal Floating Point:: Numbers in Decimal Floating Point format
16093 @subsubsection C and C@t{++} Operators
16095 @cindex C and C@t{++} operators
16097 Operators must be defined on values of specific types. For instance,
16098 @code{+} is defined on numbers, but not on structures. Operators are
16099 often defined on groups of types.
16101 For the purposes of C and C@t{++}, the following definitions hold:
16106 @emph{Integral types} include @code{int} with any of its storage-class
16107 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
16110 @emph{Floating-point types} include @code{float}, @code{double}, and
16111 @code{long double} (if supported by the target platform).
16114 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
16117 @emph{Scalar types} include all of the above.
16122 The following operators are supported. They are listed here
16123 in order of increasing precedence:
16127 The comma or sequencing operator. Expressions in a comma-separated list
16128 are evaluated from left to right, with the result of the entire
16129 expression being the last expression evaluated.
16132 Assignment. The value of an assignment expression is the value
16133 assigned. Defined on scalar types.
16136 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
16137 and translated to @w{@code{@var{a} = @var{a op b}}}.
16138 @w{@code{@var{op}=}} and @code{=} have the same precedence. The operator
16139 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
16140 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
16143 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
16144 of as: if @var{a} then @var{b} else @var{c}. The argument @var{a}
16145 should be of an integral type.
16148 Logical @sc{or}. Defined on integral types.
16151 Logical @sc{and}. Defined on integral types.
16154 Bitwise @sc{or}. Defined on integral types.
16157 Bitwise exclusive-@sc{or}. Defined on integral types.
16160 Bitwise @sc{and}. Defined on integral types.
16163 Equality and inequality. Defined on scalar types. The value of these
16164 expressions is 0 for false and non-zero for true.
16166 @item <@r{, }>@r{, }<=@r{, }>=
16167 Less than, greater than, less than or equal, greater than or equal.
16168 Defined on scalar types. The value of these expressions is 0 for false
16169 and non-zero for true.
16172 left shift, and right shift. Defined on integral types.
16175 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
16178 Addition and subtraction. Defined on integral types, floating-point types and
16181 @item *@r{, }/@r{, }%
16182 Multiplication, division, and modulus. Multiplication and division are
16183 defined on integral and floating-point types. Modulus is defined on
16187 Increment and decrement. When appearing before a variable, the
16188 operation is performed before the variable is used in an expression;
16189 when appearing after it, the variable's value is used before the
16190 operation takes place.
16193 Pointer dereferencing. Defined on pointer types. Same precedence as
16197 Address operator. Defined on variables. Same precedence as @code{++}.
16199 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
16200 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
16201 to examine the address
16202 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
16206 Negative. Defined on integral and floating-point types. Same
16207 precedence as @code{++}.
16210 Logical negation. Defined on integral types. Same precedence as
16214 Bitwise complement operator. Defined on integral types. Same precedence as
16219 Structure member, and pointer-to-structure member. For convenience,
16220 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
16221 pointer based on the stored type information.
16222 Defined on @code{struct} and @code{union} data.
16225 Dereferences of pointers to members.
16228 Array indexing. @code{@var{a}[@var{i}]} is defined as
16229 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
16232 Function parameter list. Same precedence as @code{->}.
16235 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
16236 and @code{class} types.
16239 Doubled colons also represent the @value{GDBN} scope operator
16240 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
16244 If an operator is redefined in the user code, @value{GDBN} usually
16245 attempts to invoke the redefined version instead of using the operator's
16246 predefined meaning.
16249 @subsubsection C and C@t{++} Constants
16251 @cindex C and C@t{++} constants
16253 @value{GDBN} allows you to express the constants of C and C@t{++} in the
16258 Integer constants are a sequence of digits. Octal constants are
16259 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
16260 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
16261 @samp{l}, specifying that the constant should be treated as a
16265 Floating point constants are a sequence of digits, followed by a decimal
16266 point, followed by a sequence of digits, and optionally followed by an
16267 exponent. An exponent is of the form:
16268 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
16269 sequence of digits. The @samp{+} is optional for positive exponents.
16270 A floating-point constant may also end with a letter @samp{f} or
16271 @samp{F}, specifying that the constant should be treated as being of
16272 the @code{float} (as opposed to the default @code{double}) type; or with
16273 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
16277 Enumerated constants consist of enumerated identifiers, or their
16278 integral equivalents.
16281 Character constants are a single character surrounded by single quotes
16282 (@code{'}), or a number---the ordinal value of the corresponding character
16283 (usually its @sc{ascii} value). Within quotes, the single character may
16284 be represented by a letter or by @dfn{escape sequences}, which are of
16285 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
16286 of the character's ordinal value; or of the form @samp{\@var{x}}, where
16287 @samp{@var{x}} is a predefined special character---for example,
16288 @samp{\n} for newline.
16290 Wide character constants can be written by prefixing a character
16291 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
16292 form of @samp{x}. The target wide character set is used when
16293 computing the value of this constant (@pxref{Character Sets}).
16296 String constants are a sequence of character constants surrounded by
16297 double quotes (@code{"}). Any valid character constant (as described
16298 above) may appear. Double quotes within the string must be preceded by
16299 a backslash, so for instance @samp{"a\"b'c"} is a string of five
16302 Wide string constants can be written by prefixing a string constant
16303 with @samp{L}, as in C. The target wide character set is used when
16304 computing the value of this constant (@pxref{Character Sets}).
16307 Pointer constants are an integral value. You can also write pointers
16308 to constants using the C operator @samp{&}.
16311 Array constants are comma-separated lists surrounded by braces @samp{@{}
16312 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
16313 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
16314 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
16317 @node C Plus Plus Expressions
16318 @subsubsection C@t{++} Expressions
16320 @cindex expressions in C@t{++}
16321 @value{GDBN} expression handling can interpret most C@t{++} expressions.
16323 @cindex debugging C@t{++} programs
16324 @cindex C@t{++} compilers
16325 @cindex debug formats and C@t{++}
16326 @cindex @value{NGCC} and C@t{++}
16328 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
16329 the proper compiler and the proper debug format. Currently,
16330 @value{GDBN} works best when debugging C@t{++} code that is compiled
16331 with the most recent version of @value{NGCC} possible. The DWARF
16332 debugging format is preferred; @value{NGCC} defaults to this on most
16333 popular platforms. Other compilers and/or debug formats are likely to
16334 work badly or not at all when using @value{GDBN} to debug C@t{++}
16335 code. @xref{Compilation}.
16340 @cindex member functions
16342 Member function calls are allowed; you can use expressions like
16345 count = aml->GetOriginal(x, y)
16348 @vindex this@r{, inside C@t{++} member functions}
16349 @cindex namespace in C@t{++}
16351 While a member function is active (in the selected stack frame), your
16352 expressions have the same namespace available as the member function;
16353 that is, @value{GDBN} allows implicit references to the class instance
16354 pointer @code{this} following the same rules as C@t{++}. @code{using}
16355 declarations in the current scope are also respected by @value{GDBN}.
16357 @cindex call overloaded functions
16358 @cindex overloaded functions, calling
16359 @cindex type conversions in C@t{++}
16361 You can call overloaded functions; @value{GDBN} resolves the function
16362 call to the right definition, with some restrictions. @value{GDBN} does not
16363 perform overload resolution involving user-defined type conversions,
16364 calls to constructors, or instantiations of templates that do not exist
16365 in the program. It also cannot handle ellipsis argument lists or
16368 It does perform integral conversions and promotions, floating-point
16369 promotions, arithmetic conversions, pointer conversions, conversions of
16370 class objects to base classes, and standard conversions such as those of
16371 functions or arrays to pointers; it requires an exact match on the
16372 number of function arguments.
16374 Overload resolution is always performed, unless you have specified
16375 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
16376 ,@value{GDBN} Features for C@t{++}}.
16378 You must specify @code{set overload-resolution off} in order to use an
16379 explicit function signature to call an overloaded function, as in
16381 p 'foo(char,int)'('x', 13)
16384 The @value{GDBN} command-completion facility can simplify this;
16385 see @ref{Completion, ,Command Completion}.
16387 @cindex reference declarations
16389 @value{GDBN} understands variables declared as C@t{++} lvalue or rvalue
16390 references; you can use them in expressions just as you do in C@t{++}
16391 source---they are automatically dereferenced.
16393 In the parameter list shown when @value{GDBN} displays a frame, the values of
16394 reference variables are not displayed (unlike other variables); this
16395 avoids clutter, since references are often used for large structures.
16396 The @emph{address} of a reference variable is always shown, unless
16397 you have specified @samp{set print address off}.
16400 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
16401 expressions can use it just as expressions in your program do. Since
16402 one scope may be defined in another, you can use @code{::} repeatedly if
16403 necessary, for example in an expression like
16404 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
16405 resolving name scope by reference to source files, in both C and C@t{++}
16406 debugging (@pxref{Variables, ,Program Variables}).
16409 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
16414 @subsubsection C and C@t{++} Defaults
16416 @cindex C and C@t{++} defaults
16418 If you allow @value{GDBN} to set range checking automatically, it
16419 defaults to @code{off} whenever the working language changes to
16420 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
16421 selects the working language.
16423 If you allow @value{GDBN} to set the language automatically, it
16424 recognizes source files whose names end with @file{.c}, @file{.C}, or
16425 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
16426 these files, it sets the working language to C or C@t{++}.
16427 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
16428 for further details.
16431 @subsubsection C and C@t{++} Type and Range Checks
16433 @cindex C and C@t{++} checks
16435 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
16436 checking is used. However, if you turn type checking off, @value{GDBN}
16437 will allow certain non-standard conversions, such as promoting integer
16438 constants to pointers.
16440 Range checking, if turned on, is done on mathematical operations. Array
16441 indices are not checked, since they are often used to index a pointer
16442 that is not itself an array.
16445 @subsubsection @value{GDBN} and C
16447 The @code{set print union} and @code{show print union} commands apply to
16448 the @code{union} type. When set to @samp{on}, any @code{union} that is
16449 inside a @code{struct} or @code{class} is also printed. Otherwise, it
16450 appears as @samp{@{...@}}.
16452 The @code{@@} operator aids in the debugging of dynamic arrays, formed
16453 with pointers and a memory allocation function. @xref{Expressions,
16456 @node Debugging C Plus Plus
16457 @subsubsection @value{GDBN} Features for C@t{++}
16459 @cindex commands for C@t{++}
16461 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
16462 designed specifically for use with C@t{++}. Here is a summary:
16465 @cindex break in overloaded functions
16466 @item @r{breakpoint menus}
16467 When you want a breakpoint in a function whose name is overloaded,
16468 @value{GDBN} has the capability to display a menu of possible breakpoint
16469 locations to help you specify which function definition you want.
16470 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
16472 @cindex overloading in C@t{++}
16473 @item rbreak @var{regex}
16474 Setting breakpoints using regular expressions is helpful for setting
16475 breakpoints on overloaded functions that are not members of any special
16477 @xref{Set Breaks, ,Setting Breakpoints}.
16479 @cindex C@t{++} exception handling
16481 @itemx catch rethrow
16483 Debug C@t{++} exception handling using these commands. @xref{Set
16484 Catchpoints, , Setting Catchpoints}.
16486 @cindex inheritance
16487 @item ptype @var{typename}
16488 Print inheritance relationships as well as other information for type
16490 @xref{Symbols, ,Examining the Symbol Table}.
16492 @item info vtbl @var{expression}.
16493 The @code{info vtbl} command can be used to display the virtual
16494 method tables of the object computed by @var{expression}. This shows
16495 one entry per virtual table; there may be multiple virtual tables when
16496 multiple inheritance is in use.
16498 @cindex C@t{++} demangling
16499 @item demangle @var{name}
16500 Demangle @var{name}.
16501 @xref{Symbols}, for a more complete description of the @code{demangle} command.
16503 @cindex C@t{++} symbol display
16504 @item set print demangle
16505 @itemx show print demangle
16506 @itemx set print asm-demangle
16507 @itemx show print asm-demangle
16508 Control whether C@t{++} symbols display in their source form, both when
16509 displaying code as C@t{++} source and when displaying disassemblies.
16510 @xref{Print Settings, ,Print Settings}.
16512 @item set print object
16513 @itemx show print object
16514 Choose whether to print derived (actual) or declared types of objects.
16515 @xref{Print Settings, ,Print Settings}.
16517 @item set print vtbl
16518 @itemx show print vtbl
16519 Control the format for printing virtual function tables.
16520 @xref{Print Settings, ,Print Settings}.
16521 (The @code{vtbl} commands do not work on programs compiled with the HP
16522 ANSI C@t{++} compiler (@code{aCC}).)
16524 @kindex set overload-resolution
16525 @cindex overloaded functions, overload resolution
16526 @item set overload-resolution on
16527 Enable overload resolution for C@t{++} expression evaluation. The default
16528 is on. For overloaded functions, @value{GDBN} evaluates the arguments
16529 and searches for a function whose signature matches the argument types,
16530 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
16531 Expressions, ,C@t{++} Expressions}, for details).
16532 If it cannot find a match, it emits a message.
16534 @item set overload-resolution off
16535 Disable overload resolution for C@t{++} expression evaluation. For
16536 overloaded functions that are not class member functions, @value{GDBN}
16537 chooses the first function of the specified name that it finds in the
16538 symbol table, whether or not its arguments are of the correct type. For
16539 overloaded functions that are class member functions, @value{GDBN}
16540 searches for a function whose signature @emph{exactly} matches the
16543 @kindex show overload-resolution
16544 @item show overload-resolution
16545 Show the current setting of overload resolution.
16547 @item @r{Overloaded symbol names}
16548 You can specify a particular definition of an overloaded symbol, using
16549 the same notation that is used to declare such symbols in C@t{++}: type
16550 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
16551 also use the @value{GDBN} command-line word completion facilities to list the
16552 available choices, or to finish the type list for you.
16553 @xref{Completion,, Command Completion}, for details on how to do this.
16555 @item @r{Breakpoints in functions with ABI tags}
16557 The GNU C@t{++} compiler introduced the notion of ABI ``tags'', which
16558 correspond to changes in the ABI of a type, function, or variable that
16559 would not otherwise be reflected in a mangled name. See
16560 @url{https://developers.redhat.com/blog/2015/02/05/gcc5-and-the-c11-abi/}
16563 The ABI tags are visible in C@t{++} demangled names. For example, a
16564 function that returns a std::string:
16567 std::string function(int);
16571 when compiled for the C++11 ABI is marked with the @code{cxx11} ABI
16572 tag, and @value{GDBN} displays the symbol like this:
16575 function[abi:cxx11](int)
16578 You can set a breakpoint on such functions simply as if they had no
16582 (@value{GDBP}) b function(int)
16583 Breakpoint 2 at 0x40060d: file main.cc, line 10.
16584 (@value{GDBP}) info breakpoints
16585 Num Type Disp Enb Address What
16586 1 breakpoint keep y 0x0040060d in function[abi:cxx11](int)
16590 On the rare occasion you need to disambiguate between different ABI
16591 tags, you can do so by simply including the ABI tag in the function
16595 (@value{GDBP}) b ambiguous[abi:other_tag](int)
16599 @node Decimal Floating Point
16600 @subsubsection Decimal Floating Point format
16601 @cindex decimal floating point format
16603 @value{GDBN} can examine, set and perform computations with numbers in
16604 decimal floating point format, which in the C language correspond to the
16605 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
16606 specified by the extension to support decimal floating-point arithmetic.
16608 There are two encodings in use, depending on the architecture: BID (Binary
16609 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
16610 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
16613 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
16614 to manipulate decimal floating point numbers, it is not possible to convert
16615 (using a cast, for example) integers wider than 32-bit to decimal float.
16617 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
16618 point computations, error checking in decimal float operations ignores
16619 underflow, overflow and divide by zero exceptions.
16621 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
16622 to inspect @code{_Decimal128} values stored in floating point registers.
16623 See @ref{PowerPC,,PowerPC} for more details.
16629 @value{GDBN} can be used to debug programs written in D and compiled with
16630 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
16631 specific feature --- dynamic arrays.
16636 @cindex Go (programming language)
16637 @value{GDBN} can be used to debug programs written in Go and compiled with
16638 @file{gccgo} or @file{6g} compilers.
16640 Here is a summary of the Go-specific features and restrictions:
16643 @cindex current Go package
16644 @item The current Go package
16645 The name of the current package does not need to be specified when
16646 specifying global variables and functions.
16648 For example, given the program:
16652 var myglob = "Shall we?"
16658 When stopped inside @code{main} either of these work:
16661 (@value{GDBP}) p myglob
16662 (@value{GDBP}) p main.myglob
16665 @cindex builtin Go types
16666 @item Builtin Go types
16667 The @code{string} type is recognized by @value{GDBN} and is printed
16670 @cindex builtin Go functions
16671 @item Builtin Go functions
16672 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
16673 function and handles it internally.
16675 @cindex restrictions on Go expressions
16676 @item Restrictions on Go expressions
16677 All Go operators are supported except @code{&^}.
16678 The Go @code{_} ``blank identifier'' is not supported.
16679 Automatic dereferencing of pointers is not supported.
16683 @subsection Objective-C
16685 @cindex Objective-C
16686 This section provides information about some commands and command
16687 options that are useful for debugging Objective-C code. See also
16688 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
16689 few more commands specific to Objective-C support.
16692 * Method Names in Commands::
16693 * The Print Command with Objective-C::
16696 @node Method Names in Commands
16697 @subsubsection Method Names in Commands
16699 The following commands have been extended to accept Objective-C method
16700 names as line specifications:
16702 @kindex clear@r{, and Objective-C}
16703 @kindex break@r{, and Objective-C}
16704 @kindex info line@r{, and Objective-C}
16705 @kindex jump@r{, and Objective-C}
16706 @kindex list@r{, and Objective-C}
16710 @item @code{info line}
16715 A fully qualified Objective-C method name is specified as
16718 -[@var{Class} @var{methodName}]
16721 where the minus sign is used to indicate an instance method and a
16722 plus sign (not shown) is used to indicate a class method. The class
16723 name @var{Class} and method name @var{methodName} are enclosed in
16724 brackets, similar to the way messages are specified in Objective-C
16725 source code. For example, to set a breakpoint at the @code{create}
16726 instance method of class @code{Fruit} in the program currently being
16730 break -[Fruit create]
16733 To list ten program lines around the @code{initialize} class method,
16737 list +[NSText initialize]
16740 In the current version of @value{GDBN}, the plus or minus sign is
16741 required. In future versions of @value{GDBN}, the plus or minus
16742 sign will be optional, but you can use it to narrow the search. It
16743 is also possible to specify just a method name:
16749 You must specify the complete method name, including any colons. If
16750 your program's source files contain more than one @code{create} method,
16751 you'll be presented with a numbered list of classes that implement that
16752 method. Indicate your choice by number, or type @samp{0} to exit if
16755 As another example, to clear a breakpoint established at the
16756 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
16759 clear -[NSWindow makeKeyAndOrderFront:]
16762 @node The Print Command with Objective-C
16763 @subsubsection The Print Command With Objective-C
16764 @cindex Objective-C, print objects
16765 @kindex print-object
16766 @kindex po @r{(@code{print-object})}
16768 The print command has also been extended to accept methods. For example:
16771 print -[@var{object} hash]
16774 @cindex print an Objective-C object description
16775 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
16777 will tell @value{GDBN} to send the @code{hash} message to @var{object}
16778 and print the result. Also, an additional command has been added,
16779 @code{print-object} or @code{po} for short, which is meant to print
16780 the description of an object. However, this command may only work
16781 with certain Objective-C libraries that have a particular hook
16782 function, @code{_NSPrintForDebugger}, defined.
16785 @subsection OpenCL C
16788 This section provides information about @value{GDBN}s OpenCL C support.
16791 * OpenCL C Datatypes::
16792 * OpenCL C Expressions::
16793 * OpenCL C Operators::
16796 @node OpenCL C Datatypes
16797 @subsubsection OpenCL C Datatypes
16799 @cindex OpenCL C Datatypes
16800 @value{GDBN} supports the builtin scalar and vector datatypes specified
16801 by OpenCL 1.1. In addition the half- and double-precision floating point
16802 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
16803 extensions are also known to @value{GDBN}.
16805 @node OpenCL C Expressions
16806 @subsubsection OpenCL C Expressions
16808 @cindex OpenCL C Expressions
16809 @value{GDBN} supports accesses to vector components including the access as
16810 lvalue where possible. Since OpenCL C is based on C99 most C expressions
16811 supported by @value{GDBN} can be used as well.
16813 @node OpenCL C Operators
16814 @subsubsection OpenCL C Operators
16816 @cindex OpenCL C Operators
16817 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
16821 @subsection Fortran
16822 @cindex Fortran-specific support in @value{GDBN}
16824 @value{GDBN} can be used to debug programs written in Fortran, but it
16825 currently supports only the features of Fortran 77 language.
16827 @cindex trailing underscore, in Fortran symbols
16828 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
16829 among them) append an underscore to the names of variables and
16830 functions. When you debug programs compiled by those compilers, you
16831 will need to refer to variables and functions with a trailing
16835 * Fortran Operators:: Fortran operators and expressions
16836 * Fortran Defaults:: Default settings for Fortran
16837 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
16840 @node Fortran Operators
16841 @subsubsection Fortran Operators and Expressions
16843 @cindex Fortran operators and expressions
16845 Operators must be defined on values of specific types. For instance,
16846 @code{+} is defined on numbers, but not on characters or other non-
16847 arithmetic types. Operators are often defined on groups of types.
16851 The exponentiation operator. It raises the first operand to the power
16855 The range operator. Normally used in the form of array(low:high) to
16856 represent a section of array.
16859 The access component operator. Normally used to access elements in derived
16860 types. Also suitable for unions. As unions aren't part of regular Fortran,
16861 this can only happen when accessing a register that uses a gdbarch-defined
16864 The scope operator. Normally used to access variables in modules or
16865 to set breakpoints on subroutines nested in modules or in other
16866 subroutines (internal subroutines).
16869 @node Fortran Defaults
16870 @subsubsection Fortran Defaults
16872 @cindex Fortran Defaults
16874 Fortran symbols are usually case-insensitive, so @value{GDBN} by
16875 default uses case-insensitive matches for Fortran symbols. You can
16876 change that with the @samp{set case-insensitive} command, see
16877 @ref{Symbols}, for the details.
16879 @node Special Fortran Commands
16880 @subsubsection Special Fortran Commands
16882 @cindex Special Fortran commands
16884 @value{GDBN} has some commands to support Fortran-specific features,
16885 such as displaying common blocks.
16888 @cindex @code{COMMON} blocks, Fortran
16889 @kindex info common
16890 @item info common @r{[}@var{common-name}@r{]}
16891 This command prints the values contained in the Fortran @code{COMMON}
16892 block whose name is @var{common-name}. With no argument, the names of
16893 all @code{COMMON} blocks visible at the current program location are
16900 @cindex Pascal support in @value{GDBN}, limitations
16901 Debugging Pascal programs which use sets, subranges, file variables, or
16902 nested functions does not currently work. @value{GDBN} does not support
16903 entering expressions, printing values, or similar features using Pascal
16906 The Pascal-specific command @code{set print pascal_static-members}
16907 controls whether static members of Pascal objects are displayed.
16908 @xref{Print Settings, pascal_static-members}.
16913 @value{GDBN} supports the @url{https://www.rust-lang.org/, Rust
16914 Programming Language}. Type- and value-printing, and expression
16915 parsing, are reasonably complete. However, there are a few
16916 peculiarities and holes to be aware of.
16920 Linespecs (@pxref{Specify Location}) are never relative to the current
16921 crate. Instead, they act as if there were a global namespace of
16922 crates, somewhat similar to the way @code{extern crate} behaves.
16924 That is, if @value{GDBN} is stopped at a breakpoint in a function in
16925 crate @samp{A}, module @samp{B}, then @code{break B::f} will attempt
16926 to set a breakpoint in a function named @samp{f} in a crate named
16929 As a consequence of this approach, linespecs also cannot refer to
16930 items using @samp{self::} or @samp{super::}.
16933 Because @value{GDBN} implements Rust name-lookup semantics in
16934 expressions, it will sometimes prepend the current crate to a name.
16935 For example, if @value{GDBN} is stopped at a breakpoint in the crate
16936 @samp{K}, then @code{print ::x::y} will try to find the symbol
16939 However, since it is useful to be able to refer to other crates when
16940 debugging, @value{GDBN} provides the @code{extern} extension to
16941 circumvent this. To use the extension, just put @code{extern} before
16942 a path expression to refer to the otherwise unavailable ``global''
16945 In the above example, if you wanted to refer to the symbol @samp{y} in
16946 the crate @samp{x}, you would use @code{print extern x::y}.
16949 The Rust expression evaluator does not support ``statement-like''
16950 expressions such as @code{if} or @code{match}, or lambda expressions.
16953 Tuple expressions are not implemented.
16956 The Rust expression evaluator does not currently implement the
16957 @code{Drop} trait. Objects that may be created by the evaluator will
16958 never be destroyed.
16961 @value{GDBN} does not implement type inference for generics. In order
16962 to call generic functions or otherwise refer to generic items, you
16963 will have to specify the type parameters manually.
16966 @value{GDBN} currently uses the C@t{++} demangler for Rust. In most
16967 cases this does not cause any problems. However, in an expression
16968 context, completing a generic function name will give syntactically
16969 invalid results. This happens because Rust requires the @samp{::}
16970 operator between the function name and its generic arguments. For
16971 example, @value{GDBN} might provide a completion like
16972 @code{crate::f<u32>}, where the parser would require
16973 @code{crate::f::<u32>}.
16976 As of this writing, the Rust compiler (version 1.8) has a few holes in
16977 the debugging information it generates. These holes prevent certain
16978 features from being implemented by @value{GDBN}:
16982 Method calls cannot be made via traits.
16985 Operator overloading is not implemented.
16988 When debugging in a monomorphized function, you cannot use the generic
16992 The type @code{Self} is not available.
16995 @code{use} statements are not available, so some names may not be
16996 available in the crate.
17001 @subsection Modula-2
17003 @cindex Modula-2, @value{GDBN} support
17005 The extensions made to @value{GDBN} to support Modula-2 only support
17006 output from the @sc{gnu} Modula-2 compiler (which is currently being
17007 developed). Other Modula-2 compilers are not currently supported, and
17008 attempting to debug executables produced by them is most likely
17009 to give an error as @value{GDBN} reads in the executable's symbol
17012 @cindex expressions in Modula-2
17014 * M2 Operators:: Built-in operators
17015 * Built-In Func/Proc:: Built-in functions and procedures
17016 * M2 Constants:: Modula-2 constants
17017 * M2 Types:: Modula-2 types
17018 * M2 Defaults:: Default settings for Modula-2
17019 * Deviations:: Deviations from standard Modula-2
17020 * M2 Checks:: Modula-2 type and range checks
17021 * M2 Scope:: The scope operators @code{::} and @code{.}
17022 * GDB/M2:: @value{GDBN} and Modula-2
17026 @subsubsection Operators
17027 @cindex Modula-2 operators
17029 Operators must be defined on values of specific types. For instance,
17030 @code{+} is defined on numbers, but not on structures. Operators are
17031 often defined on groups of types. For the purposes of Modula-2, the
17032 following definitions hold:
17037 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
17041 @emph{Character types} consist of @code{CHAR} and its subranges.
17044 @emph{Floating-point types} consist of @code{REAL}.
17047 @emph{Pointer types} consist of anything declared as @code{POINTER TO
17051 @emph{Scalar types} consist of all of the above.
17054 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
17057 @emph{Boolean types} consist of @code{BOOLEAN}.
17061 The following operators are supported, and appear in order of
17062 increasing precedence:
17066 Function argument or array index separator.
17069 Assignment. The value of @var{var} @code{:=} @var{value} is
17073 Less than, greater than on integral, floating-point, or enumerated
17077 Less than or equal to, greater than or equal to
17078 on integral, floating-point and enumerated types, or set inclusion on
17079 set types. Same precedence as @code{<}.
17081 @item =@r{, }<>@r{, }#
17082 Equality and two ways of expressing inequality, valid on scalar types.
17083 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
17084 available for inequality, since @code{#} conflicts with the script
17088 Set membership. Defined on set types and the types of their members.
17089 Same precedence as @code{<}.
17092 Boolean disjunction. Defined on boolean types.
17095 Boolean conjunction. Defined on boolean types.
17098 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
17101 Addition and subtraction on integral and floating-point types, or union
17102 and difference on set types.
17105 Multiplication on integral and floating-point types, or set intersection
17109 Division on floating-point types, or symmetric set difference on set
17110 types. Same precedence as @code{*}.
17113 Integer division and remainder. Defined on integral types. Same
17114 precedence as @code{*}.
17117 Negative. Defined on @code{INTEGER} and @code{REAL} data.
17120 Pointer dereferencing. Defined on pointer types.
17123 Boolean negation. Defined on boolean types. Same precedence as
17127 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
17128 precedence as @code{^}.
17131 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
17134 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
17138 @value{GDBN} and Modula-2 scope operators.
17142 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
17143 treats the use of the operator @code{IN}, or the use of operators
17144 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
17145 @code{<=}, and @code{>=} on sets as an error.
17149 @node Built-In Func/Proc
17150 @subsubsection Built-in Functions and Procedures
17151 @cindex Modula-2 built-ins
17153 Modula-2 also makes available several built-in procedures and functions.
17154 In describing these, the following metavariables are used:
17159 represents an @code{ARRAY} variable.
17162 represents a @code{CHAR} constant or variable.
17165 represents a variable or constant of integral type.
17168 represents an identifier that belongs to a set. Generally used in the
17169 same function with the metavariable @var{s}. The type of @var{s} should
17170 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
17173 represents a variable or constant of integral or floating-point type.
17176 represents a variable or constant of floating-point type.
17182 represents a variable.
17185 represents a variable or constant of one of many types. See the
17186 explanation of the function for details.
17189 All Modula-2 built-in procedures also return a result, described below.
17193 Returns the absolute value of @var{n}.
17196 If @var{c} is a lower case letter, it returns its upper case
17197 equivalent, otherwise it returns its argument.
17200 Returns the character whose ordinal value is @var{i}.
17203 Decrements the value in the variable @var{v} by one. Returns the new value.
17205 @item DEC(@var{v},@var{i})
17206 Decrements the value in the variable @var{v} by @var{i}. Returns the
17209 @item EXCL(@var{m},@var{s})
17210 Removes the element @var{m} from the set @var{s}. Returns the new
17213 @item FLOAT(@var{i})
17214 Returns the floating point equivalent of the integer @var{i}.
17216 @item HIGH(@var{a})
17217 Returns the index of the last member of @var{a}.
17220 Increments the value in the variable @var{v} by one. Returns the new value.
17222 @item INC(@var{v},@var{i})
17223 Increments the value in the variable @var{v} by @var{i}. Returns the
17226 @item INCL(@var{m},@var{s})
17227 Adds the element @var{m} to the set @var{s} if it is not already
17228 there. Returns the new set.
17231 Returns the maximum value of the type @var{t}.
17234 Returns the minimum value of the type @var{t}.
17237 Returns boolean TRUE if @var{i} is an odd number.
17240 Returns the ordinal value of its argument. For example, the ordinal
17241 value of a character is its @sc{ascii} value (on machines supporting
17242 the @sc{ascii} character set). The argument @var{x} must be of an
17243 ordered type, which include integral, character and enumerated types.
17245 @item SIZE(@var{x})
17246 Returns the size of its argument. The argument @var{x} can be a
17247 variable or a type.
17249 @item TRUNC(@var{r})
17250 Returns the integral part of @var{r}.
17252 @item TSIZE(@var{x})
17253 Returns the size of its argument. The argument @var{x} can be a
17254 variable or a type.
17256 @item VAL(@var{t},@var{i})
17257 Returns the member of the type @var{t} whose ordinal value is @var{i}.
17261 @emph{Warning:} Sets and their operations are not yet supported, so
17262 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
17266 @cindex Modula-2 constants
17268 @subsubsection Constants
17270 @value{GDBN} allows you to express the constants of Modula-2 in the following
17276 Integer constants are simply a sequence of digits. When used in an
17277 expression, a constant is interpreted to be type-compatible with the
17278 rest of the expression. Hexadecimal integers are specified by a
17279 trailing @samp{H}, and octal integers by a trailing @samp{B}.
17282 Floating point constants appear as a sequence of digits, followed by a
17283 decimal point and another sequence of digits. An optional exponent can
17284 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
17285 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
17286 digits of the floating point constant must be valid decimal (base 10)
17290 Character constants consist of a single character enclosed by a pair of
17291 like quotes, either single (@code{'}) or double (@code{"}). They may
17292 also be expressed by their ordinal value (their @sc{ascii} value, usually)
17293 followed by a @samp{C}.
17296 String constants consist of a sequence of characters enclosed by a
17297 pair of like quotes, either single (@code{'}) or double (@code{"}).
17298 Escape sequences in the style of C are also allowed. @xref{C
17299 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
17303 Enumerated constants consist of an enumerated identifier.
17306 Boolean constants consist of the identifiers @code{TRUE} and
17310 Pointer constants consist of integral values only.
17313 Set constants are not yet supported.
17317 @subsubsection Modula-2 Types
17318 @cindex Modula-2 types
17320 Currently @value{GDBN} can print the following data types in Modula-2
17321 syntax: array types, record types, set types, pointer types, procedure
17322 types, enumerated types, subrange types and base types. You can also
17323 print the contents of variables declared using these type.
17324 This section gives a number of simple source code examples together with
17325 sample @value{GDBN} sessions.
17327 The first example contains the following section of code:
17336 and you can request @value{GDBN} to interrogate the type and value of
17337 @code{r} and @code{s}.
17340 (@value{GDBP}) print s
17342 (@value{GDBP}) ptype s
17344 (@value{GDBP}) print r
17346 (@value{GDBP}) ptype r
17351 Likewise if your source code declares @code{s} as:
17355 s: SET ['A'..'Z'] ;
17359 then you may query the type of @code{s} by:
17362 (@value{GDBP}) ptype s
17363 type = SET ['A'..'Z']
17367 Note that at present you cannot interactively manipulate set
17368 expressions using the debugger.
17370 The following example shows how you might declare an array in Modula-2
17371 and how you can interact with @value{GDBN} to print its type and contents:
17375 s: ARRAY [-10..10] OF CHAR ;
17379 (@value{GDBP}) ptype s
17380 ARRAY [-10..10] OF CHAR
17383 Note that the array handling is not yet complete and although the type
17384 is printed correctly, expression handling still assumes that all
17385 arrays have a lower bound of zero and not @code{-10} as in the example
17388 Here are some more type related Modula-2 examples:
17392 colour = (blue, red, yellow, green) ;
17393 t = [blue..yellow] ;
17401 The @value{GDBN} interaction shows how you can query the data type
17402 and value of a variable.
17405 (@value{GDBP}) print s
17407 (@value{GDBP}) ptype t
17408 type = [blue..yellow]
17412 In this example a Modula-2 array is declared and its contents
17413 displayed. Observe that the contents are written in the same way as
17414 their @code{C} counterparts.
17418 s: ARRAY [1..5] OF CARDINAL ;
17424 (@value{GDBP}) print s
17425 $1 = @{1, 0, 0, 0, 0@}
17426 (@value{GDBP}) ptype s
17427 type = ARRAY [1..5] OF CARDINAL
17430 The Modula-2 language interface to @value{GDBN} also understands
17431 pointer types as shown in this example:
17435 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
17442 and you can request that @value{GDBN} describes the type of @code{s}.
17445 (@value{GDBP}) ptype s
17446 type = POINTER TO ARRAY [1..5] OF CARDINAL
17449 @value{GDBN} handles compound types as we can see in this example.
17450 Here we combine array types, record types, pointer types and subrange
17461 myarray = ARRAY myrange OF CARDINAL ;
17462 myrange = [-2..2] ;
17464 s: POINTER TO ARRAY myrange OF foo ;
17468 and you can ask @value{GDBN} to describe the type of @code{s} as shown
17472 (@value{GDBP}) ptype s
17473 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
17476 f3 : ARRAY [-2..2] OF CARDINAL;
17481 @subsubsection Modula-2 Defaults
17482 @cindex Modula-2 defaults
17484 If type and range checking are set automatically by @value{GDBN}, they
17485 both default to @code{on} whenever the working language changes to
17486 Modula-2. This happens regardless of whether you or @value{GDBN}
17487 selected the working language.
17489 If you allow @value{GDBN} to set the language automatically, then entering
17490 code compiled from a file whose name ends with @file{.mod} sets the
17491 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
17492 Infer the Source Language}, for further details.
17495 @subsubsection Deviations from Standard Modula-2
17496 @cindex Modula-2, deviations from
17498 A few changes have been made to make Modula-2 programs easier to debug.
17499 This is done primarily via loosening its type strictness:
17503 Unlike in standard Modula-2, pointer constants can be formed by
17504 integers. This allows you to modify pointer variables during
17505 debugging. (In standard Modula-2, the actual address contained in a
17506 pointer variable is hidden from you; it can only be modified
17507 through direct assignment to another pointer variable or expression that
17508 returned a pointer.)
17511 C escape sequences can be used in strings and characters to represent
17512 non-printable characters. @value{GDBN} prints out strings with these
17513 escape sequences embedded. Single non-printable characters are
17514 printed using the @samp{CHR(@var{nnn})} format.
17517 The assignment operator (@code{:=}) returns the value of its right-hand
17521 All built-in procedures both modify @emph{and} return their argument.
17525 @subsubsection Modula-2 Type and Range Checks
17526 @cindex Modula-2 checks
17529 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
17532 @c FIXME remove warning when type/range checks added
17534 @value{GDBN} considers two Modula-2 variables type equivalent if:
17538 They are of types that have been declared equivalent via a @code{TYPE
17539 @var{t1} = @var{t2}} statement
17542 They have been declared on the same line. (Note: This is true of the
17543 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
17546 As long as type checking is enabled, any attempt to combine variables
17547 whose types are not equivalent is an error.
17549 Range checking is done on all mathematical operations, assignment, array
17550 index bounds, and all built-in functions and procedures.
17553 @subsubsection The Scope Operators @code{::} and @code{.}
17555 @cindex @code{.}, Modula-2 scope operator
17556 @cindex colon, doubled as scope operator
17558 @vindex colon-colon@r{, in Modula-2}
17559 @c Info cannot handle :: but TeX can.
17562 @vindex ::@r{, in Modula-2}
17565 There are a few subtle differences between the Modula-2 scope operator
17566 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
17571 @var{module} . @var{id}
17572 @var{scope} :: @var{id}
17576 where @var{scope} is the name of a module or a procedure,
17577 @var{module} the name of a module, and @var{id} is any declared
17578 identifier within your program, except another module.
17580 Using the @code{::} operator makes @value{GDBN} search the scope
17581 specified by @var{scope} for the identifier @var{id}. If it is not
17582 found in the specified scope, then @value{GDBN} searches all scopes
17583 enclosing the one specified by @var{scope}.
17585 Using the @code{.} operator makes @value{GDBN} search the current scope for
17586 the identifier specified by @var{id} that was imported from the
17587 definition module specified by @var{module}. With this operator, it is
17588 an error if the identifier @var{id} was not imported from definition
17589 module @var{module}, or if @var{id} is not an identifier in
17593 @subsubsection @value{GDBN} and Modula-2
17595 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
17596 Five subcommands of @code{set print} and @code{show print} apply
17597 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
17598 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
17599 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
17600 analogue in Modula-2.
17602 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
17603 with any language, is not useful with Modula-2. Its
17604 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
17605 created in Modula-2 as they can in C or C@t{++}. However, because an
17606 address can be specified by an integral constant, the construct
17607 @samp{@{@var{type}@}@var{adrexp}} is still useful.
17609 @cindex @code{#} in Modula-2
17610 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
17611 interpreted as the beginning of a comment. Use @code{<>} instead.
17617 The extensions made to @value{GDBN} for Ada only support
17618 output from the @sc{gnu} Ada (GNAT) compiler.
17619 Other Ada compilers are not currently supported, and
17620 attempting to debug executables produced by them is most likely
17624 @cindex expressions in Ada
17626 * Ada Mode Intro:: General remarks on the Ada syntax
17627 and semantics supported by Ada mode
17629 * Omissions from Ada:: Restrictions on the Ada expression syntax.
17630 * Additions to Ada:: Extensions of the Ada expression syntax.
17631 * Overloading support for Ada:: Support for expressions involving overloaded
17633 * Stopping Before Main Program:: Debugging the program during elaboration.
17634 * Ada Exceptions:: Ada Exceptions
17635 * Ada Tasks:: Listing and setting breakpoints in tasks.
17636 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
17637 * Ravenscar Profile:: Tasking Support when using the Ravenscar
17639 * Ada Settings:: New settable GDB parameters for Ada.
17640 * Ada Glitches:: Known peculiarities of Ada mode.
17643 @node Ada Mode Intro
17644 @subsubsection Introduction
17645 @cindex Ada mode, general
17647 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
17648 syntax, with some extensions.
17649 The philosophy behind the design of this subset is
17653 That @value{GDBN} should provide basic literals and access to operations for
17654 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
17655 leaving more sophisticated computations to subprograms written into the
17656 program (which therefore may be called from @value{GDBN}).
17659 That type safety and strict adherence to Ada language restrictions
17660 are not particularly important to the @value{GDBN} user.
17663 That brevity is important to the @value{GDBN} user.
17666 Thus, for brevity, the debugger acts as if all names declared in
17667 user-written packages are directly visible, even if they are not visible
17668 according to Ada rules, thus making it unnecessary to fully qualify most
17669 names with their packages, regardless of context. Where this causes
17670 ambiguity, @value{GDBN} asks the user's intent.
17672 The debugger will start in Ada mode if it detects an Ada main program.
17673 As for other languages, it will enter Ada mode when stopped in a program that
17674 was translated from an Ada source file.
17676 While in Ada mode, you may use `@t{--}' for comments. This is useful
17677 mostly for documenting command files. The standard @value{GDBN} comment
17678 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
17679 middle (to allow based literals).
17681 @node Omissions from Ada
17682 @subsubsection Omissions from Ada
17683 @cindex Ada, omissions from
17685 Here are the notable omissions from the subset:
17689 Only a subset of the attributes are supported:
17693 @t{'First}, @t{'Last}, and @t{'Length}
17694 on array objects (not on types and subtypes).
17697 @t{'Min} and @t{'Max}.
17700 @t{'Pos} and @t{'Val}.
17706 @t{'Range} on array objects (not subtypes), but only as the right
17707 operand of the membership (@code{in}) operator.
17710 @t{'Access}, @t{'Unchecked_Access}, and
17711 @t{'Unrestricted_Access} (a GNAT extension).
17719 @code{Characters.Latin_1} are not available and
17720 concatenation is not implemented. Thus, escape characters in strings are
17721 not currently available.
17724 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
17725 equality of representations. They will generally work correctly
17726 for strings and arrays whose elements have integer or enumeration types.
17727 They may not work correctly for arrays whose element
17728 types have user-defined equality, for arrays of real values
17729 (in particular, IEEE-conformant floating point, because of negative
17730 zeroes and NaNs), and for arrays whose elements contain unused bits with
17731 indeterminate values.
17734 The other component-by-component array operations (@code{and}, @code{or},
17735 @code{xor}, @code{not}, and relational tests other than equality)
17736 are not implemented.
17739 @cindex array aggregates (Ada)
17740 @cindex record aggregates (Ada)
17741 @cindex aggregates (Ada)
17742 There is limited support for array and record aggregates. They are
17743 permitted only on the right sides of assignments, as in these examples:
17746 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
17747 (@value{GDBP}) set An_Array := (1, others => 0)
17748 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
17749 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
17750 (@value{GDBP}) set A_Record := (1, "Peter", True);
17751 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
17755 discriminant's value by assigning an aggregate has an
17756 undefined effect if that discriminant is used within the record.
17757 However, you can first modify discriminants by directly assigning to
17758 them (which normally would not be allowed in Ada), and then performing an
17759 aggregate assignment. For example, given a variable @code{A_Rec}
17760 declared to have a type such as:
17763 type Rec (Len : Small_Integer := 0) is record
17765 Vals : IntArray (1 .. Len);
17769 you can assign a value with a different size of @code{Vals} with two
17773 (@value{GDBP}) set A_Rec.Len := 4
17774 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
17777 As this example also illustrates, @value{GDBN} is very loose about the usual
17778 rules concerning aggregates. You may leave out some of the
17779 components of an array or record aggregate (such as the @code{Len}
17780 component in the assignment to @code{A_Rec} above); they will retain their
17781 original values upon assignment. You may freely use dynamic values as
17782 indices in component associations. You may even use overlapping or
17783 redundant component associations, although which component values are
17784 assigned in such cases is not defined.
17787 Calls to dispatching subprograms are not implemented.
17790 The overloading algorithm is much more limited (i.e., less selective)
17791 than that of real Ada. It makes only limited use of the context in
17792 which a subexpression appears to resolve its meaning, and it is much
17793 looser in its rules for allowing type matches. As a result, some
17794 function calls will be ambiguous, and the user will be asked to choose
17795 the proper resolution.
17798 The @code{new} operator is not implemented.
17801 Entry calls are not implemented.
17804 Aside from printing, arithmetic operations on the native VAX floating-point
17805 formats are not supported.
17808 It is not possible to slice a packed array.
17811 The names @code{True} and @code{False}, when not part of a qualified name,
17812 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
17814 Should your program
17815 redefine these names in a package or procedure (at best a dubious practice),
17816 you will have to use fully qualified names to access their new definitions.
17819 @node Additions to Ada
17820 @subsubsection Additions to Ada
17821 @cindex Ada, deviations from
17823 As it does for other languages, @value{GDBN} makes certain generic
17824 extensions to Ada (@pxref{Expressions}):
17828 If the expression @var{E} is a variable residing in memory (typically
17829 a local variable or array element) and @var{N} is a positive integer,
17830 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
17831 @var{N}-1 adjacent variables following it in memory as an array. In
17832 Ada, this operator is generally not necessary, since its prime use is
17833 in displaying parts of an array, and slicing will usually do this in
17834 Ada. However, there are occasional uses when debugging programs in
17835 which certain debugging information has been optimized away.
17838 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
17839 appears in function or file @var{B}.'' When @var{B} is a file name,
17840 you must typically surround it in single quotes.
17843 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
17844 @var{type} that appears at address @var{addr}.''
17847 A name starting with @samp{$} is a convenience variable
17848 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
17851 In addition, @value{GDBN} provides a few other shortcuts and outright
17852 additions specific to Ada:
17856 The assignment statement is allowed as an expression, returning
17857 its right-hand operand as its value. Thus, you may enter
17860 (@value{GDBP}) set x := y + 3
17861 (@value{GDBP}) print A(tmp := y + 1)
17865 The semicolon is allowed as an ``operator,'' returning as its value
17866 the value of its right-hand operand.
17867 This allows, for example,
17868 complex conditional breaks:
17871 (@value{GDBP}) break f
17872 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
17876 Rather than use catenation and symbolic character names to introduce special
17877 characters into strings, one may instead use a special bracket notation,
17878 which is also used to print strings. A sequence of characters of the form
17879 @samp{["@var{XX}"]} within a string or character literal denotes the
17880 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
17881 sequence of characters @samp{["""]} also denotes a single quotation mark
17882 in strings. For example,
17884 "One line.["0a"]Next line.["0a"]"
17887 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
17891 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
17892 @t{'Max} is optional (and is ignored in any case). For example, it is valid
17896 (@value{GDBP}) print 'max(x, y)
17900 When printing arrays, @value{GDBN} uses positional notation when the
17901 array has a lower bound of 1, and uses a modified named notation otherwise.
17902 For example, a one-dimensional array of three integers with a lower bound
17903 of 3 might print as
17910 That is, in contrast to valid Ada, only the first component has a @code{=>}
17914 You may abbreviate attributes in expressions with any unique,
17915 multi-character subsequence of
17916 their names (an exact match gets preference).
17917 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
17918 in place of @t{a'length}.
17921 @cindex quoting Ada internal identifiers
17922 Since Ada is case-insensitive, the debugger normally maps identifiers you type
17923 to lower case. The GNAT compiler uses upper-case characters for
17924 some of its internal identifiers, which are normally of no interest to users.
17925 For the rare occasions when you actually have to look at them,
17926 enclose them in angle brackets to avoid the lower-case mapping.
17929 (@value{GDBP}) print <JMPBUF_SAVE>[0]
17933 Printing an object of class-wide type or dereferencing an
17934 access-to-class-wide value will display all the components of the object's
17935 specific type (as indicated by its run-time tag). Likewise, component
17936 selection on such a value will operate on the specific type of the
17941 @node Overloading support for Ada
17942 @subsubsection Overloading support for Ada
17943 @cindex overloading, Ada
17945 The debugger supports limited overloading. Given a subprogram call in which
17946 the function symbol has multiple definitions, it will use the number of
17947 actual parameters and some information about their types to attempt to narrow
17948 the set of definitions. It also makes very limited use of context, preferring
17949 procedures to functions in the context of the @code{call} command, and
17950 functions to procedures elsewhere.
17952 If, after narrowing, the set of matching definitions still contains more than
17953 one definition, @value{GDBN} will display a menu to query which one it should
17957 (@value{GDBP}) print f(1)
17958 Multiple matches for f
17960 [1] foo.f (integer) return boolean at foo.adb:23
17961 [2] foo.f (foo.new_integer) return boolean at foo.adb:28
17965 In this case, just select one menu entry either to cancel expression evaluation
17966 (type @kbd{0} and press @key{RET}) or to continue evaluation with a specific
17967 instance (type the corresponding number and press @key{RET}).
17969 Here are a couple of commands to customize @value{GDBN}'s behavior in this
17974 @kindex set ada print-signatures
17975 @item set ada print-signatures
17976 Control whether parameter types and return types are displayed in overloads
17977 selection menus. It is @code{on} by default.
17978 @xref{Overloading support for Ada}.
17980 @kindex show ada print-signatures
17981 @item show ada print-signatures
17982 Show the current setting for displaying parameter types and return types in
17983 overloads selection menu.
17984 @xref{Overloading support for Ada}.
17988 @node Stopping Before Main Program
17989 @subsubsection Stopping at the Very Beginning
17991 @cindex breakpointing Ada elaboration code
17992 It is sometimes necessary to debug the program during elaboration, and
17993 before reaching the main procedure.
17994 As defined in the Ada Reference
17995 Manual, the elaboration code is invoked from a procedure called
17996 @code{adainit}. To run your program up to the beginning of
17997 elaboration, simply use the following two commands:
17998 @code{tbreak adainit} and @code{run}.
18000 @node Ada Exceptions
18001 @subsubsection Ada Exceptions
18003 A command is provided to list all Ada exceptions:
18006 @kindex info exceptions
18007 @item info exceptions
18008 @itemx info exceptions @var{regexp}
18009 The @code{info exceptions} command allows you to list all Ada exceptions
18010 defined within the program being debugged, as well as their addresses.
18011 With a regular expression, @var{regexp}, as argument, only those exceptions
18012 whose names match @var{regexp} are listed.
18015 Below is a small example, showing how the command can be used, first
18016 without argument, and next with a regular expression passed as an
18020 (@value{GDBP}) info exceptions
18021 All defined Ada exceptions:
18022 constraint_error: 0x613da0
18023 program_error: 0x613d20
18024 storage_error: 0x613ce0
18025 tasking_error: 0x613ca0
18026 const.aint_global_e: 0x613b00
18027 (@value{GDBP}) info exceptions const.aint
18028 All Ada exceptions matching regular expression "const.aint":
18029 constraint_error: 0x613da0
18030 const.aint_global_e: 0x613b00
18033 It is also possible to ask @value{GDBN} to stop your program's execution
18034 when an exception is raised. For more details, see @ref{Set Catchpoints}.
18037 @subsubsection Extensions for Ada Tasks
18038 @cindex Ada, tasking
18040 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
18041 @value{GDBN} provides the following task-related commands:
18046 This command shows a list of current Ada tasks, as in the following example:
18053 (@value{GDBP}) info tasks
18054 ID TID P-ID Pri State Name
18055 1 8088000 0 15 Child Activation Wait main_task
18056 2 80a4000 1 15 Accept Statement b
18057 3 809a800 1 15 Child Activation Wait a
18058 * 4 80ae800 3 15 Runnable c
18063 In this listing, the asterisk before the last task indicates it to be the
18064 task currently being inspected.
18068 Represents @value{GDBN}'s internal task number.
18074 The parent's task ID (@value{GDBN}'s internal task number).
18077 The base priority of the task.
18080 Current state of the task.
18084 The task has been created but has not been activated. It cannot be
18088 The task is not blocked for any reason known to Ada. (It may be waiting
18089 for a mutex, though.) It is conceptually "executing" in normal mode.
18092 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
18093 that were waiting on terminate alternatives have been awakened and have
18094 terminated themselves.
18096 @item Child Activation Wait
18097 The task is waiting for created tasks to complete activation.
18099 @item Accept Statement
18100 The task is waiting on an accept or selective wait statement.
18102 @item Waiting on entry call
18103 The task is waiting on an entry call.
18105 @item Async Select Wait
18106 The task is waiting to start the abortable part of an asynchronous
18110 The task is waiting on a select statement with only a delay
18113 @item Child Termination Wait
18114 The task is sleeping having completed a master within itself, and is
18115 waiting for the tasks dependent on that master to become terminated or
18116 waiting on a terminate Phase.
18118 @item Wait Child in Term Alt
18119 The task is sleeping waiting for tasks on terminate alternatives to
18120 finish terminating.
18122 @item Accepting RV with @var{taskno}
18123 The task is accepting a rendez-vous with the task @var{taskno}.
18127 Name of the task in the program.
18131 @kindex info task @var{taskno}
18132 @item info task @var{taskno}
18133 This command shows detailed informations on the specified task, as in
18134 the following example:
18139 (@value{GDBP}) info tasks
18140 ID TID P-ID Pri State Name
18141 1 8077880 0 15 Child Activation Wait main_task
18142 * 2 807c468 1 15 Runnable task_1
18143 (@value{GDBP}) info task 2
18144 Ada Task: 0x807c468
18148 Parent: 1 ("main_task")
18154 @kindex task@r{ (Ada)}
18155 @cindex current Ada task ID
18156 This command prints the ID and name of the current task.
18162 (@value{GDBP}) info tasks
18163 ID TID P-ID Pri State Name
18164 1 8077870 0 15 Child Activation Wait main_task
18165 * 2 807c458 1 15 Runnable some_task
18166 (@value{GDBP}) task
18167 [Current task is 2 "some_task"]
18170 @item task @var{taskno}
18171 @cindex Ada task switching
18172 This command is like the @code{thread @var{thread-id}}
18173 command (@pxref{Threads}). It switches the context of debugging
18174 from the current task to the given task.
18180 (@value{GDBP}) info tasks
18181 ID TID P-ID Pri State Name
18182 1 8077870 0 15 Child Activation Wait main_task
18183 * 2 807c458 1 15 Runnable some_task
18184 (@value{GDBP}) task 1
18185 [Switching to task 1 "main_task"]
18186 #0 0x8067726 in pthread_cond_wait ()
18188 #0 0x8067726 in pthread_cond_wait ()
18189 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
18190 #2 0x805cb63 in system.task_primitives.operations.sleep ()
18191 #3 0x806153e in system.tasking.stages.activate_tasks ()
18192 #4 0x804aacc in un () at un.adb:5
18195 @item break @var{location} task @var{taskno}
18196 @itemx break @var{location} task @var{taskno} if @dots{}
18197 @cindex breakpoints and tasks, in Ada
18198 @cindex task breakpoints, in Ada
18199 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
18200 These commands are like the @code{break @dots{} thread @dots{}}
18201 command (@pxref{Thread Stops}). The
18202 @var{location} argument specifies source lines, as described
18203 in @ref{Specify Location}.
18205 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
18206 to specify that you only want @value{GDBN} to stop the program when a
18207 particular Ada task reaches this breakpoint. The @var{taskno} is one of the
18208 numeric task identifiers assigned by @value{GDBN}, shown in the first
18209 column of the @samp{info tasks} display.
18211 If you do not specify @samp{task @var{taskno}} when you set a
18212 breakpoint, the breakpoint applies to @emph{all} tasks of your
18215 You can use the @code{task} qualifier on conditional breakpoints as
18216 well; in this case, place @samp{task @var{taskno}} before the
18217 breakpoint condition (before the @code{if}).
18225 (@value{GDBP}) info tasks
18226 ID TID P-ID Pri State Name
18227 1 140022020 0 15 Child Activation Wait main_task
18228 2 140045060 1 15 Accept/Select Wait t2
18229 3 140044840 1 15 Runnable t1
18230 * 4 140056040 1 15 Runnable t3
18231 (@value{GDBP}) b 15 task 2
18232 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
18233 (@value{GDBP}) cont
18238 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
18240 (@value{GDBP}) info tasks
18241 ID TID P-ID Pri State Name
18242 1 140022020 0 15 Child Activation Wait main_task
18243 * 2 140045060 1 15 Runnable t2
18244 3 140044840 1 15 Runnable t1
18245 4 140056040 1 15 Delay Sleep t3
18249 @node Ada Tasks and Core Files
18250 @subsubsection Tasking Support when Debugging Core Files
18251 @cindex Ada tasking and core file debugging
18253 When inspecting a core file, as opposed to debugging a live program,
18254 tasking support may be limited or even unavailable, depending on
18255 the platform being used.
18256 For instance, on x86-linux, the list of tasks is available, but task
18257 switching is not supported.
18259 On certain platforms, the debugger needs to perform some
18260 memory writes in order to provide Ada tasking support. When inspecting
18261 a core file, this means that the core file must be opened with read-write
18262 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
18263 Under these circumstances, you should make a backup copy of the core
18264 file before inspecting it with @value{GDBN}.
18266 @node Ravenscar Profile
18267 @subsubsection Tasking Support when using the Ravenscar Profile
18268 @cindex Ravenscar Profile
18270 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
18271 specifically designed for systems with safety-critical real-time
18275 @kindex set ravenscar task-switching on
18276 @cindex task switching with program using Ravenscar Profile
18277 @item set ravenscar task-switching on
18278 Allows task switching when debugging a program that uses the Ravenscar
18279 Profile. This is the default.
18281 @kindex set ravenscar task-switching off
18282 @item set ravenscar task-switching off
18283 Turn off task switching when debugging a program that uses the Ravenscar
18284 Profile. This is mostly intended to disable the code that adds support
18285 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
18286 the Ravenscar runtime is preventing @value{GDBN} from working properly.
18287 To be effective, this command should be run before the program is started.
18289 @kindex show ravenscar task-switching
18290 @item show ravenscar task-switching
18291 Show whether it is possible to switch from task to task in a program
18292 using the Ravenscar Profile.
18297 @subsubsection Ada Settings
18298 @cindex Ada settings
18301 @kindex set varsize-limit
18302 @item set varsize-limit @var{size}
18303 Prevent @value{GDBN} from attempting to evaluate objects whose size
18304 is above the given limit (@var{size}) when those sizes are computed
18305 from run-time quantities. This is typically the case when the object
18306 has a variable size, such as an array whose bounds are not known at
18307 compile time for example. Setting @var{size} to @code{unlimited}
18308 removes the size limitation. By default, the limit is about 65KB.
18310 The purpose of having such a limit is to prevent @value{GDBN} from
18311 trying to grab enormous chunks of virtual memory when asked to evaluate
18312 a quantity whose bounds have been corrupted or have not yet been fully
18313 initialized. The limit applies to the results of some subexpressions
18314 as well as to complete expressions. For example, an expression denoting
18315 a simple integer component, such as @code{x.y.z}, may fail if the size of
18316 @code{x.y} is variable and exceeds @code{size}. On the other hand,
18317 @value{GDBN} is sometimes clever; the expression @code{A(i)}, where
18318 @code{A} is an array variable with non-constant size, will generally
18319 succeed regardless of the bounds on @code{A}, as long as the component
18320 size is less than @var{size}.
18322 @kindex show varsize-limit
18323 @item show varsize-limit
18324 Show the limit on types whose size is determined by run-time quantities.
18328 @subsubsection Known Peculiarities of Ada Mode
18329 @cindex Ada, problems
18331 Besides the omissions listed previously (@pxref{Omissions from Ada}),
18332 we know of several problems with and limitations of Ada mode in
18334 some of which will be fixed with planned future releases of the debugger
18335 and the GNU Ada compiler.
18339 Static constants that the compiler chooses not to materialize as objects in
18340 storage are invisible to the debugger.
18343 Named parameter associations in function argument lists are ignored (the
18344 argument lists are treated as positional).
18347 Many useful library packages are currently invisible to the debugger.
18350 Fixed-point arithmetic, conversions, input, and output is carried out using
18351 floating-point arithmetic, and may give results that only approximate those on
18355 The GNAT compiler never generates the prefix @code{Standard} for any of
18356 the standard symbols defined by the Ada language. @value{GDBN} knows about
18357 this: it will strip the prefix from names when you use it, and will never
18358 look for a name you have so qualified among local symbols, nor match against
18359 symbols in other packages or subprograms. If you have
18360 defined entities anywhere in your program other than parameters and
18361 local variables whose simple names match names in @code{Standard},
18362 GNAT's lack of qualification here can cause confusion. When this happens,
18363 you can usually resolve the confusion
18364 by qualifying the problematic names with package
18365 @code{Standard} explicitly.
18368 Older versions of the compiler sometimes generate erroneous debugging
18369 information, resulting in the debugger incorrectly printing the value
18370 of affected entities. In some cases, the debugger is able to work
18371 around an issue automatically. In other cases, the debugger is able
18372 to work around the issue, but the work-around has to be specifically
18375 @kindex set ada trust-PAD-over-XVS
18376 @kindex show ada trust-PAD-over-XVS
18379 @item set ada trust-PAD-over-XVS on
18380 Configure GDB to strictly follow the GNAT encoding when computing the
18381 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
18382 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
18383 a complete description of the encoding used by the GNAT compiler).
18384 This is the default.
18386 @item set ada trust-PAD-over-XVS off
18387 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
18388 sometimes prints the wrong value for certain entities, changing @code{ada
18389 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
18390 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
18391 @code{off}, but this incurs a slight performance penalty, so it is
18392 recommended to leave this setting to @code{on} unless necessary.
18396 @cindex GNAT descriptive types
18397 @cindex GNAT encoding
18398 Internally, the debugger also relies on the compiler following a number
18399 of conventions known as the @samp{GNAT Encoding}, all documented in
18400 @file{gcc/ada/exp_dbug.ads} in the GCC sources. This encoding describes
18401 how the debugging information should be generated for certain types.
18402 In particular, this convention makes use of @dfn{descriptive types},
18403 which are artificial types generated purely to help the debugger.
18405 These encodings were defined at a time when the debugging information
18406 format used was not powerful enough to describe some of the more complex
18407 types available in Ada. Since DWARF allows us to express nearly all
18408 Ada features, the long-term goal is to slowly replace these descriptive
18409 types by their pure DWARF equivalent. To facilitate that transition,
18410 a new maintenance option is available to force the debugger to ignore
18411 those descriptive types. It allows the user to quickly evaluate how
18412 well @value{GDBN} works without them.
18416 @kindex maint ada set ignore-descriptive-types
18417 @item maintenance ada set ignore-descriptive-types [on|off]
18418 Control whether the debugger should ignore descriptive types.
18419 The default is not to ignore descriptives types (@code{off}).
18421 @kindex maint ada show ignore-descriptive-types
18422 @item maintenance ada show ignore-descriptive-types
18423 Show if descriptive types are ignored by @value{GDBN}.
18431 @value{GDBN} supports the
18432 @url{https://github.com/ROCm-Developer-Tools/HIP/blob/master/docs/markdown/hip_kernel_language.md,
18433 HIP Programming Language}.
18435 @c TODO: Add any language specific differences.
18437 @node Unsupported Languages
18438 @section Unsupported Languages
18440 @cindex unsupported languages
18441 @cindex minimal language
18442 In addition to the other fully-supported programming languages,
18443 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
18444 It does not represent a real programming language, but provides a set
18445 of capabilities close to what the C or assembly languages provide.
18446 This should allow most simple operations to be performed while debugging
18447 an application that uses a language currently not supported by @value{GDBN}.
18449 If the language is set to @code{auto}, @value{GDBN} will automatically
18450 select this language if the current frame corresponds to an unsupported
18454 @chapter Examining the Symbol Table
18456 The commands described in this chapter allow you to inquire about the
18457 symbols (names of variables, functions and types) defined in your
18458 program. This information is inherent in the text of your program and
18459 does not change as your program executes. @value{GDBN} finds it in your
18460 program's symbol table, in the file indicated when you started @value{GDBN}
18461 (@pxref{File Options, ,Choosing Files}), or by one of the
18462 file-management commands (@pxref{Files, ,Commands to Specify Files}).
18464 @cindex symbol names
18465 @cindex names of symbols
18466 @cindex quoting names
18467 @anchor{quoting names}
18468 Occasionally, you may need to refer to symbols that contain unusual
18469 characters, which @value{GDBN} ordinarily treats as word delimiters. The
18470 most frequent case is in referring to static variables in other
18471 source files (@pxref{Variables,,Program Variables}). File names
18472 are recorded in object files as debugging symbols, but @value{GDBN} would
18473 ordinarily parse a typical file name, like @file{foo.c}, as the three words
18474 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
18475 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
18482 looks up the value of @code{x} in the scope of the file @file{foo.c}.
18485 @cindex case-insensitive symbol names
18486 @cindex case sensitivity in symbol names
18487 @kindex set case-sensitive
18488 @item set case-sensitive on
18489 @itemx set case-sensitive off
18490 @itemx set case-sensitive auto
18491 Normally, when @value{GDBN} looks up symbols, it matches their names
18492 with case sensitivity determined by the current source language.
18493 Occasionally, you may wish to control that. The command @code{set
18494 case-sensitive} lets you do that by specifying @code{on} for
18495 case-sensitive matches or @code{off} for case-insensitive ones. If
18496 you specify @code{auto}, case sensitivity is reset to the default
18497 suitable for the source language. The default is case-sensitive
18498 matches for all languages except for Fortran, for which the default is
18499 case-insensitive matches.
18501 @kindex show case-sensitive
18502 @item show case-sensitive
18503 This command shows the current setting of case sensitivity for symbols
18506 @kindex set print type methods
18507 @item set print type methods
18508 @itemx set print type methods on
18509 @itemx set print type methods off
18510 Normally, when @value{GDBN} prints a class, it displays any methods
18511 declared in that class. You can control this behavior either by
18512 passing the appropriate flag to @code{ptype}, or using @command{set
18513 print type methods}. Specifying @code{on} will cause @value{GDBN} to
18514 display the methods; this is the default. Specifying @code{off} will
18515 cause @value{GDBN} to omit the methods.
18517 @kindex show print type methods
18518 @item show print type methods
18519 This command shows the current setting of method display when printing
18522 @kindex set print type nested-type-limit
18523 @item set print type nested-type-limit @var{limit}
18524 @itemx set print type nested-type-limit unlimited
18525 Set the limit of displayed nested types that the type printer will
18526 show. A @var{limit} of @code{unlimited} or @code{-1} will show all
18527 nested definitions. By default, the type printer will not show any nested
18528 types defined in classes.
18530 @kindex show print type nested-type-limit
18531 @item show print type nested-type-limit
18532 This command shows the current display limit of nested types when
18535 @kindex set print type typedefs
18536 @item set print type typedefs
18537 @itemx set print type typedefs on
18538 @itemx set print type typedefs off
18540 Normally, when @value{GDBN} prints a class, it displays any typedefs
18541 defined in that class. You can control this behavior either by
18542 passing the appropriate flag to @code{ptype}, or using @command{set
18543 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
18544 display the typedef definitions; this is the default. Specifying
18545 @code{off} will cause @value{GDBN} to omit the typedef definitions.
18546 Note that this controls whether the typedef definition itself is
18547 printed, not whether typedef names are substituted when printing other
18550 @kindex show print type typedefs
18551 @item show print type typedefs
18552 This command shows the current setting of typedef display when
18555 @kindex info address
18556 @cindex address of a symbol
18557 @item info address @var{symbol}
18558 Describe where the data for @var{symbol} is stored. For a register
18559 variable, this says which register it is kept in. For a non-register
18560 local variable, this prints the stack-frame offset at which the variable
18563 Note the contrast with @samp{print &@var{symbol}}, which does not work
18564 at all for a register variable, and for a stack local variable prints
18565 the exact address of the current instantiation of the variable.
18567 @kindex info symbol
18568 @cindex symbol from address
18569 @cindex closest symbol and offset for an address
18570 @item info symbol @var{addr}
18571 Print the name of a symbol which is stored at the address @var{addr}.
18572 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
18573 nearest symbol and an offset from it:
18576 (@value{GDBP}) info symbol 0x54320
18577 _initialize_vx + 396 in section .text
18581 This is the opposite of the @code{info address} command. You can use
18582 it to find out the name of a variable or a function given its address.
18584 For dynamically linked executables, the name of executable or shared
18585 library containing the symbol is also printed:
18588 (@value{GDBP}) info symbol 0x400225
18589 _start + 5 in section .text of /tmp/a.out
18590 (@value{GDBP}) info symbol 0x2aaaac2811cf
18591 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
18596 @item demangle @r{[}-l @var{language}@r{]} @r{[}@var{--}@r{]} @var{name}
18597 Demangle @var{name}.
18598 If @var{language} is provided it is the name of the language to demangle
18599 @var{name} in. Otherwise @var{name} is demangled in the current language.
18601 The @samp{--} option specifies the end of options,
18602 and is useful when @var{name} begins with a dash.
18604 The parameter @code{demangle-style} specifies how to interpret the kind
18605 of mangling used. @xref{Print Settings}.
18608 @item whatis[/@var{flags}] [@var{arg}]
18609 Print the data type of @var{arg}, which can be either an expression
18610 or a name of a data type. With no argument, print the data type of
18611 @code{$}, the last value in the value history.
18613 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
18614 is not actually evaluated, and any side-effecting operations (such as
18615 assignments or function calls) inside it do not take place.
18617 If @var{arg} is a variable or an expression, @code{whatis} prints its
18618 literal type as it is used in the source code. If the type was
18619 defined using a @code{typedef}, @code{whatis} will @emph{not} print
18620 the data type underlying the @code{typedef}. If the type of the
18621 variable or the expression is a compound data type, such as
18622 @code{struct} or @code{class}, @code{whatis} never prints their
18623 fields or methods. It just prints the @code{struct}/@code{class}
18624 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
18625 such a compound data type, use @code{ptype}.
18627 If @var{arg} is a type name that was defined using @code{typedef},
18628 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
18629 Unrolling means that @code{whatis} will show the underlying type used
18630 in the @code{typedef} declaration of @var{arg}. However, if that
18631 underlying type is also a @code{typedef}, @code{whatis} will not
18634 For C code, the type names may also have the form @samp{class
18635 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
18636 @var{union-tag}} or @samp{enum @var{enum-tag}}.
18638 @var{flags} can be used to modify how the type is displayed.
18639 Available flags are:
18643 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
18644 parameters and typedefs defined in a class when printing the class'
18645 members. The @code{/r} flag disables this.
18648 Do not print methods defined in the class.
18651 Print methods defined in the class. This is the default, but the flag
18652 exists in case you change the default with @command{set print type methods}.
18655 Do not print typedefs defined in the class. Note that this controls
18656 whether the typedef definition itself is printed, not whether typedef
18657 names are substituted when printing other types.
18660 Print typedefs defined in the class. This is the default, but the flag
18661 exists in case you change the default with @command{set print type typedefs}.
18664 Print the offsets and sizes of fields in a struct, similar to what the
18665 @command{pahole} tool does. This option implies the @code{/tm} flags.
18667 For example, given the following declarations:
18703 Issuing a @kbd{ptype /o struct tuv} command would print:
18706 (@value{GDBP}) ptype /o struct tuv
18707 /* offset | size */ type = struct tuv @{
18708 /* 0 | 4 */ int a1;
18709 /* XXX 4-byte hole */
18710 /* 8 | 8 */ char *a2;
18711 /* 16 | 4 */ int a3;
18713 /* total size (bytes): 24 */
18717 Notice the format of the first column of comments. There, you can
18718 find two parts separated by the @samp{|} character: the @emph{offset},
18719 which indicates where the field is located inside the struct, in
18720 bytes, and the @emph{size} of the field. Another interesting line is
18721 the marker of a @emph{hole} in the struct, indicating that it may be
18722 possible to pack the struct and make it use less space by reorganizing
18725 It is also possible to print offsets inside an union:
18728 (@value{GDBP}) ptype /o union qwe
18729 /* offset | size */ type = union qwe @{
18730 /* 24 */ struct tuv @{
18731 /* 0 | 4 */ int a1;
18732 /* XXX 4-byte hole */
18733 /* 8 | 8 */ char *a2;
18734 /* 16 | 4 */ int a3;
18736 /* total size (bytes): 24 */
18738 /* 40 */ struct xyz @{
18739 /* 0 | 4 */ int f1;
18740 /* 4 | 1 */ char f2;
18741 /* XXX 3-byte hole */
18742 /* 8 | 8 */ void *f3;
18743 /* 16 | 24 */ struct tuv @{
18744 /* 16 | 4 */ int a1;
18745 /* XXX 4-byte hole */
18746 /* 24 | 8 */ char *a2;
18747 /* 32 | 4 */ int a3;
18749 /* total size (bytes): 24 */
18752 /* total size (bytes): 40 */
18755 /* total size (bytes): 40 */
18759 In this case, since @code{struct tuv} and @code{struct xyz} occupy the
18760 same space (because we are dealing with an union), the offset is not
18761 printed for them. However, you can still examine the offset of each
18762 of these structures' fields.
18764 Another useful scenario is printing the offsets of a struct containing
18768 (@value{GDBP}) ptype /o struct tyu
18769 /* offset | size */ type = struct tyu @{
18770 /* 0:31 | 4 */ int a1 : 1;
18771 /* 0:28 | 4 */ int a2 : 3;
18772 /* 0: 5 | 4 */ int a3 : 23;
18773 /* 3: 3 | 1 */ signed char a4 : 2;
18774 /* XXX 3-bit hole */
18775 /* XXX 4-byte hole */
18776 /* 8 | 8 */ int64_t a5;
18777 /* 16: 0 | 4 */ int a6 : 5;
18778 /* 16: 5 | 8 */ int64_t a7 : 3;
18779 "/* XXX 7-byte padding */
18781 /* total size (bytes): 24 */
18785 Note how the offset information is now extended to also include the
18786 first bit of the bitfield.
18790 @item ptype[/@var{flags}] [@var{arg}]
18791 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
18792 detailed description of the type, instead of just the name of the type.
18793 @xref{Expressions, ,Expressions}.
18795 Contrary to @code{whatis}, @code{ptype} always unrolls any
18796 @code{typedef}s in its argument declaration, whether the argument is
18797 a variable, expression, or a data type. This means that @code{ptype}
18798 of a variable or an expression will not print literally its type as
18799 present in the source code---use @code{whatis} for that. @code{typedef}s at
18800 the pointer or reference targets are also unrolled. Only @code{typedef}s of
18801 fields, methods and inner @code{class typedef}s of @code{struct}s,
18802 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
18804 For example, for this variable declaration:
18807 typedef double real_t;
18808 struct complex @{ real_t real; double imag; @};
18809 typedef struct complex complex_t;
18811 real_t *real_pointer_var;
18815 the two commands give this output:
18819 (@value{GDBP}) whatis var
18821 (@value{GDBP}) ptype var
18822 type = struct complex @{
18826 (@value{GDBP}) whatis complex_t
18827 type = struct complex
18828 (@value{GDBP}) whatis struct complex
18829 type = struct complex
18830 (@value{GDBP}) ptype struct complex
18831 type = struct complex @{
18835 (@value{GDBP}) whatis real_pointer_var
18837 (@value{GDBP}) ptype real_pointer_var
18843 As with @code{whatis}, using @code{ptype} without an argument refers to
18844 the type of @code{$}, the last value in the value history.
18846 @cindex incomplete type
18847 Sometimes, programs use opaque data types or incomplete specifications
18848 of complex data structure. If the debug information included in the
18849 program does not allow @value{GDBN} to display a full declaration of
18850 the data type, it will say @samp{<incomplete type>}. For example,
18851 given these declarations:
18855 struct foo *fooptr;
18859 but no definition for @code{struct foo} itself, @value{GDBN} will say:
18862 (@value{GDBP}) ptype foo
18863 $1 = <incomplete type>
18867 ``Incomplete type'' is C terminology for data types that are not
18868 completely specified.
18870 @cindex unknown type
18871 Othertimes, information about a variable's type is completely absent
18872 from the debug information included in the program. This most often
18873 happens when the program or library where the variable is defined
18874 includes no debug information at all. @value{GDBN} knows the variable
18875 exists from inspecting the linker/loader symbol table (e.g., the ELF
18876 dynamic symbol table), but such symbols do not contain type
18877 information. Inspecting the type of a (global) variable for which
18878 @value{GDBN} has no type information shows:
18881 (@value{GDBP}) ptype var
18882 type = <data variable, no debug info>
18885 @xref{Variables, no debug info variables}, for how to print the values
18889 @item info types [-q] [@var{regexp}]
18890 Print a brief description of all types whose names match the regular
18891 expression @var{regexp} (or all types in your program, if you supply
18892 no argument). Each complete typename is matched as though it were a
18893 complete line; thus, @samp{i type value} gives information on all
18894 types in your program whose names include the string @code{value}, but
18895 @samp{i type ^value$} gives information only on types whose complete
18896 name is @code{value}.
18898 In programs using different languages, @value{GDBN} chooses the syntax
18899 to print the type description according to the
18900 @samp{set language} value: using @samp{set language auto}
18901 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18902 language of the type, other values mean to use
18903 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18905 This command differs from @code{ptype} in two ways: first, like
18906 @code{whatis}, it does not print a detailed description; second, it
18907 lists all source files and line numbers where a type is defined.
18909 The output from @samp{into types} is proceeded with a header line
18910 describing what types are being listed. The optional flag @samp{-q},
18911 which stands for @samp{quiet}, disables printing this header
18914 @kindex info type-printers
18915 @item info type-printers
18916 Versions of @value{GDBN} that ship with Python scripting enabled may
18917 have ``type printers'' available. When using @command{ptype} or
18918 @command{whatis}, these printers are consulted when the name of a type
18919 is needed. @xref{Type Printing API}, for more information on writing
18922 @code{info type-printers} displays all the available type printers.
18924 @kindex enable type-printer
18925 @kindex disable type-printer
18926 @item enable type-printer @var{name}@dots{}
18927 @item disable type-printer @var{name}@dots{}
18928 These commands can be used to enable or disable type printers.
18931 @cindex local variables
18932 @item info scope @var{location}
18933 List all the variables local to a particular scope. This command
18934 accepts a @var{location} argument---a function name, a source line, or
18935 an address preceded by a @samp{*}, and prints all the variables local
18936 to the scope defined by that location. (@xref{Specify Location}, for
18937 details about supported forms of @var{location}.) For example:
18940 (@value{GDBP}) @b{info scope command_line_handler}
18941 Scope for command_line_handler:
18942 Symbol rl is an argument at stack/frame offset 8, length 4.
18943 Symbol linebuffer is in static storage at address 0x150a18, length 4.
18944 Symbol linelength is in static storage at address 0x150a1c, length 4.
18945 Symbol p is a local variable in register $esi, length 4.
18946 Symbol p1 is a local variable in register $ebx, length 4.
18947 Symbol nline is a local variable in register $edx, length 4.
18948 Symbol repeat is a local variable at frame offset -8, length 4.
18952 This command is especially useful for determining what data to collect
18953 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
18956 @kindex info source
18958 Show information about the current source file---that is, the source file for
18959 the function containing the current point of execution:
18962 the name of the source file, and the directory containing it,
18964 the directory it was compiled in,
18966 its length, in lines,
18968 which programming language it is written in,
18970 if the debug information provides it, the program that compiled the file
18971 (which may include, e.g., the compiler version and command line arguments),
18973 whether the executable includes debugging information for that file, and
18974 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
18976 whether the debugging information includes information about
18977 preprocessor macros.
18981 @kindex info sources
18983 Print the names of all source files in your program for which there is
18984 debugging information, organized into two lists: files whose symbols
18985 have already been read, and files whose symbols will be read when needed.
18987 @item info sources [-dirname | -basename] [--] [@var{regexp}]
18988 Like @samp{info sources}, but only print the names of the files
18989 matching the provided @var{regexp}.
18990 By default, the @var{regexp} is used to match anywhere in the filename.
18991 If @code{-dirname}, only files having a dirname matching @var{regexp} are shown.
18992 If @code{-basename}, only files having a basename matching @var{regexp}
18994 The matching is case-sensitive, except on operating systems that
18995 have case-insensitive filesystem (e.g., MS-Windows).
18997 @kindex info functions
18998 @item info functions [-q] [-n]
18999 Print the names and data types of all defined functions.
19000 Similarly to @samp{info types}, this command groups its output by source
19001 files and annotates each function definition with its source line
19004 In programs using different languages, @value{GDBN} chooses the syntax
19005 to print the function name and type according to the
19006 @samp{set language} value: using @samp{set language auto}
19007 (see @ref{Automatically, ,Set Language Automatically}) means to use the
19008 language of the function, other values mean to use
19009 the manually specified language (see @ref{Manually, ,Set Language Manually}).
19011 The @samp{-n} flag excludes @dfn{non-debugging symbols} from the
19012 results. A non-debugging symbol is a symbol that comes from the
19013 executable's symbol table, not from the debug information (for
19014 example, DWARF) associated with the executable.
19016 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
19017 printing header information and messages explaining why no functions
19020 @item info functions [-q] [-n] [-t @var{type_regexp}] [@var{regexp}]
19021 Like @samp{info functions}, but only print the names and data types
19022 of the functions selected with the provided regexp(s).
19024 If @var{regexp} is provided, print only the functions whose names
19025 match the regular expression @var{regexp}.
19026 Thus, @samp{info fun step} finds all functions whose
19027 names include @code{step}; @samp{info fun ^step} finds those whose names
19028 start with @code{step}. If a function name contains characters that
19029 conflict with the regular expression language (e.g.@:
19030 @samp{operator*()}), they may be quoted with a backslash.
19032 If @var{type_regexp} is provided, print only the functions whose
19033 types, as printed by the @code{whatis} command, match
19034 the regular expression @var{type_regexp}.
19035 If @var{type_regexp} contains space(s), it should be enclosed in
19036 quote characters. If needed, use backslash to escape the meaning
19037 of special characters or quotes.
19038 Thus, @samp{info fun -t '^int ('} finds the functions that return
19039 an integer; @samp{info fun -t '(.*int.*'} finds the functions that
19040 have an argument type containing int; @samp{info fun -t '^int (' ^step}
19041 finds the functions whose names start with @code{step} and that return
19044 If both @var{regexp} and @var{type_regexp} are provided, a function
19045 is printed only if its name matches @var{regexp} and its type matches
19049 @kindex info variables
19050 @item info variables [-q] [-n]
19051 Print the names and data types of all variables that are defined
19052 outside of functions (i.e.@: excluding local variables).
19053 The printed variables are grouped by source files and annotated with
19054 their respective source line numbers.
19056 In programs using different languages, @value{GDBN} chooses the syntax
19057 to print the variable name and type according to the
19058 @samp{set language} value: using @samp{set language auto}
19059 (see @ref{Automatically, ,Set Language Automatically}) means to use the
19060 language of the variable, other values mean to use
19061 the manually specified language (see @ref{Manually, ,Set Language Manually}).
19063 The @samp{-n} flag excludes non-debugging symbols from the results.
19065 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
19066 printing header information and messages explaining why no variables
19069 @item info variables [-q] [-n] [-t @var{type_regexp}] [@var{regexp}]
19070 Like @kbd{info variables}, but only print the variables selected
19071 with the provided regexp(s).
19073 If @var{regexp} is provided, print only the variables whose names
19074 match the regular expression @var{regexp}.
19076 If @var{type_regexp} is provided, print only the variables whose
19077 types, as printed by the @code{whatis} command, match
19078 the regular expression @var{type_regexp}.
19079 If @var{type_regexp} contains space(s), it should be enclosed in
19080 quote characters. If needed, use backslash to escape the meaning
19081 of special characters or quotes.
19083 If both @var{regexp} and @var{type_regexp} are provided, an argument
19084 is printed only if its name matches @var{regexp} and its type matches
19087 @kindex info modules
19089 @item info modules @r{[}-q@r{]} @r{[}@var{regexp}@r{]}
19090 List all Fortran modules in the program, or all modules matching the
19091 optional regular expression @var{regexp}.
19093 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
19094 printing header information and messages explaining why no modules
19097 @kindex info module
19098 @cindex Fortran modules, information about
19099 @cindex functions and variables by Fortran module
19100 @cindex module functions and variables
19101 @item info module functions @r{[}-q@r{]} @r{[}-m @var{module-regexp}@r{]} @r{[}-t @var{type-regexp}@r{]} @r{[}@var{regexp}@r{]}
19102 @itemx info module variables @r{[}-q@r{]} @r{[}-m @var{module-regexp}@r{]} @r{[}-t @var{type-regexp}@r{]} @r{[}@var{regexp}@r{]}
19103 List all functions or variables within all Fortran modules. The set
19104 of functions or variables listed can be limited by providing some or
19105 all of the optional regular expressions. If @var{module-regexp} is
19106 provided, then only Fortran modules matching @var{module-regexp} will
19107 be searched. Only functions or variables whose type matches the
19108 optional regular expression @var{type-regexp} will be listed. And
19109 only functions or variables whose name matches the optional regular
19110 expression @var{regexp} will be listed.
19112 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
19113 printing header information and messages explaining why no functions
19114 or variables have been printed.
19116 @kindex info classes
19117 @cindex Objective-C, classes and selectors
19119 @itemx info classes @var{regexp}
19120 Display all Objective-C classes in your program, or
19121 (with the @var{regexp} argument) all those matching a particular regular
19124 @kindex info selectors
19125 @item info selectors
19126 @itemx info selectors @var{regexp}
19127 Display all Objective-C selectors in your program, or
19128 (with the @var{regexp} argument) all those matching a particular regular
19132 This was never implemented.
19133 @kindex info methods
19135 @itemx info methods @var{regexp}
19136 The @code{info methods} command permits the user to examine all defined
19137 methods within C@t{++} program, or (with the @var{regexp} argument) a
19138 specific set of methods found in the various C@t{++} classes. Many
19139 C@t{++} classes provide a large number of methods. Thus, the output
19140 from the @code{ptype} command can be overwhelming and hard to use. The
19141 @code{info-methods} command filters the methods, printing only those
19142 which match the regular-expression @var{regexp}.
19145 @cindex opaque data types
19146 @kindex set opaque-type-resolution
19147 @item set opaque-type-resolution on
19148 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
19149 declared as a pointer to a @code{struct}, @code{class}, or
19150 @code{union}---for example, @code{struct MyType *}---that is used in one
19151 source file although the full declaration of @code{struct MyType} is in
19152 another source file. The default is on.
19154 A change in the setting of this subcommand will not take effect until
19155 the next time symbols for a file are loaded.
19157 @item set opaque-type-resolution off
19158 Tell @value{GDBN} not to resolve opaque types. In this case, the type
19159 is printed as follows:
19161 @{<no data fields>@}
19164 @kindex show opaque-type-resolution
19165 @item show opaque-type-resolution
19166 Show whether opaque types are resolved or not.
19168 @kindex set print symbol-loading
19169 @cindex print messages when symbols are loaded
19170 @item set print symbol-loading
19171 @itemx set print symbol-loading full
19172 @itemx set print symbol-loading brief
19173 @itemx set print symbol-loading off
19174 The @code{set print symbol-loading} command allows you to control the
19175 printing of messages when @value{GDBN} loads symbol information.
19176 By default a message is printed for the executable and one for each
19177 shared library, and normally this is what you want. However, when
19178 debugging apps with large numbers of shared libraries these messages
19180 When set to @code{brief} a message is printed for each executable,
19181 and when @value{GDBN} loads a collection of shared libraries at once
19182 it will only print one message regardless of the number of shared
19183 libraries. When set to @code{off} no messages are printed.
19185 @kindex show print symbol-loading
19186 @item show print symbol-loading
19187 Show whether messages will be printed when a @value{GDBN} command
19188 entered from the keyboard causes symbol information to be loaded.
19190 @kindex maint print symbols
19191 @cindex symbol dump
19192 @kindex maint print psymbols
19193 @cindex partial symbol dump
19194 @kindex maint print msymbols
19195 @cindex minimal symbol dump
19196 @item maint print symbols @r{[}-pc @var{address}@r{]} @r{[}@var{filename}@r{]}
19197 @itemx maint print symbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
19198 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-pc @var{address}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
19199 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
19200 @itemx maint print msymbols @r{[}-objfile @var{objfile}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
19201 Write a dump of debugging symbol data into the file @var{filename} or
19202 the terminal if @var{filename} is unspecified.
19203 If @code{-objfile @var{objfile}} is specified, only dump symbols for
19205 If @code{-pc @var{address}} is specified, only dump symbols for the file
19206 with code at that address. Note that @var{address} may be a symbol like
19208 If @code{-source @var{source}} is specified, only dump symbols for that
19211 These commands are used to debug the @value{GDBN} symbol-reading code.
19212 These commands do not modify internal @value{GDBN} state, therefore
19213 @samp{maint print symbols} will only print symbols for already expanded symbol
19215 You can use the command @code{info sources} to find out which files these are.
19216 If you use @samp{maint print psymbols} instead, the dump shows information
19217 about symbols that @value{GDBN} only knows partially---that is, symbols
19218 defined in files that @value{GDBN} has skimmed, but not yet read completely.
19219 Finally, @samp{maint print msymbols} just dumps ``minimal symbols'', e.g.,
19222 @xref{Files, ,Commands to Specify Files}, for a discussion of how
19223 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
19225 @kindex maint info symtabs
19226 @kindex maint info psymtabs
19227 @cindex listing @value{GDBN}'s internal symbol tables
19228 @cindex symbol tables, listing @value{GDBN}'s internal
19229 @cindex full symbol tables, listing @value{GDBN}'s internal
19230 @cindex partial symbol tables, listing @value{GDBN}'s internal
19231 @item maint info symtabs @r{[} @var{regexp} @r{]}
19232 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
19234 List the @code{struct symtab} or @code{struct partial_symtab}
19235 structures whose names match @var{regexp}. If @var{regexp} is not
19236 given, list them all. The output includes expressions which you can
19237 copy into a @value{GDBN} debugging this one to examine a particular
19238 structure in more detail. For example:
19241 (@value{GDBP}) maint info psymtabs dwarf2read
19242 @{ objfile /home/gnu/build/gdb/gdb
19243 ((struct objfile *) 0x82e69d0)
19244 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
19245 ((struct partial_symtab *) 0x8474b10)
19248 text addresses 0x814d3c8 -- 0x8158074
19249 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
19250 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
19251 dependencies (none)
19254 (@value{GDBP}) maint info symtabs
19258 We see that there is one partial symbol table whose filename contains
19259 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
19260 and we see that @value{GDBN} has not read in any symtabs yet at all.
19261 If we set a breakpoint on a function, that will cause @value{GDBN} to
19262 read the symtab for the compilation unit containing that function:
19265 (@value{GDBP}) break dwarf2_psymtab_to_symtab
19266 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
19268 (@value{GDBP}) maint info symtabs
19269 @{ objfile /home/gnu/build/gdb/gdb
19270 ((struct objfile *) 0x82e69d0)
19271 @{ symtab /home/gnu/src/gdb/dwarf2read.c
19272 ((struct symtab *) 0x86c1f38)
19275 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
19276 linetable ((struct linetable *) 0x8370fa0)
19277 debugformat DWARF 2
19283 @kindex maint info line-table
19284 @cindex listing @value{GDBN}'s internal line tables
19285 @cindex line tables, listing @value{GDBN}'s internal
19286 @item maint info line-table @r{[} @var{regexp} @r{]}
19288 List the @code{struct linetable} from all @code{struct symtab}
19289 instances whose name matches @var{regexp}. If @var{regexp} is not
19290 given, list the @code{struct linetable} from all @code{struct symtab}.
19292 @kindex maint set symbol-cache-size
19293 @cindex symbol cache size
19294 @item maint set symbol-cache-size @var{size}
19295 Set the size of the symbol cache to @var{size}.
19296 The default size is intended to be good enough for debugging
19297 most applications. This option exists to allow for experimenting
19298 with different sizes.
19300 @kindex maint show symbol-cache-size
19301 @item maint show symbol-cache-size
19302 Show the size of the symbol cache.
19304 @kindex maint print symbol-cache
19305 @cindex symbol cache, printing its contents
19306 @item maint print symbol-cache
19307 Print the contents of the symbol cache.
19308 This is useful when debugging symbol cache issues.
19310 @kindex maint print symbol-cache-statistics
19311 @cindex symbol cache, printing usage statistics
19312 @item maint print symbol-cache-statistics
19313 Print symbol cache usage statistics.
19314 This helps determine how well the cache is being utilized.
19316 @kindex maint flush-symbol-cache
19317 @cindex symbol cache, flushing
19318 @item maint flush-symbol-cache
19319 Flush the contents of the symbol cache, all entries are removed.
19320 This command is useful when debugging the symbol cache.
19321 It is also useful when collecting performance data.
19326 @chapter Altering Execution
19328 Once you think you have found an error in your program, you might want to
19329 find out for certain whether correcting the apparent error would lead to
19330 correct results in the rest of the run. You can find the answer by
19331 experiment, using the @value{GDBN} features for altering execution of the
19334 For example, you can store new values into variables or memory
19335 locations, give your program a signal, restart it at a different
19336 address, or even return prematurely from a function.
19339 * Assignment:: Assignment to variables
19340 * Jumping:: Continuing at a different address
19341 * Signaling:: Giving your program a signal
19342 * Returning:: Returning from a function
19343 * Calling:: Calling your program's functions
19344 * Patching:: Patching your program
19345 * Compiling and Injecting Code:: Compiling and injecting code in @value{GDBN}
19349 @section Assignment to Variables
19352 @cindex setting variables
19353 To alter the value of a variable, evaluate an assignment expression.
19354 @xref{Expressions, ,Expressions}. For example,
19361 stores the value 4 into the variable @code{x}, and then prints the
19362 value of the assignment expression (which is 4).
19363 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
19364 information on operators in supported languages.
19366 @kindex set variable
19367 @cindex variables, setting
19368 If you are not interested in seeing the value of the assignment, use the
19369 @code{set} command instead of the @code{print} command. @code{set} is
19370 really the same as @code{print} except that the expression's value is
19371 not printed and is not put in the value history (@pxref{Value History,
19372 ,Value History}). The expression is evaluated only for its effects.
19374 If the beginning of the argument string of the @code{set} command
19375 appears identical to a @code{set} subcommand, use the @code{set
19376 variable} command instead of just @code{set}. This command is identical
19377 to @code{set} except for its lack of subcommands. For example, if your
19378 program has a variable @code{width}, you get an error if you try to set
19379 a new value with just @samp{set width=13}, because @value{GDBN} has the
19380 command @code{set width}:
19383 (@value{GDBP}) whatis width
19385 (@value{GDBP}) p width
19387 (@value{GDBP}) set width=47
19388 Invalid syntax in expression.
19392 The invalid expression, of course, is @samp{=47}. In
19393 order to actually set the program's variable @code{width}, use
19396 (@value{GDBP}) set var width=47
19399 Because the @code{set} command has many subcommands that can conflict
19400 with the names of program variables, it is a good idea to use the
19401 @code{set variable} command instead of just @code{set}. For example, if
19402 your program has a variable @code{g}, you run into problems if you try
19403 to set a new value with just @samp{set g=4}, because @value{GDBN} has
19404 the command @code{set gnutarget}, abbreviated @code{set g}:
19408 (@value{GDBP}) whatis g
19412 (@value{GDBP}) set g=4
19416 The program being debugged has been started already.
19417 Start it from the beginning? (y or n) y
19418 Starting program: /home/smith/cc_progs/a.out
19419 "/home/smith/cc_progs/a.out": can't open to read symbols:
19420 Invalid bfd target.
19421 (@value{GDBP}) show g
19422 The current BFD target is "=4".
19427 The program variable @code{g} did not change, and you silently set the
19428 @code{gnutarget} to an invalid value. In order to set the variable
19432 (@value{GDBP}) set var g=4
19435 @value{GDBN} allows more implicit conversions in assignments than C; you can
19436 freely store an integer value into a pointer variable or vice versa,
19437 and you can convert any structure to any other structure that is the
19438 same length or shorter.
19439 @comment FIXME: how do structs align/pad in these conversions?
19440 @comment /doc@cygnus.com 18dec1990
19442 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
19443 construct to generate a value of specified type at a specified address
19444 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
19445 to memory location @code{0x83040} as an integer (which implies a certain size
19446 and representation in memory), and
19449 set @{int@}0x83040 = 4
19453 stores the value 4 into that memory location.
19456 @section Continuing at a Different Address
19458 Ordinarily, when you continue your program, you do so at the place where
19459 it stopped, with the @code{continue} command. You can instead continue at
19460 an address of your own choosing, with the following commands:
19464 @kindex j @r{(@code{jump})}
19465 @item jump @var{location}
19466 @itemx j @var{location}
19467 Resume execution at @var{location}. Execution stops again immediately
19468 if there is a breakpoint there. @xref{Specify Location}, for a description
19469 of the different forms of @var{location}. It is common
19470 practice to use the @code{tbreak} command in conjunction with
19471 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
19473 The @code{jump} command does not change the current stack frame, or
19474 the stack pointer, or the contents of any memory location or any
19475 register other than the program counter. If @var{location} is in
19476 a different function from the one currently executing, the results may
19477 be bizarre if the two functions expect different patterns of arguments or
19478 of local variables. For this reason, the @code{jump} command requests
19479 confirmation if the specified line is not in the function currently
19480 executing. However, even bizarre results are predictable if you are
19481 well acquainted with the machine-language code of your program.
19484 On many systems, you can get much the same effect as the @code{jump}
19485 command by storing a new value into the register @code{$pc}. The
19486 difference is that this does not start your program running; it only
19487 changes the address of where it @emph{will} run when you continue. For
19495 makes the next @code{continue} command or stepping command execute at
19496 address @code{0x485}, rather than at the address where your program stopped.
19497 @xref{Continuing and Stepping, ,Continuing and Stepping}.
19499 The most common occasion to use the @code{jump} command is to back
19500 up---perhaps with more breakpoints set---over a portion of a program
19501 that has already executed, in order to examine its execution in more
19506 @section Giving your Program a Signal
19507 @cindex deliver a signal to a program
19511 @item signal @var{signal}
19512 Resume execution where your program is stopped, but immediately give it the
19513 signal @var{signal}. The @var{signal} can be the name or the number of a
19514 signal. For example, on many systems @code{signal 2} and @code{signal
19515 SIGINT} are both ways of sending an interrupt signal.
19517 Alternatively, if @var{signal} is zero, continue execution without
19518 giving a signal. This is useful when your program stopped on account of
19519 a signal and would ordinarily see the signal when resumed with the
19520 @code{continue} command; @samp{signal 0} causes it to resume without a
19523 @emph{Note:} When resuming a multi-threaded program, @var{signal} is
19524 delivered to the currently selected thread, not the thread that last
19525 reported a stop. This includes the situation where a thread was
19526 stopped due to a signal. So if you want to continue execution
19527 suppressing the signal that stopped a thread, you should select that
19528 same thread before issuing the @samp{signal 0} command. If you issue
19529 the @samp{signal 0} command with another thread as the selected one,
19530 @value{GDBN} detects that and asks for confirmation.
19532 Invoking the @code{signal} command is not the same as invoking the
19533 @code{kill} utility from the shell. Sending a signal with @code{kill}
19534 causes @value{GDBN} to decide what to do with the signal depending on
19535 the signal handling tables (@pxref{Signals}). The @code{signal} command
19536 passes the signal directly to your program.
19538 @code{signal} does not repeat when you press @key{RET} a second time
19539 after executing the command.
19541 @kindex queue-signal
19542 @item queue-signal @var{signal}
19543 Queue @var{signal} to be delivered immediately to the current thread
19544 when execution of the thread resumes. The @var{signal} can be the name or
19545 the number of a signal. For example, on many systems @code{signal 2} and
19546 @code{signal SIGINT} are both ways of sending an interrupt signal.
19547 The handling of the signal must be set to pass the signal to the program,
19548 otherwise @value{GDBN} will report an error.
19549 You can control the handling of signals from @value{GDBN} with the
19550 @code{handle} command (@pxref{Signals}).
19552 Alternatively, if @var{signal} is zero, any currently queued signal
19553 for the current thread is discarded and when execution resumes no signal
19554 will be delivered. This is useful when your program stopped on account
19555 of a signal and would ordinarily see the signal when resumed with the
19556 @code{continue} command.
19558 This command differs from the @code{signal} command in that the signal
19559 is just queued, execution is not resumed. And @code{queue-signal} cannot
19560 be used to pass a signal whose handling state has been set to @code{nopass}
19565 @xref{stepping into signal handlers}, for information on how stepping
19566 commands behave when the thread has a signal queued.
19569 @section Returning from a Function
19572 @cindex returning from a function
19575 @itemx return @var{expression}
19576 You can cancel execution of a function call with the @code{return}
19577 command. If you give an
19578 @var{expression} argument, its value is used as the function's return
19582 When you use @code{return}, @value{GDBN} discards the selected stack frame
19583 (and all frames within it). You can think of this as making the
19584 discarded frame return prematurely. If you wish to specify a value to
19585 be returned, give that value as the argument to @code{return}.
19587 This pops the selected stack frame (@pxref{Selection, ,Selecting a
19588 Frame}), and any other frames inside of it, leaving its caller as the
19589 innermost remaining frame. That frame becomes selected. The
19590 specified value is stored in the registers used for returning values
19593 The @code{return} command does not resume execution; it leaves the
19594 program stopped in the state that would exist if the function had just
19595 returned. In contrast, the @code{finish} command (@pxref{Continuing
19596 and Stepping, ,Continuing and Stepping}) resumes execution until the
19597 selected stack frame returns naturally.
19599 @value{GDBN} needs to know how the @var{expression} argument should be set for
19600 the inferior. The concrete registers assignment depends on the OS ABI and the
19601 type being returned by the selected stack frame. For example it is common for
19602 OS ABI to return floating point values in FPU registers while integer values in
19603 CPU registers. Still some ABIs return even floating point values in CPU
19604 registers. Larger integer widths (such as @code{long long int}) also have
19605 specific placement rules. @value{GDBN} already knows the OS ABI from its
19606 current target so it needs to find out also the type being returned to make the
19607 assignment into the right register(s).
19609 Normally, the selected stack frame has debug info. @value{GDBN} will always
19610 use the debug info instead of the implicit type of @var{expression} when the
19611 debug info is available. For example, if you type @kbd{return -1}, and the
19612 function in the current stack frame is declared to return a @code{long long
19613 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
19614 into a @code{long long int}:
19617 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
19619 (@value{GDBP}) return -1
19620 Make func return now? (y or n) y
19621 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
19622 43 printf ("result=%lld\n", func ());
19626 However, if the selected stack frame does not have a debug info, e.g., if the
19627 function was compiled without debug info, @value{GDBN} has to find out the type
19628 to return from user. Specifying a different type by mistake may set the value
19629 in different inferior registers than the caller code expects. For example,
19630 typing @kbd{return -1} with its implicit type @code{int} would set only a part
19631 of a @code{long long int} result for a debug info less function (on 32-bit
19632 architectures). Therefore the user is required to specify the return type by
19633 an appropriate cast explicitly:
19636 Breakpoint 2, 0x0040050b in func ()
19637 (@value{GDBP}) return -1
19638 Return value type not available for selected stack frame.
19639 Please use an explicit cast of the value to return.
19640 (@value{GDBP}) return (long long int) -1
19641 Make selected stack frame return now? (y or n) y
19642 #0 0x00400526 in main ()
19647 @section Calling Program Functions
19650 @cindex calling functions
19651 @cindex inferior functions, calling
19652 @item print @var{expr}
19653 Evaluate the expression @var{expr} and display the resulting value.
19654 The expression may include calls to functions in the program being
19658 @item call @var{expr}
19659 Evaluate the expression @var{expr} without displaying @code{void}
19662 You can use this variant of the @code{print} command if you want to
19663 execute a function from your program that does not return anything
19664 (a.k.a.@: @dfn{a void function}), but without cluttering the output
19665 with @code{void} returned values that @value{GDBN} will otherwise
19666 print. If the result is not void, it is printed and saved in the
19670 It is possible for the function you call via the @code{print} or
19671 @code{call} command to generate a signal (e.g., if there's a bug in
19672 the function, or if you passed it incorrect arguments). What happens
19673 in that case is controlled by the @code{set unwindonsignal} command.
19675 Similarly, with a C@t{++} program it is possible for the function you
19676 call via the @code{print} or @code{call} command to generate an
19677 exception that is not handled due to the constraints of the dummy
19678 frame. In this case, any exception that is raised in the frame, but has
19679 an out-of-frame exception handler will not be found. GDB builds a
19680 dummy-frame for the inferior function call, and the unwinder cannot
19681 seek for exception handlers outside of this dummy-frame. What happens
19682 in that case is controlled by the
19683 @code{set unwind-on-terminating-exception} command.
19686 @item set unwindonsignal
19687 @kindex set unwindonsignal
19688 @cindex unwind stack in called functions
19689 @cindex call dummy stack unwinding
19690 Set unwinding of the stack if a signal is received while in a function
19691 that @value{GDBN} called in the program being debugged. If set to on,
19692 @value{GDBN} unwinds the stack it created for the call and restores
19693 the context to what it was before the call. If set to off (the
19694 default), @value{GDBN} stops in the frame where the signal was
19697 @item show unwindonsignal
19698 @kindex show unwindonsignal
19699 Show the current setting of stack unwinding in the functions called by
19702 @item set unwind-on-terminating-exception
19703 @kindex set unwind-on-terminating-exception
19704 @cindex unwind stack in called functions with unhandled exceptions
19705 @cindex call dummy stack unwinding on unhandled exception.
19706 Set unwinding of the stack if a C@t{++} exception is raised, but left
19707 unhandled while in a function that @value{GDBN} called in the program being
19708 debugged. If set to on (the default), @value{GDBN} unwinds the stack
19709 it created for the call and restores the context to what it was before
19710 the call. If set to off, @value{GDBN} the exception is delivered to
19711 the default C@t{++} exception handler and the inferior terminated.
19713 @item show unwind-on-terminating-exception
19714 @kindex show unwind-on-terminating-exception
19715 Show the current setting of stack unwinding in the functions called by
19718 @item set may-call-functions
19719 @kindex set may-call-functions
19720 @cindex disabling calling functions in the program
19721 @cindex calling functions in the program, disabling
19722 Set permission to call functions in the program.
19723 This controls whether @value{GDBN} will attempt to call functions in
19724 the program, such as with expressions in the @code{print} command. It
19725 defaults to @code{on}.
19727 To call a function in the program, @value{GDBN} has to temporarily
19728 modify the state of the inferior. This has potentially undesired side
19729 effects. Also, having @value{GDBN} call nested functions is likely to
19730 be erroneous and may even crash the program being debugged. You can
19731 avoid such hazards by forbidding @value{GDBN} from calling functions
19732 in the program being debugged. If calling functions in the program
19733 is forbidden, GDB will throw an error when a command (such as printing
19734 an expression) starts a function call in the program.
19736 @item show may-call-functions
19737 @kindex show may-call-functions
19738 Show permission to call functions in the program.
19742 @subsection Calling functions with no debug info
19744 @cindex no debug info functions
19745 Sometimes, a function you wish to call is missing debug information.
19746 In such case, @value{GDBN} does not know the type of the function,
19747 including the types of the function's parameters. To avoid calling
19748 the inferior function incorrectly, which could result in the called
19749 function functioning erroneously and even crash, @value{GDBN} refuses
19750 to call the function unless you tell it the type of the function.
19752 For prototyped (i.e.@: ANSI/ISO style) functions, there are two ways
19753 to do that. The simplest is to cast the call to the function's
19754 declared return type. For example:
19757 (@value{GDBP}) p getenv ("PATH")
19758 'getenv' has unknown return type; cast the call to its declared return type
19759 (@value{GDBP}) p (char *) getenv ("PATH")
19760 $1 = 0x7fffffffe7ba "/usr/local/bin:/"...
19763 Casting the return type of a no-debug function is equivalent to
19764 casting the function to a pointer to a prototyped function that has a
19765 prototype that matches the types of the passed-in arguments, and
19766 calling that. I.e., the call above is equivalent to:
19769 (@value{GDBP}) p ((char * (*) (const char *)) getenv) ("PATH")
19773 and given this prototyped C or C++ function with float parameters:
19776 float multiply (float v1, float v2) @{ return v1 * v2; @}
19780 these calls are equivalent:
19783 (@value{GDBP}) p (float) multiply (2.0f, 3.0f)
19784 (@value{GDBP}) p ((float (*) (float, float)) multiply) (2.0f, 3.0f)
19787 If the function you wish to call is declared as unprototyped (i.e.@:
19788 old K&R style), you must use the cast-to-function-pointer syntax, so
19789 that @value{GDBN} knows that it needs to apply default argument
19790 promotions (promote float arguments to double). @xref{ABI, float
19791 promotion}. For example, given this unprototyped C function with
19792 float parameters, and no debug info:
19796 multiply_noproto (v1, v2)
19804 you call it like this:
19807 (@value{GDBP}) p ((float (*) ()) multiply_noproto) (2.0f, 3.0f)
19811 @section Patching Programs
19813 @cindex patching binaries
19814 @cindex writing into executables
19815 @cindex writing into corefiles
19817 By default, @value{GDBN} opens the file containing your program's
19818 executable code (or the corefile) read-only. This prevents accidental
19819 alterations to machine code; but it also prevents you from intentionally
19820 patching your program's binary.
19822 If you'd like to be able to patch the binary, you can specify that
19823 explicitly with the @code{set write} command. For example, you might
19824 want to turn on internal debugging flags, or even to make emergency
19830 @itemx set write off
19831 If you specify @samp{set write on}, @value{GDBN} opens executable and
19832 core files for both reading and writing; if you specify @kbd{set write
19833 off} (the default), @value{GDBN} opens them read-only.
19835 If you have already loaded a file, you must load it again (using the
19836 @code{exec-file} or @code{core-file} command) after changing @code{set
19837 write}, for your new setting to take effect.
19841 Display whether executable files and core files are opened for writing
19842 as well as reading.
19845 @node Compiling and Injecting Code
19846 @section Compiling and injecting code in @value{GDBN}
19847 @cindex injecting code
19848 @cindex writing into executables
19849 @cindex compiling code
19851 @value{GDBN} supports on-demand compilation and code injection into
19852 programs running under @value{GDBN}. GCC 5.0 or higher built with
19853 @file{libcc1.so} must be installed for this functionality to be enabled.
19854 This functionality is implemented with the following commands.
19857 @kindex compile code
19858 @item compile code @var{source-code}
19859 @itemx compile code -raw @var{--} @var{source-code}
19860 Compile @var{source-code} with the compiler language found as the current
19861 language in @value{GDBN} (@pxref{Languages}). If compilation and
19862 injection is not supported with the current language specified in
19863 @value{GDBN}, or the compiler does not support this feature, an error
19864 message will be printed. If @var{source-code} compiles and links
19865 successfully, @value{GDBN} will load the object-code emitted,
19866 and execute it within the context of the currently selected inferior.
19867 It is important to note that the compiled code is executed immediately.
19868 After execution, the compiled code is removed from @value{GDBN} and any
19869 new types or variables you have defined will be deleted.
19871 The command allows you to specify @var{source-code} in two ways.
19872 The simplest method is to provide a single line of code to the command.
19876 compile code printf ("hello world\n");
19879 If you specify options on the command line as well as source code, they
19880 may conflict. The @samp{--} delimiter can be used to separate options
19881 from actual source code. E.g.:
19884 compile code -r -- printf ("hello world\n");
19887 Alternatively you can enter source code as multiple lines of text. To
19888 enter this mode, invoke the @samp{compile code} command without any text
19889 following the command. This will start the multiple-line editor and
19890 allow you to type as many lines of source code as required. When you
19891 have completed typing, enter @samp{end} on its own line to exit the
19896 >printf ("hello\n");
19897 >printf ("world\n");
19901 Specifying @samp{-raw}, prohibits @value{GDBN} from wrapping the
19902 provided @var{source-code} in a callable scope. In this case, you must
19903 specify the entry point of the code by defining a function named
19904 @code{_gdb_expr_}. The @samp{-raw} code cannot access variables of the
19905 inferior. Using @samp{-raw} option may be needed for example when
19906 @var{source-code} requires @samp{#include} lines which may conflict with
19907 inferior symbols otherwise.
19909 @kindex compile file
19910 @item compile file @var{filename}
19911 @itemx compile file -raw @var{filename}
19912 Like @code{compile code}, but take the source code from @var{filename}.
19915 compile file /home/user/example.c
19920 @item compile print [[@var{options}] --] @var{expr}
19921 @itemx compile print [[@var{options}] --] /@var{f} @var{expr}
19922 Compile and execute @var{expr} with the compiler language found as the
19923 current language in @value{GDBN} (@pxref{Languages}). By default the
19924 value of @var{expr} is printed in a format appropriate to its data type;
19925 you can choose a different format by specifying @samp{/@var{f}}, where
19926 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
19927 Formats}. The @code{compile print} command accepts the same options
19928 as the @code{print} command; see @ref{print options}.
19930 @item compile print [[@var{options}] --]
19931 @itemx compile print [[@var{options}] --] /@var{f}
19932 @cindex reprint the last value
19933 Alternatively you can enter the expression (source code producing it) as
19934 multiple lines of text. To enter this mode, invoke the @samp{compile print}
19935 command without any text following the command. This will start the
19936 multiple-line editor.
19940 The process of compiling and injecting the code can be inspected using:
19943 @anchor{set debug compile}
19944 @item set debug compile
19945 @cindex compile command debugging info
19946 Turns on or off display of @value{GDBN} process of compiling and
19947 injecting the code. The default is off.
19949 @item show debug compile
19950 Displays the current state of displaying @value{GDBN} process of
19951 compiling and injecting the code.
19953 @anchor{set debug compile-cplus-types}
19954 @item set debug compile-cplus-types
19955 @cindex compile C@t{++} type conversion
19956 Turns on or off the display of C@t{++} type conversion debugging information.
19957 The default is off.
19959 @item show debug compile-cplus-types
19960 Displays the current state of displaying debugging information for
19961 C@t{++} type conversion.
19964 @subsection Compilation options for the @code{compile} command
19966 @value{GDBN} needs to specify the right compilation options for the code
19967 to be injected, in part to make its ABI compatible with the inferior
19968 and in part to make the injected code compatible with @value{GDBN}'s
19972 The options used, in increasing precedence:
19975 @item target architecture and OS options (@code{gdbarch})
19976 These options depend on target processor type and target operating
19977 system, usually they specify at least 32-bit (@code{-m32}) or 64-bit
19978 (@code{-m64}) compilation option.
19980 @item compilation options recorded in the target
19981 @value{NGCC} (since version 4.7) stores the options used for compilation
19982 into @code{DW_AT_producer} part of DWARF debugging information according
19983 to the @value{NGCC} option @code{-grecord-gcc-switches}. One has to
19984 explicitly specify @code{-g} during inferior compilation otherwise
19985 @value{NGCC} produces no DWARF. This feature is only relevant for
19986 platforms where @code{-g} produces DWARF by default, otherwise one may
19987 try to enforce DWARF by using @code{-gdwarf-4}.
19989 @item compilation options set by @code{set compile-args}
19993 You can override compilation options using the following command:
19996 @item set compile-args
19997 @cindex compile command options override
19998 Set compilation options used for compiling and injecting code with the
19999 @code{compile} commands. These options override any conflicting ones
20000 from the target architecture and/or options stored during inferior
20003 @item show compile-args
20004 Displays the current state of compilation options override.
20005 This does not show all the options actually used during compilation,
20006 use @ref{set debug compile} for that.
20009 @subsection Caveats when using the @code{compile} command
20011 There are a few caveats to keep in mind when using the @code{compile}
20012 command. As the caveats are different per language, the table below
20013 highlights specific issues on a per language basis.
20016 @item C code examples and caveats
20017 When the language in @value{GDBN} is set to @samp{C}, the compiler will
20018 attempt to compile the source code with a @samp{C} compiler. The source
20019 code provided to the @code{compile} command will have much the same
20020 access to variables and types as it normally would if it were part of
20021 the program currently being debugged in @value{GDBN}.
20023 Below is a sample program that forms the basis of the examples that
20024 follow. This program has been compiled and loaded into @value{GDBN},
20025 much like any other normal debugging session.
20028 void function1 (void)
20031 printf ("function 1\n");
20034 void function2 (void)
20049 For the purposes of the examples in this section, the program above has
20050 been compiled, loaded into @value{GDBN}, stopped at the function
20051 @code{main}, and @value{GDBN} is awaiting input from the user.
20053 To access variables and types for any program in @value{GDBN}, the
20054 program must be compiled and packaged with debug information. The
20055 @code{compile} command is not an exception to this rule. Without debug
20056 information, you can still use the @code{compile} command, but you will
20057 be very limited in what variables and types you can access.
20059 So with that in mind, the example above has been compiled with debug
20060 information enabled. The @code{compile} command will have access to
20061 all variables and types (except those that may have been optimized
20062 out). Currently, as @value{GDBN} has stopped the program in the
20063 @code{main} function, the @code{compile} command would have access to
20064 the variable @code{k}. You could invoke the @code{compile} command
20065 and type some source code to set the value of @code{k}. You can also
20066 read it, or do anything with that variable you would normally do in
20067 @code{C}. Be aware that changes to inferior variables in the
20068 @code{compile} command are persistent. In the following example:
20071 compile code k = 3;
20075 the variable @code{k} is now 3. It will retain that value until
20076 something else in the example program changes it, or another
20077 @code{compile} command changes it.
20079 Normal scope and access rules apply to source code compiled and
20080 injected by the @code{compile} command. In the example, the variables
20081 @code{j} and @code{k} are not accessible yet, because the program is
20082 currently stopped in the @code{main} function, where these variables
20083 are not in scope. Therefore, the following command
20086 compile code j = 3;
20090 will result in a compilation error message.
20092 Once the program is continued, execution will bring these variables in
20093 scope, and they will become accessible; then the code you specify via
20094 the @code{compile} command will be able to access them.
20096 You can create variables and types with the @code{compile} command as
20097 part of your source code. Variables and types that are created as part
20098 of the @code{compile} command are not visible to the rest of the program for
20099 the duration of its run. This example is valid:
20102 compile code int ff = 5; printf ("ff is %d\n", ff);
20105 However, if you were to type the following into @value{GDBN} after that
20106 command has completed:
20109 compile code printf ("ff is %d\n'', ff);
20113 a compiler error would be raised as the variable @code{ff} no longer
20114 exists. Object code generated and injected by the @code{compile}
20115 command is removed when its execution ends. Caution is advised
20116 when assigning to program variables values of variables created by the
20117 code submitted to the @code{compile} command. This example is valid:
20120 compile code int ff = 5; k = ff;
20123 The value of the variable @code{ff} is assigned to @code{k}. The variable
20124 @code{k} does not require the existence of @code{ff} to maintain the value
20125 it has been assigned. However, pointers require particular care in
20126 assignment. If the source code compiled with the @code{compile} command
20127 changed the address of a pointer in the example program, perhaps to a
20128 variable created in the @code{compile} command, that pointer would point
20129 to an invalid location when the command exits. The following example
20130 would likely cause issues with your debugged program:
20133 compile code int ff = 5; p = &ff;
20136 In this example, @code{p} would point to @code{ff} when the
20137 @code{compile} command is executing the source code provided to it.
20138 However, as variables in the (example) program persist with their
20139 assigned values, the variable @code{p} would point to an invalid
20140 location when the command exists. A general rule should be followed
20141 in that you should either assign @code{NULL} to any assigned pointers,
20142 or restore a valid location to the pointer before the command exits.
20144 Similar caution must be exercised with any structs, unions, and typedefs
20145 defined in @code{compile} command. Types defined in the @code{compile}
20146 command will no longer be available in the next @code{compile} command.
20147 Therefore, if you cast a variable to a type defined in the
20148 @code{compile} command, care must be taken to ensure that any future
20149 need to resolve the type can be achieved.
20152 (@value{GDBP}) compile code static struct a @{ int a; @} v = @{ 42 @}; argv = &v;
20153 (@value{GDBP}) compile code printf ("%d\n", ((struct a *) argv)->a);
20154 gdb command line:1:36: error: dereferencing pointer to incomplete type ‘struct a’
20155 Compilation failed.
20156 (@value{GDBP}) compile code struct a @{ int a; @}; printf ("%d\n", ((struct a *) argv)->a);
20160 Variables that have been optimized away by the compiler are not
20161 accessible to the code submitted to the @code{compile} command.
20162 Access to those variables will generate a compiler error which @value{GDBN}
20163 will print to the console.
20166 @subsection Compiler search for the @code{compile} command
20168 @value{GDBN} needs to find @value{NGCC} for the inferior being debugged
20169 which may not be obvious for remote targets of different architecture
20170 than where @value{GDBN} is running. Environment variable @code{PATH} on
20171 @value{GDBN} host is searched for @value{NGCC} binary matching the
20172 target architecture and operating system. This search can be overriden
20173 by @code{set compile-gcc} @value{GDBN} command below. @code{PATH} is
20174 taken from shell that executed @value{GDBN}, it is not the value set by
20175 @value{GDBN} command @code{set environment}). @xref{Environment}.
20178 Specifically @code{PATH} is searched for binaries matching regular expression
20179 @code{@var{arch}(-[^-]*)?-@var{os}-gcc} according to the inferior target being
20180 debugged. @var{arch} is processor name --- multiarch is supported, so for
20181 example both @code{i386} and @code{x86_64} targets look for pattern
20182 @code{(x86_64|i.86)} and both @code{s390} and @code{s390x} targets look
20183 for pattern @code{s390x?}. @var{os} is currently supported only for
20184 pattern @code{linux(-gnu)?}.
20186 On Posix hosts the compiler driver @value{GDBN} needs to find also
20187 shared library @file{libcc1.so} from the compiler. It is searched in
20188 default shared library search path (overridable with usual environment
20189 variable @code{LD_LIBRARY_PATH}), unrelated to @code{PATH} or @code{set
20190 compile-gcc} settings. Contrary to it @file{libcc1plugin.so} is found
20191 according to the installation of the found compiler --- as possibly
20192 specified by the @code{set compile-gcc} command.
20195 @item set compile-gcc
20196 @cindex compile command driver filename override
20197 Set compilation command used for compiling and injecting code with the
20198 @code{compile} commands. If this option is not set (it is set to
20199 an empty string), the search described above will occur --- that is the
20202 @item show compile-gcc
20203 Displays the current compile command @value{NGCC} driver filename.
20204 If set, it is the main command @command{gcc}, found usually for example
20205 under name @file{x86_64-linux-gnu-gcc}.
20209 @chapter @value{GDBN} Files
20211 @value{GDBN} needs to know the file name of the program to be debugged,
20212 both in order to read its symbol table and in order to start your
20213 program. To debug a core dump of a previous run, you must also tell
20214 @value{GDBN} the name of the core dump file.
20217 * Files:: Commands to specify files
20218 * File Caching:: Information about @value{GDBN}'s file caching
20219 * Separate Debug Files:: Debugging information in separate files
20220 * MiniDebugInfo:: Debugging information in a special section
20221 * Index Files:: Index files speed up @value{GDBN}
20222 * Symbol Errors:: Errors reading symbol files
20223 * Data Files:: @value{GDBN} data files
20227 @section Commands to Specify Files
20229 @cindex symbol table
20230 @cindex core dump file
20232 You may want to specify executable and core dump file names. The usual
20233 way to do this is at start-up time, using the arguments to
20234 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
20235 Out of @value{GDBN}}).
20237 Occasionally it is necessary to change to a different file during a
20238 @value{GDBN} session. Or you may run @value{GDBN} and forget to
20239 specify a file you want to use. Or you are debugging a remote target
20240 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
20241 Program}). In these situations the @value{GDBN} commands to specify
20242 new files are useful.
20245 @cindex executable file
20247 @item file @var{filename}
20248 Use @var{filename} as the program to be debugged. It is read for its
20249 symbols and for the contents of pure memory. It is also the program
20250 executed when you use the @code{run} command. If you do not specify a
20251 directory and the file is not found in the @value{GDBN} working directory,
20252 @value{GDBN} uses the environment variable @code{PATH} as a list of
20253 directories to search, just as the shell does when looking for a program
20254 to run. You can change the value of this variable, for both @value{GDBN}
20255 and your program, using the @code{path} command.
20257 @cindex unlinked object files
20258 @cindex patching object files
20259 You can load unlinked object @file{.o} files into @value{GDBN} using
20260 the @code{file} command. You will not be able to ``run'' an object
20261 file, but you can disassemble functions and inspect variables. Also,
20262 if the underlying BFD functionality supports it, you could use
20263 @kbd{gdb -write} to patch object files using this technique. Note
20264 that @value{GDBN} can neither interpret nor modify relocations in this
20265 case, so branches and some initialized variables will appear to go to
20266 the wrong place. But this feature is still handy from time to time.
20269 @code{file} with no argument makes @value{GDBN} discard any information it
20270 has on both executable file and the symbol table.
20273 @item exec-file @r{[} @var{filename} @r{]}
20274 Specify that the program to be run (but not the symbol table) is found
20275 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
20276 if necessary to locate your program. Omitting @var{filename} means to
20277 discard information on the executable file.
20279 @kindex symbol-file
20280 @item symbol-file @r{[} @var{filename} @r{[} -o @var{offset} @r{]]}
20281 Read symbol table information from file @var{filename}. @code{PATH} is
20282 searched when necessary. Use the @code{file} command to get both symbol
20283 table and program to run from the same file.
20285 If an optional @var{offset} is specified, it is added to the start
20286 address of each section in the symbol file. This is useful if the
20287 program is relocated at runtime, such as the Linux kernel with kASLR
20290 @code{symbol-file} with no argument clears out @value{GDBN} information on your
20291 program's symbol table.
20293 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
20294 some breakpoints and auto-display expressions. This is because they may
20295 contain pointers to the internal data recording symbols and data types,
20296 which are part of the old symbol table data being discarded inside
20299 @code{symbol-file} does not repeat if you press @key{RET} again after
20302 When @value{GDBN} is configured for a particular environment, it
20303 understands debugging information in whatever format is the standard
20304 generated for that environment; you may use either a @sc{gnu} compiler, or
20305 other compilers that adhere to the local conventions.
20306 Best results are usually obtained from @sc{gnu} compilers; for example,
20307 using @code{@value{NGCC}} you can generate debugging information for
20310 For most kinds of object files, with the exception of old SVR3 systems
20311 using COFF, the @code{symbol-file} command does not normally read the
20312 symbol table in full right away. Instead, it scans the symbol table
20313 quickly to find which source files and which symbols are present. The
20314 details are read later, one source file at a time, as they are needed.
20316 The purpose of this two-stage reading strategy is to make @value{GDBN}
20317 start up faster. For the most part, it is invisible except for
20318 occasional pauses while the symbol table details for a particular source
20319 file are being read. (The @code{set verbose} command can turn these
20320 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
20321 Warnings and Messages}.)
20323 We have not implemented the two-stage strategy for COFF yet. When the
20324 symbol table is stored in COFF format, @code{symbol-file} reads the
20325 symbol table data in full right away. Note that ``stabs-in-COFF''
20326 still does the two-stage strategy, since the debug info is actually
20330 @cindex reading symbols immediately
20331 @cindex symbols, reading immediately
20332 @item symbol-file @r{[} -readnow @r{]} @var{filename}
20333 @itemx file @r{[} -readnow @r{]} @var{filename}
20334 You can override the @value{GDBN} two-stage strategy for reading symbol
20335 tables by using the @samp{-readnow} option with any of the commands that
20336 load symbol table information, if you want to be sure @value{GDBN} has the
20337 entire symbol table available.
20339 @cindex @code{-readnever}, option for symbol-file command
20340 @cindex never read symbols
20341 @cindex symbols, never read
20342 @item symbol-file @r{[} -readnever @r{]} @var{filename}
20343 @itemx file @r{[} -readnever @r{]} @var{filename}
20344 You can instruct @value{GDBN} to never read the symbolic information
20345 contained in @var{filename} by using the @samp{-readnever} option.
20346 @xref{--readnever}.
20348 @c FIXME: for now no mention of directories, since this seems to be in
20349 @c flux. 13mar1992 status is that in theory GDB would look either in
20350 @c current dir or in same dir as myprog; but issues like competing
20351 @c GDB's, or clutter in system dirs, mean that in practice right now
20352 @c only current dir is used. FFish says maybe a special GDB hierarchy
20353 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
20357 @item core-file @r{[}@var{filename}@r{]}
20359 Specify the whereabouts of a core dump file to be used as the ``contents
20360 of memory''. Traditionally, core files contain only some parts of the
20361 address space of the process that generated them; @value{GDBN} can access the
20362 executable file itself for other parts.
20364 @code{core-file} with no argument specifies that no core file is
20367 Note that the core file is ignored when your program is actually running
20368 under @value{GDBN}. So, if you have been running your program and you
20369 wish to debug a core file instead, you must kill the subprocess in which
20370 the program is running. To do this, use the @code{kill} command
20371 (@pxref{Kill Process, ,Killing the Child Process}).
20373 @kindex add-symbol-file
20374 @cindex dynamic linking
20375 @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{]}
20376 The @code{add-symbol-file} command reads additional symbol table
20377 information from the file @var{filename}. You would use this command
20378 when @var{filename} has been dynamically loaded (by some other means)
20379 into the program that is running. The @var{textaddress} parameter gives
20380 the memory address at which the file's text section has been loaded.
20381 You can additionally specify the base address of other sections using
20382 an arbitrary number of @samp{-s @var{section} @var{address}} pairs.
20383 If a section is omitted, @value{GDBN} will use its default addresses
20384 as found in @var{filename}. Any @var{address} or @var{textaddress}
20385 can be given as an expression.
20387 If an optional @var{offset} is specified, it is added to the start
20388 address of each section, except those for which the address was
20389 specified explicitly.
20391 The symbol table of the file @var{filename} is added to the symbol table
20392 originally read with the @code{symbol-file} command. You can use the
20393 @code{add-symbol-file} command any number of times; the new symbol data
20394 thus read is kept in addition to the old.
20396 Changes can be reverted using the command @code{remove-symbol-file}.
20398 @cindex relocatable object files, reading symbols from
20399 @cindex object files, relocatable, reading symbols from
20400 @cindex reading symbols from relocatable object files
20401 @cindex symbols, reading from relocatable object files
20402 @cindex @file{.o} files, reading symbols from
20403 Although @var{filename} is typically a shared library file, an
20404 executable file, or some other object file which has been fully
20405 relocated for loading into a process, you can also load symbolic
20406 information from relocatable @file{.o} files, as long as:
20410 the file's symbolic information refers only to linker symbols defined in
20411 that file, not to symbols defined by other object files,
20413 every section the file's symbolic information refers to has actually
20414 been loaded into the inferior, as it appears in the file, and
20416 you can determine the address at which every section was loaded, and
20417 provide these to the @code{add-symbol-file} command.
20421 Some embedded operating systems, like Sun Chorus and VxWorks, can load
20422 relocatable files into an already running program; such systems
20423 typically make the requirements above easy to meet. However, it's
20424 important to recognize that many native systems use complex link
20425 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
20426 assembly, for example) that make the requirements difficult to meet. In
20427 general, one cannot assume that using @code{add-symbol-file} to read a
20428 relocatable object file's symbolic information will have the same effect
20429 as linking the relocatable object file into the program in the normal
20432 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
20434 @kindex remove-symbol-file
20435 @item remove-symbol-file @var{filename}
20436 @item remove-symbol-file -a @var{address}
20437 Remove a symbol file added via the @code{add-symbol-file} command. The
20438 file to remove can be identified by its @var{filename} or by an @var{address}
20439 that lies within the boundaries of this symbol file in memory. Example:
20442 (@value{GDBP}) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
20443 add symbol table from file "/home/user/gdb/mylib.so" at
20444 .text_addr = 0x7ffff7ff9480
20446 Reading symbols from /home/user/gdb/mylib.so...done.
20447 (@value{GDBP}) remove-symbol-file -a 0x7ffff7ff9480
20448 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
20453 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
20455 @kindex add-symbol-file-from-memory
20456 @cindex @code{syscall DSO}
20457 @cindex load symbols from memory
20458 @item add-symbol-file-from-memory @var{address}
20459 Load symbols from the given @var{address} in a dynamically loaded
20460 object file whose image is mapped directly into the inferior's memory.
20461 For example, the Linux kernel maps a @code{syscall DSO} into each
20462 process's address space; this DSO provides kernel-specific code for
20463 some system calls. The argument can be any expression whose
20464 evaluation yields the address of the file's shared object file header.
20465 For this command to work, you must have used @code{symbol-file} or
20466 @code{exec-file} commands in advance.
20469 @item section @var{section} @var{addr}
20470 The @code{section} command changes the base address of the named
20471 @var{section} of the exec file to @var{addr}. This can be used if the
20472 exec file does not contain section addresses, (such as in the
20473 @code{a.out} format), or when the addresses specified in the file
20474 itself are wrong. Each section must be changed separately. The
20475 @code{info files} command, described below, lists all the sections and
20479 @kindex info target
20482 @code{info files} and @code{info target} are synonymous; both print the
20483 current target (@pxref{Targets, ,Specifying a Debugging Target}),
20484 including the names of the executable and core dump files currently in
20485 use by @value{GDBN}, and the files from which symbols were loaded. The
20486 command @code{help target} lists all possible targets rather than
20489 @kindex maint info sections
20490 @item maint info sections
20491 Another command that can give you extra information about program sections
20492 is @code{maint info sections}. In addition to the section information
20493 displayed by @code{info files}, this command displays the flags and file
20494 offset of each section in the executable and core dump files. In addition,
20495 @code{maint info sections} provides the following command options (which
20496 may be arbitrarily combined):
20500 Display sections for all loaded object files, including shared libraries.
20501 @item @var{sections}
20502 Display info only for named @var{sections}.
20503 @item @var{section-flags}
20504 Display info only for sections for which @var{section-flags} are true.
20505 The section flags that @value{GDBN} currently knows about are:
20508 Section will have space allocated in the process when loaded.
20509 Set for all sections except those containing debug information.
20511 Section will be loaded from the file into the child process memory.
20512 Set for pre-initialized code and data, clear for @code{.bss} sections.
20514 Section needs to be relocated before loading.
20516 Section cannot be modified by the child process.
20518 Section contains executable code only.
20520 Section contains data only (no executable code).
20522 Section will reside in ROM.
20524 Section contains data for constructor/destructor lists.
20526 Section is not empty.
20528 An instruction to the linker to not output the section.
20529 @item COFF_SHARED_LIBRARY
20530 A notification to the linker that the section contains
20531 COFF shared library information.
20533 Section contains common symbols.
20536 @kindex set trust-readonly-sections
20537 @cindex read-only sections
20538 @item set trust-readonly-sections on
20539 Tell @value{GDBN} that readonly sections in your object file
20540 really are read-only (i.e.@: that their contents will not change).
20541 In that case, @value{GDBN} can fetch values from these sections
20542 out of the object file, rather than from the target program.
20543 For some targets (notably embedded ones), this can be a significant
20544 enhancement to debugging performance.
20546 The default is off.
20548 @item set trust-readonly-sections off
20549 Tell @value{GDBN} not to trust readonly sections. This means that
20550 the contents of the section might change while the program is running,
20551 and must therefore be fetched from the target when needed.
20553 @item show trust-readonly-sections
20554 Show the current setting of trusting readonly sections.
20557 All file-specifying commands allow both absolute and relative file names
20558 as arguments. @value{GDBN} always converts the file name to an absolute file
20559 name and remembers it that way.
20561 @cindex shared libraries
20562 @anchor{Shared Libraries}
20563 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, SunOS,
20564 Darwin/Mach-O, SVr4, IBM RS/6000 AIX, QNX Neutrino, FDPIC (FR-V), and
20565 DSBT (TIC6X) shared libraries.
20567 On MS-Windows @value{GDBN} must be linked with the Expat library to support
20568 shared libraries. @xref{Expat}.
20570 @value{GDBN} automatically loads symbol definitions from shared libraries
20571 when you use the @code{run} command, or when you examine a core file.
20572 (Before you issue the @code{run} command, @value{GDBN} does not understand
20573 references to a function in a shared library, however---unless you are
20574 debugging a core file).
20576 @c FIXME: some @value{GDBN} release may permit some refs to undef
20577 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
20578 @c FIXME...lib; check this from time to time when updating manual
20580 There are times, however, when you may wish to not automatically load
20581 symbol definitions from shared libraries, such as when they are
20582 particularly large or there are many of them.
20584 To control the automatic loading of shared library symbols, use the
20588 @kindex set auto-solib-add
20589 @item set auto-solib-add @var{mode}
20590 If @var{mode} is @code{on}, symbols from all shared object libraries
20591 will be loaded automatically when the inferior begins execution, you
20592 attach to an independently started inferior, or when the dynamic linker
20593 informs @value{GDBN} that a new library has been loaded. If @var{mode}
20594 is @code{off}, symbols must be loaded manually, using the
20595 @code{sharedlibrary} command. The default value is @code{on}.
20597 @cindex memory used for symbol tables
20598 If your program uses lots of shared libraries with debug info that
20599 takes large amounts of memory, you can decrease the @value{GDBN}
20600 memory footprint by preventing it from automatically loading the
20601 symbols from shared libraries. To that end, type @kbd{set
20602 auto-solib-add off} before running the inferior, then load each
20603 library whose debug symbols you do need with @kbd{sharedlibrary
20604 @var{regexp}}, where @var{regexp} is a regular expression that matches
20605 the libraries whose symbols you want to be loaded.
20607 @kindex show auto-solib-add
20608 @item show auto-solib-add
20609 Display the current autoloading mode.
20612 @cindex load shared library
20613 To explicitly load shared library symbols, use the @code{sharedlibrary}
20617 @kindex info sharedlibrary
20619 @item info share @var{regex}
20620 @itemx info sharedlibrary @var{regex}
20621 Print the names of the shared libraries which are currently loaded
20622 that match @var{regex}. If @var{regex} is omitted then print
20623 all shared libraries that are loaded.
20626 @item info dll @var{regex}
20627 This is an alias of @code{info sharedlibrary}.
20629 @kindex sharedlibrary
20631 @item sharedlibrary @var{regex}
20632 @itemx share @var{regex}
20633 Load shared object library symbols for files matching a
20634 Unix regular expression.
20635 As with files loaded automatically, it only loads shared libraries
20636 required by your program for a core file or after typing @code{run}. If
20637 @var{regex} is omitted all shared libraries required by your program are
20640 @item nosharedlibrary
20641 @kindex nosharedlibrary
20642 @cindex unload symbols from shared libraries
20643 Unload all shared object library symbols. This discards all symbols
20644 that have been loaded from all shared libraries. Symbols from shared
20645 libraries that were loaded by explicit user requests are not
20649 Sometimes you may wish that @value{GDBN} stops and gives you control
20650 when any of shared library events happen. The best way to do this is
20651 to use @code{catch load} and @code{catch unload} (@pxref{Set
20654 @value{GDBN} also supports the the @code{set stop-on-solib-events}
20655 command for this. This command exists for historical reasons. It is
20656 less useful than setting a catchpoint, because it does not allow for
20657 conditions or commands as a catchpoint does.
20660 @item set stop-on-solib-events
20661 @kindex set stop-on-solib-events
20662 This command controls whether @value{GDBN} should give you control
20663 when the dynamic linker notifies it about some shared library event.
20664 The most common event of interest is loading or unloading of a new
20667 @item show stop-on-solib-events
20668 @kindex show stop-on-solib-events
20669 Show whether @value{GDBN} stops and gives you control when shared
20670 library events happen.
20673 Shared libraries are also supported in many cross or remote debugging
20674 configurations. @value{GDBN} needs to have access to the target's libraries;
20675 this can be accomplished either by providing copies of the libraries
20676 on the host system, or by asking @value{GDBN} to automatically retrieve the
20677 libraries from the target. If copies of the target libraries are
20678 provided, they need to be the same as the target libraries, although the
20679 copies on the target can be stripped as long as the copies on the host are
20682 @cindex where to look for shared libraries
20683 For remote debugging, you need to tell @value{GDBN} where the target
20684 libraries are, so that it can load the correct copies---otherwise, it
20685 may try to load the host's libraries. @value{GDBN} has two variables
20686 to specify the search directories for target libraries.
20689 @cindex prefix for executable and shared library file names
20690 @cindex system root, alternate
20691 @kindex set solib-absolute-prefix
20692 @kindex set sysroot
20693 @item set sysroot @var{path}
20694 Use @var{path} as the system root for the program being debugged. Any
20695 absolute shared library paths will be prefixed with @var{path}; many
20696 runtime loaders store the absolute paths to the shared library in the
20697 target program's memory. When starting processes remotely, and when
20698 attaching to already-running processes (local or remote), their
20699 executable filenames will be prefixed with @var{path} if reported to
20700 @value{GDBN} as absolute by the operating system. If you use
20701 @code{set sysroot} to find executables and shared libraries, they need
20702 to be laid out in the same way that they are on the target, with
20703 e.g.@: a @file{/bin}, @file{/lib} and @file{/usr/lib} hierarchy under
20706 If @var{path} starts with the sequence @file{target:} and the target
20707 system is remote then @value{GDBN} will retrieve the target binaries
20708 from the remote system. This is only supported when using a remote
20709 target that supports the @code{remote get} command (@pxref{File
20710 Transfer,,Sending files to a remote system}). The part of @var{path}
20711 following the initial @file{target:} (if present) is used as system
20712 root prefix on the remote file system. If @var{path} starts with the
20713 sequence @file{remote:} this is converted to the sequence
20714 @file{target:} by @code{set sysroot}@footnote{Historically the
20715 functionality to retrieve binaries from the remote system was
20716 provided by prefixing @var{path} with @file{remote:}}. If you want
20717 to specify a local system root using a directory that happens to be
20718 named @file{target:} or @file{remote:}, you need to use some
20719 equivalent variant of the name like @file{./target:}.
20721 For targets with an MS-DOS based filesystem, such as MS-Windows and
20722 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
20723 absolute file name with @var{path}. But first, on Unix hosts,
20724 @value{GDBN} converts all backslash directory separators into forward
20725 slashes, because the backslash is not a directory separator on Unix:
20728 c:\foo\bar.dll @result{} c:/foo/bar.dll
20731 Then, @value{GDBN} attempts prefixing the target file name with
20732 @var{path}, and looks for the resulting file name in the host file
20736 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
20739 If that does not find the binary, @value{GDBN} tries removing
20740 the @samp{:} character from the drive spec, both for convenience, and,
20741 for the case of the host file system not supporting file names with
20745 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
20748 This makes it possible to have a system root that mirrors a target
20749 with more than one drive. E.g., you may want to setup your local
20750 copies of the target system shared libraries like so (note @samp{c} vs
20754 @file{/path/to/sysroot/c/sys/bin/foo.dll}
20755 @file{/path/to/sysroot/c/sys/bin/bar.dll}
20756 @file{/path/to/sysroot/z/sys/bin/bar.dll}
20760 and point the system root at @file{/path/to/sysroot}, so that
20761 @value{GDBN} can find the correct copies of both
20762 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
20764 If that still does not find the binary, @value{GDBN} tries
20765 removing the whole drive spec from the target file name:
20768 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
20771 This last lookup makes it possible to not care about the drive name,
20772 if you don't want or need to.
20774 The @code{set solib-absolute-prefix} command is an alias for @code{set
20777 @cindex default system root
20778 @cindex @samp{--with-sysroot}
20779 You can set the default system root by using the configure-time
20780 @samp{--with-sysroot} option. If the system root is inside
20781 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
20782 @samp{--exec-prefix}), then the default system root will be updated
20783 automatically if the installed @value{GDBN} is moved to a new
20786 @kindex show sysroot
20788 Display the current executable and shared library prefix.
20790 @kindex set solib-search-path
20791 @item set solib-search-path @var{path}
20792 If this variable is set, @var{path} is a colon-separated list of
20793 directories to search for shared libraries. @samp{solib-search-path}
20794 is used after @samp{sysroot} fails to locate the library, or if the
20795 path to the library is relative instead of absolute. If you want to
20796 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
20797 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
20798 finding your host's libraries. @samp{sysroot} is preferred; setting
20799 it to a nonexistent directory may interfere with automatic loading
20800 of shared library symbols.
20802 @kindex show solib-search-path
20803 @item show solib-search-path
20804 Display the current shared library search path.
20806 @cindex DOS file-name semantics of file names.
20807 @kindex set target-file-system-kind (unix|dos-based|auto)
20808 @kindex show target-file-system-kind
20809 @item set target-file-system-kind @var{kind}
20810 Set assumed file system kind for target reported file names.
20812 Shared library file names as reported by the target system may not
20813 make sense as is on the system @value{GDBN} is running on. For
20814 example, when remote debugging a target that has MS-DOS based file
20815 system semantics, from a Unix host, the target may be reporting to
20816 @value{GDBN} a list of loaded shared libraries with file names such as
20817 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
20818 drive letters, so the @samp{c:\} prefix is not normally understood as
20819 indicating an absolute file name, and neither is the backslash
20820 normally considered a directory separator character. In that case,
20821 the native file system would interpret this whole absolute file name
20822 as a relative file name with no directory components. This would make
20823 it impossible to point @value{GDBN} at a copy of the remote target's
20824 shared libraries on the host using @code{set sysroot}, and impractical
20825 with @code{set solib-search-path}. Setting
20826 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
20827 to interpret such file names similarly to how the target would, and to
20828 map them to file names valid on @value{GDBN}'s native file system
20829 semantics. The value of @var{kind} can be @code{"auto"}, in addition
20830 to one of the supported file system kinds. In that case, @value{GDBN}
20831 tries to determine the appropriate file system variant based on the
20832 current target's operating system (@pxref{ABI, ,Configuring the
20833 Current ABI}). The supported file system settings are:
20837 Instruct @value{GDBN} to assume the target file system is of Unix
20838 kind. Only file names starting the forward slash (@samp{/}) character
20839 are considered absolute, and the directory separator character is also
20843 Instruct @value{GDBN} to assume the target file system is DOS based.
20844 File names starting with either a forward slash, or a drive letter
20845 followed by a colon (e.g., @samp{c:}), are considered absolute, and
20846 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
20847 considered directory separators.
20850 Instruct @value{GDBN} to use the file system kind associated with the
20851 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
20852 This is the default.
20856 @cindex file name canonicalization
20857 @cindex base name differences
20858 When processing file names provided by the user, @value{GDBN}
20859 frequently needs to compare them to the file names recorded in the
20860 program's debug info. Normally, @value{GDBN} compares just the
20861 @dfn{base names} of the files as strings, which is reasonably fast
20862 even for very large programs. (The base name of a file is the last
20863 portion of its name, after stripping all the leading directories.)
20864 This shortcut in comparison is based upon the assumption that files
20865 cannot have more than one base name. This is usually true, but
20866 references to files that use symlinks or similar filesystem
20867 facilities violate that assumption. If your program records files
20868 using such facilities, or if you provide file names to @value{GDBN}
20869 using symlinks etc., you can set @code{basenames-may-differ} to
20870 @code{true} to instruct @value{GDBN} to completely canonicalize each
20871 pair of file names it needs to compare. This will make file-name
20872 comparisons accurate, but at a price of a significant slowdown.
20875 @item set basenames-may-differ
20876 @kindex set basenames-may-differ
20877 Set whether a source file may have multiple base names.
20879 @item show basenames-may-differ
20880 @kindex show basenames-may-differ
20881 Show whether a source file may have multiple base names.
20885 @section File Caching
20886 @cindex caching of opened files
20887 @cindex caching of bfd objects
20889 To speed up file loading, and reduce memory usage, @value{GDBN} will
20890 reuse the @code{bfd} objects used to track open files. @xref{Top, ,
20891 BFD, bfd, The Binary File Descriptor Library}. The following commands
20892 allow visibility and control of the caching behavior.
20895 @kindex maint info bfds
20896 @item maint info bfds
20897 This prints information about each @code{bfd} object that is known to
20900 @kindex maint set bfd-sharing
20901 @kindex maint show bfd-sharing
20902 @kindex bfd caching
20903 @item maint set bfd-sharing
20904 @item maint show bfd-sharing
20905 Control whether @code{bfd} objects can be shared. When sharing is
20906 enabled @value{GDBN} reuses already open @code{bfd} objects rather
20907 than reopening the same file. Turning sharing off does not cause
20908 already shared @code{bfd} objects to be unshared, but all future files
20909 that are opened will create a new @code{bfd} object. Similarly,
20910 re-enabling sharing does not cause multiple existing @code{bfd}
20911 objects to be collapsed into a single shared @code{bfd} object.
20913 @kindex set debug bfd-cache @var{level}
20914 @kindex bfd caching
20915 @item set debug bfd-cache @var{level}
20916 Turns on debugging of the bfd cache, setting the level to @var{level}.
20918 @kindex show debug bfd-cache
20919 @kindex bfd caching
20920 @item show debug bfd-cache
20921 Show the current debugging level of the bfd cache.
20924 @node Separate Debug Files
20925 @section Debugging Information in Separate Files
20926 @cindex separate debugging information files
20927 @cindex debugging information in separate files
20928 @cindex @file{.debug} subdirectories
20929 @cindex debugging information directory, global
20930 @cindex global debugging information directories
20931 @cindex build ID, and separate debugging files
20932 @cindex @file{.build-id} directory
20934 @value{GDBN} allows you to put a program's debugging information in a
20935 file separate from the executable itself, in a way that allows
20936 @value{GDBN} to find and load the debugging information automatically.
20937 Since debugging information can be very large---sometimes larger
20938 than the executable code itself---some systems distribute debugging
20939 information for their executables in separate files, which users can
20940 install only when they need to debug a problem.
20942 @value{GDBN} supports two ways of specifying the separate debug info
20947 The executable contains a @dfn{debug link} that specifies the name of
20948 the separate debug info file. The separate debug file's name is
20949 usually @file{@var{executable}.debug}, where @var{executable} is the
20950 name of the corresponding executable file without leading directories
20951 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
20952 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
20953 checksum for the debug file, which @value{GDBN} uses to validate that
20954 the executable and the debug file came from the same build.
20957 The executable contains a @dfn{build ID}, a unique bit string that is
20958 also present in the corresponding debug info file. (This is supported
20959 only on some operating systems, when using the ELF or PE file formats
20960 for binary files and the @sc{gnu} Binutils.) For more details about
20961 this feature, see the description of the @option{--build-id}
20962 command-line option in @ref{Options, , Command Line Options, ld,
20963 The GNU Linker}. The debug info file's name is not specified
20964 explicitly by the build ID, but can be computed from the build ID, see
20968 Depending on the way the debug info file is specified, @value{GDBN}
20969 uses two different methods of looking for the debug file:
20973 For the ``debug link'' method, @value{GDBN} looks up the named file in
20974 the directory of the executable file, then in a subdirectory of that
20975 directory named @file{.debug}, and finally under each one of the
20976 global debug directories, in a subdirectory whose name is identical to
20977 the leading directories of the executable's absolute file name. (On
20978 MS-Windows/MS-DOS, the drive letter of the executable's leading
20979 directories is converted to a one-letter subdirectory, i.e.@:
20980 @file{d:/usr/bin/} is converted to @file{/d/usr/bin/}, because Windows
20981 filesystems disallow colons in file names.)
20984 For the ``build ID'' method, @value{GDBN} looks in the
20985 @file{.build-id} subdirectory of each one of the global debug directories for
20986 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
20987 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
20988 are the rest of the bit string. (Real build ID strings are 32 or more
20989 hex characters, not 10.)
20992 So, for example, suppose you ask @value{GDBN} to debug
20993 @file{/usr/bin/ls}, which has a debug link that specifies the
20994 file @file{ls.debug}, and a build ID whose value in hex is
20995 @code{abcdef1234}. If the list of the global debug directories includes
20996 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
20997 debug information files, in the indicated order:
21001 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
21003 @file{/usr/bin/ls.debug}
21005 @file{/usr/bin/.debug/ls.debug}
21007 @file{/usr/lib/debug/usr/bin/ls.debug}.
21010 @anchor{debug-file-directory}
21011 Global debugging info directories default to what is set by @value{GDBN}
21012 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
21013 you can also set the global debugging info directories, and view the list
21014 @value{GDBN} is currently using.
21018 @kindex set debug-file-directory
21019 @item set debug-file-directory @var{directories}
21020 Set the directories which @value{GDBN} searches for separate debugging
21021 information files to @var{directory}. Multiple path components can be set
21022 concatenating them by a path separator.
21024 @kindex show debug-file-directory
21025 @item show debug-file-directory
21026 Show the directories @value{GDBN} searches for separate debugging
21031 @cindex @code{.gnu_debuglink} sections
21032 @cindex debug link sections
21033 A debug link is a special section of the executable file named
21034 @code{.gnu_debuglink}. The section must contain:
21038 A filename, with any leading directory components removed, followed by
21041 zero to three bytes of padding, as needed to reach the next four-byte
21042 boundary within the section, and
21044 a four-byte CRC checksum, stored in the same endianness used for the
21045 executable file itself. The checksum is computed on the debugging
21046 information file's full contents by the function given below, passing
21047 zero as the @var{crc} argument.
21050 Any executable file format can carry a debug link, as long as it can
21051 contain a section named @code{.gnu_debuglink} with the contents
21054 @cindex @code{.note.gnu.build-id} sections
21055 @cindex build ID sections
21056 The build ID is a special section in the executable file (and in other
21057 ELF binary files that @value{GDBN} may consider). This section is
21058 often named @code{.note.gnu.build-id}, but that name is not mandatory.
21059 It contains unique identification for the built files---the ID remains
21060 the same across multiple builds of the same build tree. The default
21061 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
21062 content for the build ID string. The same section with an identical
21063 value is present in the original built binary with symbols, in its
21064 stripped variant, and in the separate debugging information file.
21066 The debugging information file itself should be an ordinary
21067 executable, containing a full set of linker symbols, sections, and
21068 debugging information. The sections of the debugging information file
21069 should have the same names, addresses, and sizes as the original file,
21070 but they need not contain any data---much like a @code{.bss} section
21071 in an ordinary executable.
21073 The @sc{gnu} binary utilities (Binutils) package includes the
21074 @samp{objcopy} utility that can produce
21075 the separated executable / debugging information file pairs using the
21076 following commands:
21079 @kbd{objcopy --only-keep-debug foo foo.debug}
21084 These commands remove the debugging
21085 information from the executable file @file{foo} and place it in the file
21086 @file{foo.debug}. You can use the first, second or both methods to link the
21091 The debug link method needs the following additional command to also leave
21092 behind a debug link in @file{foo}:
21095 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
21098 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
21099 a version of the @code{strip} command such that the command @kbd{strip foo -f
21100 foo.debug} has the same functionality as the two @code{objcopy} commands and
21101 the @code{ln -s} command above, together.
21104 Build ID gets embedded into the main executable using @code{ld --build-id} or
21105 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
21106 compatibility fixes for debug files separation are present in @sc{gnu} binary
21107 utilities (Binutils) package since version 2.18.
21112 @cindex CRC algorithm definition
21113 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
21114 IEEE 802.3 using the polynomial:
21116 @c TexInfo requires naked braces for multi-digit exponents for Tex
21117 @c output, but this causes HTML output to barf. HTML has to be set using
21118 @c raw commands. So we end up having to specify this equation in 2
21123 <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>
21124 + <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
21130 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
21131 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
21135 The function is computed byte at a time, taking the least
21136 significant bit of each byte first. The initial pattern
21137 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
21138 the final result is inverted to ensure trailing zeros also affect the
21141 @emph{Note:} This is the same CRC polynomial as used in handling the
21142 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{qCRC packet}).
21143 However in the case of the Remote Serial Protocol, the CRC is computed
21144 @emph{most} significant bit first, and the result is not inverted, so
21145 trailing zeros have no effect on the CRC value.
21147 To complete the description, we show below the code of the function
21148 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
21149 initially supplied @code{crc} argument means that an initial call to
21150 this function passing in zero will start computing the CRC using
21153 @kindex gnu_debuglink_crc32
21156 gnu_debuglink_crc32 (unsigned long crc,
21157 unsigned char *buf, size_t len)
21159 static const unsigned long crc32_table[256] =
21161 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
21162 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
21163 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
21164 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
21165 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
21166 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
21167 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
21168 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
21169 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
21170 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
21171 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
21172 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
21173 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
21174 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
21175 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
21176 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
21177 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
21178 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
21179 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
21180 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
21181 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
21182 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
21183 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
21184 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
21185 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
21186 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
21187 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
21188 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
21189 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
21190 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
21191 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
21192 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
21193 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
21194 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
21195 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
21196 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
21197 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
21198 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
21199 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
21200 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
21201 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
21202 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
21203 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
21204 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
21205 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
21206 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
21207 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
21208 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
21209 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
21210 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
21211 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
21214 unsigned char *end;
21216 crc = ~crc & 0xffffffff;
21217 for (end = buf + len; buf < end; ++buf)
21218 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
21219 return ~crc & 0xffffffff;
21224 This computation does not apply to the ``build ID'' method.
21226 @node MiniDebugInfo
21227 @section Debugging information in a special section
21228 @cindex separate debug sections
21229 @cindex @samp{.gnu_debugdata} section
21231 Some systems ship pre-built executables and libraries that have a
21232 special @samp{.gnu_debugdata} section. This feature is called
21233 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
21234 is used to supply extra symbols for backtraces.
21236 The intent of this section is to provide extra minimal debugging
21237 information for use in simple backtraces. It is not intended to be a
21238 replacement for full separate debugging information (@pxref{Separate
21239 Debug Files}). The example below shows the intended use; however,
21240 @value{GDBN} does not currently put restrictions on what sort of
21241 debugging information might be included in the section.
21243 @value{GDBN} has support for this extension. If the section exists,
21244 then it is used provided that no other source of debugging information
21245 can be found, and that @value{GDBN} was configured with LZMA support.
21247 This section can be easily created using @command{objcopy} and other
21248 standard utilities:
21251 # Extract the dynamic symbols from the main binary, there is no need
21252 # to also have these in the normal symbol table.
21253 nm -D @var{binary} --format=posix --defined-only \
21254 | awk '@{ print $1 @}' | sort > dynsyms
21256 # Extract all the text (i.e. function) symbols from the debuginfo.
21257 # (Note that we actually also accept "D" symbols, for the benefit
21258 # of platforms like PowerPC64 that use function descriptors.)
21259 nm @var{binary} --format=posix --defined-only \
21260 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
21263 # Keep all the function symbols not already in the dynamic symbol
21265 comm -13 dynsyms funcsyms > keep_symbols
21267 # Separate full debug info into debug binary.
21268 objcopy --only-keep-debug @var{binary} debug
21270 # Copy the full debuginfo, keeping only a minimal set of symbols and
21271 # removing some unnecessary sections.
21272 objcopy -S --remove-section .gdb_index --remove-section .comment \
21273 --keep-symbols=keep_symbols debug mini_debuginfo
21275 # Drop the full debug info from the original binary.
21276 strip --strip-all -R .comment @var{binary}
21278 # Inject the compressed data into the .gnu_debugdata section of the
21281 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
21285 @section Index Files Speed Up @value{GDBN}
21286 @cindex index files
21287 @cindex @samp{.gdb_index} section
21289 When @value{GDBN} finds a symbol file, it scans the symbols in the
21290 file in order to construct an internal symbol table. This lets most
21291 @value{GDBN} operations work quickly---at the cost of a delay early
21292 on. For large programs, this delay can be quite lengthy, so
21293 @value{GDBN} provides a way to build an index, which speeds up
21296 For convenience, @value{GDBN} comes with a program,
21297 @command{gdb-add-index}, which can be used to add the index to a
21298 symbol file. It takes the symbol file as its only argument:
21301 $ gdb-add-index symfile
21304 @xref{gdb-add-index}.
21306 It is also possible to do the work manually. Here is what
21307 @command{gdb-add-index} does behind the curtains.
21309 The index is stored as a section in the symbol file. @value{GDBN} can
21310 write the index to a file, then you can put it into the symbol file
21311 using @command{objcopy}.
21313 To create an index file, use the @code{save gdb-index} command:
21316 @item save gdb-index [-dwarf-5] @var{directory}
21317 @kindex save gdb-index
21318 Create index files for all symbol files currently known by
21319 @value{GDBN}. For each known @var{symbol-file}, this command by
21320 default creates it produces a single file
21321 @file{@var{symbol-file}.gdb-index}. If you invoke this command with
21322 the @option{-dwarf-5} option, it produces 2 files:
21323 @file{@var{symbol-file}.debug_names} and
21324 @file{@var{symbol-file}.debug_str}. The files are created in the
21325 given @var{directory}.
21328 Once you have created an index file you can merge it into your symbol
21329 file, here named @file{symfile}, using @command{objcopy}:
21332 $ objcopy --add-section .gdb_index=symfile.gdb-index \
21333 --set-section-flags .gdb_index=readonly symfile symfile
21336 Or for @code{-dwarf-5}:
21339 $ objcopy --dump-section .debug_str=symfile.debug_str.new symfile
21340 $ cat symfile.debug_str >>symfile.debug_str.new
21341 $ objcopy --add-section .debug_names=symfile.gdb-index \
21342 --set-section-flags .debug_names=readonly \
21343 --update-section .debug_str=symfile.debug_str.new symfile symfile
21346 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
21347 sections that have been deprecated. Usually they are deprecated because
21348 they are missing a new feature or have performance issues.
21349 To tell @value{GDBN} to use a deprecated index section anyway
21350 specify @code{set use-deprecated-index-sections on}.
21351 The default is @code{off}.
21352 This can speed up startup, but may result in some functionality being lost.
21353 @xref{Index Section Format}.
21355 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
21356 must be done before gdb reads the file. The following will not work:
21359 $ gdb -ex "set use-deprecated-index-sections on" <program>
21362 Instead you must do, for example,
21365 $ gdb -iex "set use-deprecated-index-sections on" <program>
21368 There are currently some limitation on indices. They only work when
21369 using DWARF debugging information, not stabs. And, only the
21370 @code{-dwarf-5} index works for programs using Ada.
21372 @subsection Automatic symbol index cache
21374 @cindex automatic symbol index cache
21375 It is possible for @value{GDBN} to automatically save a copy of this index in a
21376 cache on disk and retrieve it from there when loading the same binary in the
21377 future. This feature can be turned on with @kbd{set index-cache on}. The
21378 following commands can be used to tweak the behavior of the index cache.
21382 @kindex set index-cache
21383 @item set index-cache on
21384 @itemx set index-cache off
21385 Enable or disable the use of the symbol index cache.
21387 @item set index-cache directory @var{directory}
21388 @kindex show index-cache
21389 @itemx show index-cache directory
21390 Set/show the directory where index files will be saved.
21392 The default value for this directory depends on the host platform. On
21393 most systems, the index is cached in the @file{gdb} subdirectory of
21394 the directory pointed to by the @env{XDG_CACHE_HOME} environment
21395 variable, if it is defined, else in the @file{.cache/gdb} subdirectory
21396 of your home directory. However, on some systems, the default may
21397 differ according to local convention.
21399 There is no limit on the disk space used by index cache. It is perfectly safe
21400 to delete the content of that directory to free up disk space.
21402 @item show index-cache stats
21403 Print the number of cache hits and misses since the launch of @value{GDBN}.
21407 @node Symbol Errors
21408 @section Errors Reading Symbol Files
21410 While reading a symbol file, @value{GDBN} occasionally encounters problems,
21411 such as symbol types it does not recognize, or known bugs in compiler
21412 output. By default, @value{GDBN} does not notify you of such problems, since
21413 they are relatively common and primarily of interest to people
21414 debugging compilers. If you are interested in seeing information
21415 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
21416 only one message about each such type of problem, no matter how many
21417 times the problem occurs; or you can ask @value{GDBN} to print more messages,
21418 to see how many times the problems occur, with the @code{set
21419 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
21422 The messages currently printed, and their meanings, include:
21425 @item inner block not inside outer block in @var{symbol}
21427 The symbol information shows where symbol scopes begin and end
21428 (such as at the start of a function or a block of statements). This
21429 error indicates that an inner scope block is not fully contained
21430 in its outer scope blocks.
21432 @value{GDBN} circumvents the problem by treating the inner block as if it had
21433 the same scope as the outer block. In the error message, @var{symbol}
21434 may be shown as ``@code{(don't know)}'' if the outer block is not a
21437 @item block at @var{address} out of order
21439 The symbol information for symbol scope blocks should occur in
21440 order of increasing addresses. This error indicates that it does not
21443 @value{GDBN} does not circumvent this problem, and has trouble
21444 locating symbols in the source file whose symbols it is reading. (You
21445 can often determine what source file is affected by specifying
21446 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
21449 @item bad block start address patched
21451 The symbol information for a symbol scope block has a start address
21452 smaller than the address of the preceding source line. This is known
21453 to occur in the SunOS 4.1.1 (and earlier) C compiler.
21455 @value{GDBN} circumvents the problem by treating the symbol scope block as
21456 starting on the previous source line.
21458 @item bad string table offset in symbol @var{n}
21461 Symbol number @var{n} contains a pointer into the string table which is
21462 larger than the size of the string table.
21464 @value{GDBN} circumvents the problem by considering the symbol to have the
21465 name @code{foo}, which may cause other problems if many symbols end up
21468 @item unknown symbol type @code{0x@var{nn}}
21470 The symbol information contains new data types that @value{GDBN} does
21471 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
21472 uncomprehended information, in hexadecimal.
21474 @value{GDBN} circumvents the error by ignoring this symbol information.
21475 This usually allows you to debug your program, though certain symbols
21476 are not accessible. If you encounter such a problem and feel like
21477 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
21478 on @code{complain}, then go up to the function @code{read_dbx_symtab}
21479 and examine @code{*bufp} to see the symbol.
21481 @item stub type has NULL name
21483 @value{GDBN} could not find the full definition for a struct or class.
21485 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
21486 The symbol information for a C@t{++} member function is missing some
21487 information that recent versions of the compiler should have output for
21490 @item info mismatch between compiler and debugger
21492 @value{GDBN} could not parse a type specification output by the compiler.
21497 @section @value{GDBN} Data Files
21499 @cindex prefix for data files
21500 @value{GDBN} will sometimes read an auxiliary data file. These files
21501 are kept in a directory known as the @dfn{data directory}.
21503 You can set the data directory's name, and view the name @value{GDBN}
21504 is currently using.
21507 @kindex set data-directory
21508 @item set data-directory @var{directory}
21509 Set the directory which @value{GDBN} searches for auxiliary data files
21510 to @var{directory}.
21512 @kindex show data-directory
21513 @item show data-directory
21514 Show the directory @value{GDBN} searches for auxiliary data files.
21517 @cindex default data directory
21518 @cindex @samp{--with-gdb-datadir}
21519 You can set the default data directory by using the configure-time
21520 @samp{--with-gdb-datadir} option. If the data directory is inside
21521 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
21522 @samp{--exec-prefix}), then the default data directory will be updated
21523 automatically if the installed @value{GDBN} is moved to a new
21526 The data directory may also be specified with the
21527 @code{--data-directory} command line option.
21528 @xref{Mode Options}.
21531 @chapter Specifying a Debugging Target
21533 @cindex debugging target
21534 A @dfn{target} is the execution environment occupied by your program.
21536 Often, @value{GDBN} runs in the same host environment as your program;
21537 in that case, the debugging target is specified as a side effect when
21538 you use the @code{file} or @code{core} commands. When you need more
21539 flexibility---for example, running @value{GDBN} on a physically separate
21540 host, or controlling a standalone system over a serial port or a
21541 realtime system over a TCP/IP connection---you can use the @code{target}
21542 command to specify one of the target types configured for @value{GDBN}
21543 (@pxref{Target Commands, ,Commands for Managing Targets}).
21545 @cindex target architecture
21546 It is possible to build @value{GDBN} for several different @dfn{target
21547 architectures}. When @value{GDBN} is built like that, you can choose
21548 one of the available architectures with the @kbd{set architecture}
21552 @kindex set architecture
21553 @kindex show architecture
21554 @item set architecture @var{arch}
21555 This command sets the current target architecture to @var{arch}. The
21556 value of @var{arch} can be @code{"auto"}, in addition to one of the
21557 supported architectures.
21559 @item show architecture
21560 Show the current target architecture.
21562 @item set processor
21564 @kindex set processor
21565 @kindex show processor
21566 These are alias commands for, respectively, @code{set architecture}
21567 and @code{show architecture}.
21571 * Active Targets:: Active targets
21572 * Target Commands:: Commands for managing targets
21573 * Byte Order:: Choosing target byte order
21576 @node Active Targets
21577 @section Active Targets
21579 @cindex stacking targets
21580 @cindex active targets
21581 @cindex multiple targets
21583 There are multiple classes of targets such as: processes, executable files or
21584 recording sessions. Core files belong to the process class, making core file
21585 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
21586 on multiple active targets, one in each class. This allows you to (for
21587 example) start a process and inspect its activity, while still having access to
21588 the executable file after the process finishes. Or if you start process
21589 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
21590 presented a virtual layer of the recording target, while the process target
21591 remains stopped at the chronologically last point of the process execution.
21593 Use the @code{core-file} and @code{exec-file} commands to select a new core
21594 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
21595 specify as a target a process that is already running, use the @code{attach}
21596 command (@pxref{Attach, ,Debugging an Already-running Process}).
21598 @node Target Commands
21599 @section Commands for Managing Targets
21602 @item target @var{type} @var{parameters}
21603 Connects the @value{GDBN} host environment to a target machine or
21604 process. A target is typically a protocol for talking to debugging
21605 facilities. You use the argument @var{type} to specify the type or
21606 protocol of the target machine.
21608 Further @var{parameters} are interpreted by the target protocol, but
21609 typically include things like device names or host names to connect
21610 with, process numbers, and baud rates.
21612 The @code{target} command does not repeat if you press @key{RET} again
21613 after executing the command.
21615 @kindex help target
21617 Displays the names of all targets available. To display targets
21618 currently selected, use either @code{info target} or @code{info files}
21619 (@pxref{Files, ,Commands to Specify Files}).
21621 @item help target @var{name}
21622 Describe a particular target, including any parameters necessary to
21625 @kindex set gnutarget
21626 @item set gnutarget @var{args}
21627 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
21628 knows whether it is reading an @dfn{executable},
21629 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
21630 with the @code{set gnutarget} command. Unlike most @code{target} commands,
21631 with @code{gnutarget} the @code{target} refers to a program, not a machine.
21634 @emph{Warning:} To specify a file format with @code{set gnutarget},
21635 you must know the actual BFD name.
21639 @xref{Files, , Commands to Specify Files}.
21641 @kindex show gnutarget
21642 @item show gnutarget
21643 Use the @code{show gnutarget} command to display what file format
21644 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
21645 @value{GDBN} will determine the file format for each file automatically,
21646 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
21649 @cindex common targets
21650 Here are some common targets (available, or not, depending on the GDB
21655 @item target exec @var{program}
21656 @cindex executable file target
21657 An executable file. @samp{target exec @var{program}} is the same as
21658 @samp{exec-file @var{program}}.
21660 @item target core @var{filename}
21661 @cindex core dump file target
21662 A core dump file. @samp{target core @var{filename}} is the same as
21663 @samp{core-file @var{filename}}.
21665 @item target remote @var{medium}
21666 @cindex remote target
21667 A remote system connected to @value{GDBN} via a serial line or network
21668 connection. This command tells @value{GDBN} to use its own remote
21669 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
21671 For example, if you have a board connected to @file{/dev/ttya} on the
21672 machine running @value{GDBN}, you could say:
21675 target remote /dev/ttya
21678 @code{target remote} supports the @code{load} command. This is only
21679 useful if you have some other way of getting the stub to the target
21680 system, and you can put it somewhere in memory where it won't get
21681 clobbered by the download.
21683 @item target sim @r{[}@var{simargs}@r{]} @dots{}
21684 @cindex built-in simulator target
21685 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
21693 works; however, you cannot assume that a specific memory map, device
21694 drivers, or even basic I/O is available, although some simulators do
21695 provide these. For info about any processor-specific simulator details,
21696 see the appropriate section in @ref{Embedded Processors, ,Embedded
21699 @item target native
21700 @cindex native target
21701 Setup for local/native process debugging. Useful to make the
21702 @code{run} command spawn native processes (likewise @code{attach},
21703 etc.@:) even when @code{set auto-connect-native-target} is @code{off}
21704 (@pxref{set auto-connect-native-target}).
21708 Different targets are available on different configurations of @value{GDBN};
21709 your configuration may have more or fewer targets.
21711 Many remote targets require you to download the executable's code once
21712 you've successfully established a connection. You may wish to control
21713 various aspects of this process.
21718 @kindex set hash@r{, for remote monitors}
21719 @cindex hash mark while downloading
21720 This command controls whether a hash mark @samp{#} is displayed while
21721 downloading a file to the remote monitor. If on, a hash mark is
21722 displayed after each S-record is successfully downloaded to the
21726 @kindex show hash@r{, for remote monitors}
21727 Show the current status of displaying the hash mark.
21729 @item set debug monitor
21730 @kindex set debug monitor
21731 @cindex display remote monitor communications
21732 Enable or disable display of communications messages between
21733 @value{GDBN} and the remote monitor.
21735 @item show debug monitor
21736 @kindex show debug monitor
21737 Show the current status of displaying communications between
21738 @value{GDBN} and the remote monitor.
21743 @kindex load @var{filename} @var{offset}
21744 @item load @var{filename} @var{offset}
21746 Depending on what remote debugging facilities are configured into
21747 @value{GDBN}, the @code{load} command may be available. Where it exists, it
21748 is meant to make @var{filename} (an executable) available for debugging
21749 on the remote system---by downloading, or dynamic linking, for example.
21750 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
21751 the @code{add-symbol-file} command.
21753 If your @value{GDBN} does not have a @code{load} command, attempting to
21754 execute it gets the error message ``@code{You can't do that when your
21755 target is @dots{}}''
21757 The file is loaded at whatever address is specified in the executable.
21758 For some object file formats, you can specify the load address when you
21759 link the program; for other formats, like a.out, the object file format
21760 specifies a fixed address.
21761 @c FIXME! This would be a good place for an xref to the GNU linker doc.
21763 It is also possible to tell @value{GDBN} to load the executable file at a
21764 specific offset described by the optional argument @var{offset}. When
21765 @var{offset} is provided, @var{filename} must also be provided.
21767 Depending on the remote side capabilities, @value{GDBN} may be able to
21768 load programs into flash memory.
21770 @code{load} does not repeat if you press @key{RET} again after using it.
21775 @kindex flash-erase
21777 @anchor{flash-erase}
21779 Erases all known flash memory regions on the target.
21784 @section Choosing Target Byte Order
21786 @cindex choosing target byte order
21787 @cindex target byte order
21789 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
21790 offer the ability to run either big-endian or little-endian byte
21791 orders. Usually the executable or symbol will include a bit to
21792 designate the endian-ness, and you will not need to worry about
21793 which to use. However, you may still find it useful to adjust
21794 @value{GDBN}'s idea of processor endian-ness manually.
21798 @item set endian big
21799 Instruct @value{GDBN} to assume the target is big-endian.
21801 @item set endian little
21802 Instruct @value{GDBN} to assume the target is little-endian.
21804 @item set endian auto
21805 Instruct @value{GDBN} to use the byte order associated with the
21809 Display @value{GDBN}'s current idea of the target byte order.
21813 If the @code{set endian auto} mode is in effect and no executable has
21814 been selected, then the endianness used is the last one chosen either
21815 by one of the @code{set endian big} and @code{set endian little}
21816 commands or by inferring from the last executable used. If no
21817 endianness has been previously chosen, then the default for this mode
21818 is inferred from the target @value{GDBN} has been built for, and is
21819 @code{little} if the name of the target CPU has an @code{el} suffix
21820 and @code{big} otherwise.
21822 Note that these commands merely adjust interpretation of symbolic
21823 data on the host, and that they have absolutely no effect on the
21826 @node Heterogeneous Debugging
21827 @chapter Debugging Heterogeneous Programs
21828 @cindex heterogeneous debugging
21830 @cindex heterogeneous system
21831 @cindex heterogeneous program
21832 In some operating systems, such as Linux with @acronym{AMD}'s
21833 @acronym{ROCm, Radeon Open Compute platforM} installed, a single
21834 program may have multiple threads in the same process, executing on
21835 different devices which may have different target architectures. Such
21836 a system is termed a @dfn{heterogeneous system} and a program that
21837 uses the multiple devices is termed a @dfn{heterogeneous program}.
21839 @cindex heterogeneous agent
21840 The multiple devices of a heterogeneous system are termed
21841 @dfn{heterogeneous agents}. They can include the following kinds of
21842 devices: @acronym{CPU, Central Processing Unit}, @acronym{GPU,
21843 Graphics Processing Unit}, @acronym{DSP, Digital Signal Processor},
21844 @acronym{FPGA, Field Programmable Gate Array}, as well as other
21845 specialized hardware.
21847 @cindex heterogeneous host agent
21848 The device of a heterogeneous system that starts the execution of the
21849 program is termed the @dfn{heterogeneous host agent}.
21851 The precise way threads are created on different heterogeneous agents
21852 may vary from one heterogeneous system to another, but in general the
21853 threads behave similarly no matter what heterogeneous agent is
21854 executing them, except that the target architecture may be different.
21856 @cindex heterogeneous queue
21857 @cindex heterogeneous packet
21858 A heterogeneous program can create @dfn{heterogeneous queues}
21859 associated with a heterogeneous agent. The heterogeneous program can
21860 then place @dfn{heterogeneous packets} on a heterogeneous queue to
21861 control the actions of the associated heterogeneous agent. A
21862 heterogeneous agent removes heterogeneous packets from the
21863 heterogeneous queues assocated with it and performs the requested
21864 actions. The packet actions and scheduling of packet processing
21865 varies depending on the heterogeneous system and the target
21866 architecture of the heterogeneous agent. @xref{Architectures}.
21868 @cindex heterogeneous dispatch packet
21869 @cindex heterogeneous dispatch
21870 A @dfn{heterogeneous dispatch packet} is used to initiate code
21871 execution on a heterogeneous agent. A single heterogeneous dispatch
21872 packet may specify that the heterogeneous agent create a set of
21873 threads that are all associated with a corresponding
21874 @dfn{heterogeneous dispatch}. Each thread typically has an associated
21875 position within the heterogeneous dispatch, possibly expressed as a
21876 multi-dimensional grid position. The heterogeneous agent typically
21877 can create multiple threads that execute concurrently. If a
21878 heterogeneous dispatch is larger than the number of concurrent threads
21879 that can be created, the heterogeneous agent creates threads of the
21880 heterogeneous dispatch as other threads complete. When all the
21881 threads of a heterogeneous dispatch have been created and have
21882 completed, the heterogeneous dispatch is considered complete.
21884 @cindex heterogeneous work-group
21885 The threads of a heterogeneous dispatch may be grouped into
21886 @dfn{heterogeneous work-groups}. The threads that belong to the same
21887 heterogeneous work-group may have special shared memory, and efficient
21888 execution synchronization abilities. A thread that is part of a
21889 heterogeneous work-group typically has an associated position within
21890 the heterogeneous work-group, possibly also expressed as a
21891 multi-dimensional grid position.
21893 Other heterogeneous packets may control heterogeneous packet
21894 scheduling, memory visibility between the threads of a heterogeneous
21895 dispatch and other threads, or other services supported by the
21896 heterogeneous system.
21898 @cindex heterogeneous lane
21899 On some heterogeneous systems there can be heterogeneous agents that
21900 support @acronym{SIMD, Single Instruction Multiple Data} or
21901 @acronym{SIMT, Single Instruction Multiple Threads} machine
21902 instructions. On these target achitectures, a single machine
21903 instruction can operate in parallel on multiple @dfn{heterogeneous
21906 @cindex divergent control flow
21907 Source languages used by heterogeneous programs can be implemented on
21908 target achitectures that support multiple heterogeneous lanes by
21909 mapping a source language thread of execution onto a heterogeneous
21910 lane of a single target architecture thread. Control flow in the
21911 source language may be implemented by controlling which heterogeneous
21912 lanes are active. If the source language control flow may result in
21913 some heterogeneous lanes becoming inactive while some remain active,
21914 the control flow is said to be @dfn{divergent}. Typically, the
21915 machine code may execute different divergent paths for different sets
21916 of heterogeneous lanes, before the control flow recoverges and all
21917 heterogeneous lanes become active.
21919 Just because a target architecture supports multiple lanes, does not
21920 mean that the source language is mapped to use them to implement
21921 source language threads of execution. Therefore, a thread is only
21922 considered to have multiple heterogeneous lanes if it's current frame
21923 corresponds to a source language that does do such a mapping.
21925 @anchor{Address Space}
21926 @cindex address space
21927 On some heterogeneous systems there can be heterogeneous agents with
21928 target achitectures that support multiple @dfn{address spaces}. In
21929 these target achitectures, there may be memory that is physically
21930 disjoint from regular global virtual memory. There can also be cases
21931 when the same underlying memory can be accessed using linear addresses
21932 that map to the underlying physical memory in an interleaved manner.
21933 In these target architectures there can be distinct machine
21934 instructions to access the distinct address spaces. For example,
21935 there may be physical hardware scratch pad memory that is allocated
21936 and accessible only to the threads that are associated with the same
21937 heterogeneous work-group. There may be hardware address swizzle logic
21938 that allows regular global virtual memory to be allocated per
21939 heterogeneous lane such that they have a linear address view, which in
21940 fact maps to an interleaved global virtual memory access to improve
21943 @value{GDBN} provides these facilities for debugging heterogeneous
21948 @item @code{info sharedlibrary}, command supports code objects for
21949 multiple architectures
21951 @item debugger convenience variables for heterogeneous entities
21953 @item @code{set architecture}, @code{show architecture}, @code{x/i},
21954 @code{disassemble}, commands to disassemble multiple architectures in
21957 @item @code{info threads}, @code{thread}, commands support threads
21958 executing on multiple heterogeneous agents
21960 @item @code{info agents}, @code{info queues}, @code{info packets},
21961 @code{info dispatches}, commands to inquire about the heterogeneous
21964 @item @code{info lanes}, @code{lane}, commands support source language
21965 threads of execution that are mapped to SIMD-like lanes of a thread
21967 @item @code{$_thread_find}, @code{$_thread_find_first_gid},
21968 @code{$_lane_find}, @code{$_lane_find_first_gid} debugger convenience
21969 functions can find threads and heterogeneous lanes associated with
21970 specific heterogeneous entities
21972 @item @code{maint print address-spaces}, command together with address
21973 qualifiers supports multiple address spaces
21977 A heterogeneous system may use separate code objects for the different
21978 target architectures of the heterogeneous agents. The @code{info
21979 sharedlibrary} command lists all the code objects currently loaded,
21980 regardless of their target architecture.
21982 The following rules apply in determining the target architecture used
21983 by commands when debugging heterogeneous programs:
21988 Typically the target architecture of the heterogeneous host agent is
21989 the target architecture of the program's code object. The @code{set
21990 architecture} command (@pxref{Targets,,Specifying a Debugging Target})
21991 can be used to change this target architecture. The target
21992 architecture of other heterogeneous agents is typically the target
21993 architecture of the associated device.
21996 The target architecture of a thread is the target architecture of the
21997 selected stack frame. Typically stack frames will have the same
21998 target architecture as the heterogeneous agent on which the thread was
21999 created, however, a target may assocociate different target
22000 architectures for different stack frames.
22003 The current target architecture is the target architecture of the
22004 selected thread, or the target architecture of the heterogeneous host
22005 agent if there are no threads.
22009 @value{GDBN} handles the heterogeneous agent, queue, and dispatch
22010 entities in a similar manner to threads (@pxref{Threads}):
22015 For debugging purposes, @value{GDBN} associates its own number
22016 ---always a single integer---with each heterogeneous entity of an
22017 inferior. This number is unique between all instances of
22018 heterogeneous entities of an inferior, but not unique between
22019 heterogeneous entities of different inferiors.
22022 You can refer to a given heterogeneous entity in an inferior using the
22023 qualified @var{inferior-num}.@var{heterogeneous-entity-num} syntax,
22024 also known as a @dfn{qualified heterogeneous entity ID}, with
22025 @var{inferior-num} being the inferior number and
22026 @var{heterogeneous-entity-num} being the heterogeneous entity number
22027 of the given inferior. If you omit @var{inferior-num}, then
22028 @value{GDBN} infers you're referring to a heterogeneous entity of the
22032 Until you create a second inferior, @value{GDBN} does not show the
22033 @var{inferior-num} part of heterogeneous entity IDs, even though you
22034 can always use the full
22035 @var{inferior-num}.@var{heterogeneous-entity-num} form to refer to
22036 heterogeneous entities of inferior 1, the initial inferior.
22039 @anchor{heterogeneous entity ID list}
22040 @cindex heterogeneous entity ID list
22041 Some commands accept a space-separated @dfn{heterogeneous entity ID
22042 list} as argument. The list element has the same forms as for thread
22043 ID lists. @xref{thread ID list}.
22046 @anchor{global heterogeneous entity numbers} In addition to a
22047 @emph{per-inferior} number, each heterogeneous entity is also assigned
22048 a unique @emph{global} number, also known as @dfn{global heterogeneous
22049 entity ID}, a single integer. Unlike the heterogeneous entity number
22050 component of the heterogeneous entity ID, no two threads have the same
22051 global heterogeneous entity ID, even when you're debugging multiple
22056 The following debugger convenience variables (@pxref{Convenience
22057 Vars,,Convenience Variables}) are related to heterogeneous debugging.
22058 You may find these useful in writing breakpoint conditional
22059 expressions, command scripts, and so forth.
22065 @itemx $_thread_systag
22066 @itemx $_thread_name
22067 @xref{Convenience Vars,,Convenience Variables}.
22069 @vindex $_agent@r{, convenience variable}
22070 @vindex $_gagent@r{, convenience variable}
22071 @vindex $_queue@r{, convenience variable}
22072 @vindex $_gqueue@r{, convenience variable}
22073 @vindex $_dispatch@r{, convenience variable}
22074 @vindex $_gdispatch@r{, convenience variable}
22075 @vindex $_lane@r{, convenience variable}
22076 @vindex $_glane@r{, convenience variable}
22085 There are debugger convenience variables that contain the number of
22086 each heterogeneous entity associated with the current thread if it was
22087 created by a heterogeneous dispatch, or 0 otherwise. @code{$_agent},
22088 @code{$_queue}, and @code{$_dispatch} contain the corresponding
22089 per-inferior heterogeneous entity number. While @code{$_gagent},
22090 @code{$_gqueue}, and @code{$_gdispatch}, contain the corresponding
22091 global heterogeneous entity number.
22093 @vindex $_lane@r{, convenience variable}
22094 @vindex $_glane@r{, convenience variable}
22097 The heterogeneous lane number of the current lane of the current thread.
22098 @code{$_lane} contains the corresponding per-inferior heterogeneous lane
22099 number. While @code{$_glane} contains the corresponding global
22100 heterogeneous lane number. If the current thread does not have multiple
22101 heterogeneous lanes, it is treated as if it has a single heterogeneous
22104 @vindex $_dispatch_pos@r{, convenience variable}
22105 @item $_dispatch_pos
22106 The heterogeneous dispatch position string of the current thread within
22107 its associated heterogeneous dispatch if it is was created by a
22108 heterogeneous dispatch, or the empty string otherwise. The format
22109 varies depending on the heterogeneous system and target architecture
22110 of the heterogeneous agent. @xref{Architectures}.
22112 @vindex $_lane_name@r{, convenience variable}
22114 The heterogeneous lane name string of the current heterogeneous lane, or
22115 the empty string if no name has been assigned by the @code{lane name}
22118 @vindex $_thread_workgroup_pos@r{, convenience variable}
22119 @vindex $_lane_workgroup_pos@r{, convenience variable}
22120 @item $_thread_workgroup_pos
22121 @item $_lane_workgroup_pos
22122 The heterogeneous work-group position string of the current thread or
22123 heterogeneous lane within its associated heterogeneous dispatch if it
22124 is was created by a heterogeneous dispatch, or the empty string
22125 otherwise. The format varies depending on the heterogeneous system
22126 and target architecture of the heterogeneous agent.
22127 @xref{Architectures}.
22129 @vindex $_lane_systag@r{, convenience variable}
22130 @item $_lane_systag
22131 The target system's heterogeneous lane identifier (@var{lane_systag})
22132 string of the current heterogeneous lane. @xref{target system lane
22137 The following debugger convenience functions (@pxref{Convenience
22138 Funs,,Convenience Functions}) are related to heterogeneous debugging.
22139 Given the very large number of threads on heterogeneous systems, these
22140 may be very useful. They allow threads or thread lists to be
22141 specified based on the target system's thread identifier
22142 (@var{systag}) or thread name, and allow heterogeneous lanes or
22143 heterogeneous lane lists to be specified based on the target system's
22144 heterogeneous lane identifier (@var{lane_systag}) or heterogeneous
22149 @item $_thread_find
22150 @itemx $_thread_find_first_gid
22151 @xref{Convenience Funs,,Convenience Functions}.
22153 @findex $_lane_find@r{, convenience function}
22154 @item $_lane_find(@var{regex})
22155 Searches for heterogeneous lanes whose name or @var{lane_systag}
22156 matches the supplied regular expression. The syntax of the regular
22157 expression is that specified by @code{Python}'s regular expression
22160 Returns a string that is the space separated list of per-inferior
22161 heterogeneous lane numbers of the found heterogeneous lanes. If
22162 debugging multiple inferiors, the heterogeneous lane numbers are
22163 qualified with the inferior number. If no heterogeneous lane are
22164 found, the empty string is returned. The string can be used in
22165 commands that accept a heterogeneous lane ID list.
22166 @xref{heterogeneous entity ID list}.
22168 For example, the following command lists all heterogeneous lanes that
22169 are part of a heterogeneous work-group with work-group position
22170 @samp{(1,2,3)} (@pxref{Heterogeneous Debugging}):
22173 (@value{GDBP}) info lanes $_thread_find ("work-item(1,2,3)")
22176 @item $_lane_find_first_gid(@var{regex})
22177 @findex $_lane_find_first_gid@r{, convenience function}
22178 Similar to the @code{$_lane_find} convenience function, except it
22179 returns a number that is the global heterogeneous lane number of one
22180 of the heterogeneous lanes found, or 0 if no heterogeneous lanes were
22181 found. The number can be used in commands that accept a global
22182 heterogeneous lane number. @xref{global heterogeneous entity
22185 For example, the following command sets the current heterogeneous lane
22186 to one of the heterogeneous lanes that are part of a heterogeneous
22187 work-group with work-item position @samp{(1,2,3)}:
22190 (@value{GDBP}) lane -gid $_lane_find_first_gid ("work-item(1,2,3)")
22195 The following commands are related to heterogeneous debugging:
22199 @item info agents @r{[}-gid@r{]} @r{[}@var{agent-id-list}@r{]}
22200 @itemx info queues @r{[}-gid@r{]} @r{[}@var{queue-id-list}@r{]}
22201 @itemx info dispatches @r{[}-gid@r{]} @r{[}@var{dispatch-id-list}@r{]}
22202 @kindex info agents
22203 @kindex info queues
22204 @kindex info dispatches
22205 @code{info agents}, @code{info queues} and @code{info dispatches}
22206 commands display information about one or more heterogeneous agents,
22207 heterogeneous queues and executing heterogeneous dispatches
22208 respectively. With no arguments displays information about all
22209 corresponding heterogeneous entities. You can specify the list of
22210 heterogeneous entities that you want to display using the
22211 heterogeneous entity ID list syntax (@pxref{heterogeneous entity ID
22214 @value{GDBN} displays for each heterogeneous entity (in this order):
22218 the per-inferior heterogeneous entity number assigned by @value{GDBN}
22221 the global heterogeneous entity number assigned by @value{GDBN}, if
22222 the @w{@option{-gid}} option was specified
22225 for the @code{info queues} and @code{info dispatches} commands, the
22226 associated heterogeneous agent number assigned by @value{GDBN},
22227 displayed as a global ID if the @w{@option{-gid}} option was
22228 specified, otherwise displayed as the per-inferior ID
22231 for the @code{info dispatches} command, the associated heterogeneous
22232 queue number assigned by @value{GDBN}, displayed as a global ID if the
22233 @w{@option{-gid}} option was specified, otherwise displayed as the
22237 additional information about the heterogeneous entity that varies
22238 depending on the heterogeneous system and may vary depending on the
22239 target architecture of the heterogeneous entity
22240 (@pxref{Architectures})
22244 Some heterogeneous agents may not be listed until the inferior has
22245 started execution of the program.
22247 @item info packets @r{[}-gid@r{]} @r{[}@var{queue-id-list}@r{]}
22248 @kindex info packets
22249 Display information about the heterogeneous packets on one or more
22250 heterogeneous queues. With no arguments displays information about
22251 all heterogeneous queues. You can specify the list of heterogeneous
22252 queues that you want to display using the heterogeneous queue ID list
22253 syntax (@pxref{heterogeneous entity ID list}).
22255 Since heterogeneous agents may be processing heterogeneous packets
22256 asynchronously, the display is at best a snapshot, and may be
22257 inconsistent due to the heterogeneous queues being updated while they
22258 are being inspected.
22260 The heterogeneous packets are listed contiguously for each
22261 heterogeneous agent, and for each heterogeneous queue of that
22262 heterogeneous agent, with the oldest packet first.
22264 @value{GDBN} displays for each heterogeneous packet (in this order):
22268 the associated heterogeneous agent number assigned by @value{GDBN},
22269 displayed as a global ID if the @w{@option{-gid}} option was
22270 specified, otherwise displayed as the per-inferior ID
22273 the associated heterogeneous queue number assigned by @value{GDBN},
22274 displayed as a global ID if the @w{@option{-gid}} option was
22275 specified, otherwise displayed as the per-inferior ID
22278 the packet position in the heterogeneous queue, with the oldest one
22282 additional information about the heterogeneous packet that varies
22283 depending on the heterogeneous system and may vary depending on the
22284 target architecture of the heterogeneous entity
22285 (@pxref{Architectures})
22289 @item info threads @r{[}-gid@r{]} @r{[}@var{thread-id-list}@r{]}
22290 The @code{info threads} command (@pxref{Threads}) lists the threads
22291 created on all the heterogeneous agents.
22293 If any of the threads listed have multiple heterogeneous lanes, then
22294 an additional @emph{Lanes} column is displayed before the target
22295 system's thread identifier (@var{systag}) column. For threads that
22296 have multiple heterogeneous lanes, the number of heterogeneous lanes
22297 that are active followed by a slash and the total number of
22298 heterogeneous lanes of the current frame of the thread is displayed.
22299 Otherwise, nothing is displayed.
22301 The target system's thread identifier (@var{systag}) (@pxref{target
22302 system thread identifier}) for threads associated with heterogeneous
22303 dispatches varies depending on the heterogeneous system and target
22304 architecture of the heterogeneous agent. However, it typically will
22305 include information about the heterogeneous agent, heterogeneous
22306 queue, heterogeneous dispatch, heterogeneous work-group position
22307 within the heterogeneous dispatch, and thread position within the
22308 heterogeneous work-group. @xref{Architectures}.
22310 The stack frame summary displayed is for the active lanes of the
22311 thread. This may differ from the stack frame information for the
22312 current lane if the focus is on an inactive lane. Use the @code{info
22313 lanes} command for information about individual lanes of a thread.
22318 @c end table here to get a little more width for example
22321 (@value{GDBP}) info threads
22322 Id Lanes Target Id Frame
22323 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
22324 2 2/64 ROCm process 35 agent 1 queue 2 dispatch 3 work-group(2,3,4)/1 0x34e5 in saxpy ()
22325 3 64/64 ROCm process 65 agent 1 queue 2 dispatch 4 work-group(2,4,4)/2 0x34e5 in saxpy ()
22329 @cindex heterogeneous lane index
22330 @item thread @r{[}-gid@r{]} @var{thread-id} @r{[}@var{lane-index}@r{]}
22331 The @code{thread} command has an optional @var{lane-index} argument to
22332 specify the @dfn{heterogeneous lane index}. If the value is not
22333 between 1 and the number of heterogeneous lanes of the current frame
22334 of the thread, then @value{GDBN} will print an error. If omitted it
22337 The current thread is set to @var{thread-id} and the current
22338 heterogeneous lane is set to the heterogeneous lane corresponding to
22339 the specified heterogeneous lane index.
22341 If the thread has multiple heterogeneous lanes, @value{GDBN} responds
22342 by displaying the system identifier of the heterogeneous lane you
22343 selected, otherwise it responds with the system identifier of the
22344 thread you selected, followed by its current stack frame summary.
22346 @item thread apply @r{[}@var{thread-id-list} @r{|} all @r{[}-ascending@r{]]} @r{[}@var{flag}@r{]@dots{}} @var{command}
22347 @itemx taas [@var{option}]@dots{} @var{command}
22348 @itemx tfaas [@var{option}]@dots{} @var{command}
22351 These commands operate the same way for all threads, regardless of
22352 whether or not the thread is associated with a heterogeneous dispatch.
22354 If the thread's frame has multiple heterogeneous lanes then the
22355 heterogeneous lane index 1 is used. Use the heterogeneous lane
22356 counterpart commands if it is desired to perform the the @var{command}
22357 on each lane of a thread.
22361 @cindex lane identifier (system)
22362 @item info lanes @r{[}-gid@r{]} @var{lane-id}
22363 Display information about one or more heterogeneous lanes. With no
22364 arguments displays information about all heterogeneous lanes. You can
22365 specify the list of heterogeneous lanes that you want to display using
22366 the heterogeneous lane ID list syntax (@pxref{heterogeneous entity ID
22369 @value{GDBN} displays for each heterogeneous lane (in this order):
22373 The per-inferior heterogeneous lane number assigned by @value{GDBN}.
22376 The global heterogeneous lane number assigned by @value{GDBN}, if the
22377 @w{@option{-gid}} option was specified.
22380 The thread number assigned by @value{GDBN} for the thread that
22381 contains the heterogeneous lane. This is displayed as a global thread
22382 number if the @w{@option{-gid}} option was specified, otherwise as a
22383 per-inferior thread number. If the thread has multiple heterogeneous
22384 lanes then this is followed by a slash and the heterogeneous lane
22385 index of the heterogeneous lane within the thread with the first lane
22389 An indication of whether the heterogeneous lane is active or inactive.
22391 @anchor{target system lane identifier}
22393 The target system's heterogeneous lane identifier (@var{lane_systag}).
22394 This varies depending on the system and target architecture of the
22395 heterogeneous agent. However, for heterogeneous agents it typically
22396 will include information about the heterogeneous agent, heterogeneous
22397 queue, heterogeneous dispatch, heterogeneous work-group position
22398 within the heterogeneous dispatch, and position of the heterogeneous
22399 lane in the heterogeneous work-group. @xref{Architectures}.
22402 The heterogeneous lane's name, if one is assigned by the user (see
22403 @code{lane name}, below).
22406 The current stack frame summary for that heterogeneous lane. If the
22407 heterogeneous lane is inactive this is the source position at which the
22408 heterogeneous lane will resume.
22412 An asterisk @samp{*} to the left of the @value{GDBN} heterogeneous
22413 lane number indicates the current heterogeneous lane.
22417 @c end table here to get a little more width for example
22420 (@value{GDBP}) info lanes
22421 Id Thread Active Target Id Frame
22422 * 1 4 Y process 35 thread 13 main (argc=1, argv=0x7ffffff8)
22423 2 5/2 Y ROCm process 35 agent 1 queue 2 dispatch 3 work-group(2,3,4) work-item(1,2,4) 0x34e5 in saxpy ()
22424 3 6/12 N ROCm process 65 agent 1 queue 2 dispatch 4 work-group(2,4,4) work-item(1,2,3) 0x34e5 in saxpy ()
22427 If you're debugging multiple inferiors, @value{GDBN} displays
22428 heterogeneous lane IDs using the qualified
22429 @var{inferior-num}.@var{lane-num} format. Otherwise, only
22430 @var{lane-num} is shown.
22432 If you specify the @w{@option{-gid}} option, @value{GDBN} displays a
22433 column indicating each heterogeneous lane's global heterogeneous lane
22434 ID, and displays the thread's global thread number:
22437 (@value{GDBP}) info lanes -gid
22438 Id GId Thread Active Target Id Frame
22439 * 1.1 1 4 Y process 35 thread 13 main (argc=1, argv=0x7ffffff8)
22440 1.2 3 5/2 Y ROCm process 35 agent 1 queue 2 dispatch 3 work-group(2,3,4) work-item(1,2,4) 0x34e5 in saxpy ()
22441 2.1 1 4 Y process 65 thread 1 main (argc=1, argv=0x7ffffff8)
22442 2.2 4 6/12 N ROCm process 65 agent 1 queue 2 dispatch 4 work-group(2,4,4) work-item(1,2,3) 0x34e5 in saxpy ()
22447 @item lane @r{[}-gid@r{]} @var{lane-id}
22448 Make heterogeneous lane ID @var{lane-id} the current heterogeneous
22449 lane and the thread that contains the heterogeneous lane the current
22450 thread. The command argument @var{lane-id} is the @value{GDBN}
22451 heterogeneous lane ID: if the @w{@option{-gid}} option is given it is
22452 a global heterogeneous lane identifier, as shown in the second field
22453 of the @code{info lanes -gid} display; otherwise it is a per-inferior
22454 heterogeneous lane identifier, with or without an inferior qualifier
22455 (e.g., @samp{2.1} or @samp{1}), as shown in the first field of the
22456 @code{info lanes} display.
22458 @value{GDBN} responds by displaying the system identifier of the
22459 heterogeneous lane you selected, and its current stack frame summary:
22462 (@value{GDBP}) lane 2
22463 [Switching to lane 2 (Thread 0xb7fdab70 (LWP 12747))]
22464 #0 some_function (ignore=0x0) at example.c:8
22465 8 printf ("hello\n");
22469 As with the @samp{[New @dots{}]} message, the form of the text after
22470 @samp{Switching to} depends on your system's conventions for identifying
22471 heterogeneous lanes.
22474 @cindex name a heterogeneous lane
22475 @anchor{heterogeneous lane name}
22476 @item lane name [@var{name}]
22477 This command assigns a name to the current heterogeneous lane. If no
22478 argument is given, any existing user-specified name is removed. The
22479 heterogeneous lane name appears in the @code{info lanes} display.
22482 @cindex search for a heterogeneous lane
22484 @item lane find [@var{regexp}]
22485 Search for and display heterogeneous lane ids whose name or
22486 @var{lane_systag} matches the supplied regular expression. The syntax
22487 of the regular expression is that specified by @code{Python}'s regular
22488 expression support.
22490 As well as being the complement to the @code{lane name} command, this
22491 command also allows you to identify a heterogeneous lane by its target
22492 @var{lane_systag}. For instance, on @acronym{AMD ROCm}, the target
22493 @var{lane_systag} is the heterogeneous agent, heterogeneous queue,
22494 heterogeneous dispatch, heterogeneous work-group position and
22495 heterogeneous work-item position.
22498 (@value{GDBP}) lane find "work-group(2,3,4)"
22499 Lane 2 has lane id 'ROCm process 35 agent 1 queue 2 dispatch 3 work-group(2,3,4) work-item(1,2,4)'
22500 (@value{GDBP}) info lane 2
22501 Id Thread Active Target Id Frame
22502 2 5/2 Y ROCm process 35 agent 1 queue 2 dispatch 3 work-group(2,3,4) work-item(1,2,4) 0x34e5 in saxpy ()
22505 @c FIXME-implementors!! Perhaps better ways to find lanes and threads
22506 @c would be beneficial. If the @var{systag} and ${lane_systag} were
22507 @c considered as tuples and not a plain strings, structured queries
22508 @c could be used. Maybe that would also support the sort order of the
22509 @c returned list. SQL is an example to examine.
22511 @c User defined pretty printing functions could be allowed so that
22512 @c users can control how @var{systag} and ${lane_systag} values are
22513 @c displayed in commands that display them. This would allow cater to
22514 @c situations that benefit from full verbose output, and those where
22515 @c partial terse output is all that is needed. But the underlying
22516 @c @var{systag} and ${lane_systag} values always have the full
22519 @c Ways for commands that list lanes and threads to aggregate the
22520 @c output would be beneficial in heterogeneous systems that tend to
22521 @c have very large counts. For example, all lanes that have adjacent
22522 @c dispatch postions, and that are at the same source postion, could
22523 @c be displayed as a single row that specifies the range of postions.
22524 @c Perhaps target or user defined functions could be allowed to guide
22525 @c the aggregation, and return the aggregated range. That would allow
22526 @c different heterogeneous system to be supported that had different
22527 @c ways to represent dispatch positions. There may even be multiple
22528 @c ways to aggregate on some system.
22530 @item lane apply @r{[}@var{thread-id-list} @r{|} all @r{[}-ascending@r{]]} @r{[}@var{flag}@r{]@dots{}} @var{command}
22531 @itemx laas [@var{option}]@dots{} @var{command}
22532 @itemx lfaas [@var{option}]@dots{} @var{command}
22533 @code{lane apply}, @code{laas}, and @code{lfass} commands are simalar
22534 to their thread counterparts @code{thread apply}, @code{taas}, and
22535 @code{tfaas} respectively, except they operatate on heterogeneous
22536 lanes. @xref{Threads}.
22538 @item backtrace [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
22539 @itemx frame @r{[} @var{frame-selection-spec} @r{]}
22540 @itemx frame apply [all | @var{count} | @var{-count} | level @var{level}@dots{}] [@var{option}]@dots{} @var{command}
22541 @itemx select-frame @r{[} @var{frame-selection-spec} @r{]}
22542 @itemx up-silently @var{n}
22543 @itemx down-silently @var{n}
22545 @itemx info args [-q] [-t @var{type_regexp}] [@var{regexp}]
22546 @itemx info locals [-q] [-t @var{type_regexp}] [@var{regexp}]
22547 @itemx faas @var{command}
22548 The frame commands apply to the current heterogeneous lane.
22550 If the frame is switched from one that has multiple heterogeneous
22551 lanes to one with fewer (including only one) then the current lane is
22552 switched to the heterogeneous lane corresponding to the highest
22553 heterogeneous lane index of the new frame and @value{GDBN} responds by
22554 displaying the system identifier of the heterogeneous lane selected.
22556 @xref{Stack, ,Examining the Stack}.
22558 @item set libthread-db-search-path
22559 @itemx show libthread-db-search-path
22560 @itemx set debug libthread-db
22561 @itemx show debug libthread-db
22562 These commands only apply to threads created on the heterogeneous host
22563 agent that are not associated with a heterogeneous dispatch. There
22564 are no commands that support reporting of heterogeneous dispatch
22569 The @code{x/i} and @code{display/i} commands (@pxref{Memory,,Examining
22570 Memory}) can be used to disassemble machine instructions. They use
22571 the current target architecture.
22574 The @code{disassemble} command (@pxref{Machine Code,,Source and
22575 Machine Code}) can also be used to disassemble machine instructions.
22576 If the start address of the range is within a loaded code object, then
22577 the target architecture of the code object is used. Otherwise, the
22578 current target architecture is used.
22580 @c FIXME-implementors!! It would be more helpful if @code{set
22581 @c architecture} was an inferior setting used by both @code{x/i} and
22582 @c @code{disassemble} when not set to @code{auto}. When set to
22583 @c @code{auto} then the architecture of the code object containing the
22584 @c start address should be used by both commands. Otherwise, the
22585 @c thread target architecture should be used, or the heterogeneous host
22586 @c agent target architecture if there are no threads. That way a user
22587 @c can choose what architecture to disassemble in, and will get
22588 @c sensible behavior if they specify the default of @code{auto} even
22589 @c for heterogeneous systems.
22591 @item info registers
22592 @itemx info all-registers
22593 @itemx maint print reggroups
22594 The register commands display information about the current
22598 The @code{print} command evaluates the source language expression in
22599 the context of the current heterogeneous lane.
22607 If the current heterogeneous lane is set to an inactive heterogeneous
22608 lane, then the @code{step}, @code{next}, @code{finish} and
22609 @code{until} commands (@pxref{Continuing and Stepping, ,Continuing and
22610 Stepping}) may cause other heterogeneous lanes of the same thread to
22611 advance so that the current heterogeneous lane becomes active. This
22612 may result in other heterogeneous lanes completing whole functions.
22614 If the current heterogeneous lane is set to an inactive heterogeneous
22615 lane, then the @code{stepi} and @code{nexti} commands
22616 (@pxref{Continuing and Stepping, ,Continuing and Stepping}) may not
22617 cause the source position to appear to move until execution reaches a
22618 point that makes the current heterogeneous lane active. However,
22619 other heterogeneous lanes of the same thread will advance.
22621 @item break @r{[}-lane @var{lane-index}@r{]} @r{[}location@r{]} @r{[}if @var{cond}@r{]}
22622 @itemx tbreak @r{[}-lane @var{lane-index}@r{]} @r{[}location@r{]} @r{[}if @var{cond}@r{]}
22623 @itemx hbreak @r{[}-lane @var{lane-index}@r{]} @r{[}location@r{]} @r{[}if @var{cond}@r{]}
22624 @itemx thbreak @r{[}-lane @var{lane-index}@r{]} @r{[}location@r{]} @r{[}if @var{cond}@r{]}
22625 @itemx rbreak @r{[}-lane @var{lane-index}@r{]} @var{regex}
22626 @itemx info breakpoints @r{[}@var{list}@dots{}@r{]}
22627 @itemx watch @r{[}-lane @var{lane-index}@r{]} @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
22628 @itemx rwatch @r{[}-lane @var{lane-index}@r{]} @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
22629 @itemx awatch @r{[}-lane @var{lane-index}@r{]} @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
22630 @itemx info watchpoints @r{[}@var{list}@dots{}@r{]}
22631 @itemx catch @r{[}-lane @var{lane-index}@r{]} @var{event}
22632 @itemx tcatch @r{[}-lane @var{lane-index}@r{]} @var{event}
22633 When a breakpoint, watchpoint, or catchpoint (@pxref{Breakpoints,
22634 ,Breakpoints; Watchpoints; and Catchpoints}) is hit by a frame of a
22635 thread with multiple heterogeneous lanes, each active lane is treated
22641 The breakpoint condition, if present, is evaluated for each active
22642 heterogeneous lane.
22645 The breakpoint command, if present, is evaluated for each active
22646 heterogeneous lane that evaluates the breakpoint condition to true.
22649 If the breakpoint causes the heterogeneous lane to halt then he
22650 current heterogeneous lane is set to the halting heterogeneous lane
22651 and @value{GDBN} responds by displaying the system identifier of the
22652 heterogeneous lane selected.
22655 If the breakpoint is a temporary breakpoint, then it will be removed,
22656 and so any remaining heterogeneous lanes will not report the
22660 In non-stop mode all heterogeneous lanes that halt at the breakpoint
22664 In all-stop mode, continuing from the breakpoint will cause the next
22665 heterogeneous ative lane that hit the breakpoint to be processed.
22669 If a heterogeneous lane causes a thread to halt, then the other
22670 heterogeneous lanes of the thread will no longer execute even if in
22673 For @code{break}, @code{watch}, @code{catch}, and their variants, the
22674 @w{@option{-lane @var{lane-index}}} option can be specified. This
22675 limits @value{GDBN} to only process breakpoints if the heterogeneous
22676 lane has a heterogeneous lane index that matches @var{lane-index}.
22678 The @code{info break} and @code{info watch} commands add a @emph{Lane}
22679 column before the @emph{Address} column if any breakoint has a
22680 @var{lane-index} specified that displays the heterogeneous lane index.
22682 @c FIXME-implementors!! Should there be way to request all pending
22683 @c breakpoints to be processed? This may result in multiple
22684 @c lanes/threads being reported as halted. This would avoid the user
22685 @c having to continue a very large number of times to get all the
22686 @c threads/lanes that have unprocessed breakpoints to be processed.
22688 @c In addition, a way to list all the theards/lanes that are halted at
22689 @c a breakpoint. If this was avaiable as a conveniece function, then
22690 @c the @code{thread apply} and @code{lane apply} commands could be
22691 @c used to perform a command on all such threads in one action.
22693 @anchor{maint print address-spaces}
22694 @c FIXME-implementers!! This is not a maintenance command as it is
22695 @c displaying imformation about available address spaces that can be
22696 @c used. It has been defined as a @code{maint} command only to match
22697 @c the @code{maint print reggroups} command which also should not be a
22698 @c maintenace command for the same reason.
22699 @item maint print address-spaces @r{[}@var{file}@r{]}
22700 @code {maint print address-spaces} displays the address space names
22701 supported by each target achitecture. The optional argument
22702 @var{file} tells to what file to write the information.
22704 The address spaces info looks like this:
22707 (@value{GDBP}) @kbd{maint print address-spaces}
22715 The @var{global} address space corresponds to the default global
22716 virtual memory address space and is available for all target
22719 Every address entered or displayed can optionally specify the address
22720 space qualifier by appending an @samp{@@} followed by an address space
22721 name. @value{GDBN} will print an error if the address space name is
22722 not supported by the current architecture.
22727 (@value{GDBP}) x/x 0x10021608@@group
22728 0x10021608@@group: 0x0022fd98
22731 @c FIXME-implementors!! Perhaps the gdb internal types (such as used
22732 @c for register types) can be extended to support addresses in address
22735 If there is no current thread then the only address space that can be
22736 specified is @var{global}.
22738 If entering an address and no address space is specified, the
22739 @var{global} address space is used.
22741 If an address is displayed, the address space qualifier is omitted for
22742 the @var{global} address space.
22746 Heterogeneous systems often have very large numbers of threads.
22747 Breakpoint conditions can be used to limit the number of threads
22748 reporting breakpoint hits. For example,
22751 break kernel_foo if $_streq($_lane_workgroup_pos, "(0,0,0)")
22754 The @code{tbreak} command can be used so only one heterogeneous lane
22755 will report the breakpoint. Before continuing execution, the
22756 breakpoint will need to be set again if necessary.
22758 The @code{set scheduler-locking on} command together with the
22759 @w{@option{-lane}} breakpoint option can be used to lock @value{GDBN}
22760 to only resume the current thread, and only report breakoints for a
22761 fixed heterogeneous lane index. This avoids the overhead of resuming
22762 a large number of threads every time resuming from a breakpoint, and
22763 also avoids the focus being switched to other threads that hit the
22764 breakpoints. Note however that other threads will not be executed.
22767 @c Change command parsing so convienence variable
22768 @c substitution will work as shown. Investigate using tuples and
22769 @c lists the result of convienence variables and convienence
22771 @c Update MI commands for heterogeneous commands.
22772 @c Update Python bindings for heterogeneous commands.
22773 @c Update gdbserver remote protocol for heterogeneous commands.
22775 @node Remote Debugging
22776 @chapter Debugging Remote Programs
22777 @cindex remote debugging
22779 If you are trying to debug a program running on a machine that cannot run
22780 @value{GDBN} in the usual way, it is often useful to use remote debugging.
22781 For example, you might use remote debugging on an operating system kernel,
22782 or on a small system which does not have a general purpose operating system
22783 powerful enough to run a full-featured debugger.
22785 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
22786 to make this work with particular debugging targets. In addition,
22787 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
22788 but not specific to any particular target system) which you can use if you
22789 write the remote stubs---the code that runs on the remote system to
22790 communicate with @value{GDBN}.
22792 Other remote targets may be available in your
22793 configuration of @value{GDBN}; use @code{help target} to list them.
22796 * Connecting:: Connecting to a remote target
22797 * File Transfer:: Sending files to a remote system
22798 * Server:: Using the gdbserver program
22799 * Remote Configuration:: Remote configuration
22800 * Remote Stub:: Implementing a remote stub
22804 @section Connecting to a Remote Target
22805 @cindex remote debugging, connecting
22806 @cindex @code{gdbserver}, connecting
22807 @cindex remote debugging, types of connections
22808 @cindex @code{gdbserver}, types of connections
22809 @cindex @code{gdbserver}, @code{target remote} mode
22810 @cindex @code{gdbserver}, @code{target extended-remote} mode
22812 This section describes how to connect to a remote target, including the
22813 types of connections and their differences, how to set up executable and
22814 symbol files on the host and target, and the commands used for
22815 connecting to and disconnecting from the remote target.
22817 @subsection Types of Remote Connections
22819 @value{GDBN} supports two types of remote connections, @code{target remote}
22820 mode and @code{target extended-remote} mode. Note that many remote targets
22821 support only @code{target remote} mode. There are several major
22822 differences between the two types of connections, enumerated here:
22826 @cindex remote debugging, detach and program exit
22827 @item Result of detach or program exit
22828 @strong{With target remote mode:} When the debugged program exits or you
22829 detach from it, @value{GDBN} disconnects from the target. When using
22830 @code{gdbserver}, @code{gdbserver} will exit.
22832 @strong{With target extended-remote mode:} When the debugged program exits or
22833 you detach from it, @value{GDBN} remains connected to the target, even
22834 though no program is running. You can rerun the program, attach to a
22835 running program, or use @code{monitor} commands specific to the target.
22837 When using @code{gdbserver} in this case, it does not exit unless it was
22838 invoked using the @option{--once} option. If the @option{--once} option
22839 was not used, you can ask @code{gdbserver} to exit using the
22840 @code{monitor exit} command (@pxref{Monitor Commands for gdbserver}).
22842 @item Specifying the program to debug
22843 For both connection types you use the @code{file} command to specify the
22844 program on the host system. If you are using @code{gdbserver} there are
22845 some differences in how to specify the location of the program on the
22848 @strong{With target remote mode:} You must either specify the program to debug
22849 on the @code{gdbserver} command line or use the @option{--attach} option
22850 (@pxref{Attaching to a program,,Attaching to a Running Program}).
22852 @cindex @option{--multi}, @code{gdbserver} option
22853 @strong{With target extended-remote mode:} You may specify the program to debug
22854 on the @code{gdbserver} command line, or you can load the program or attach
22855 to it using @value{GDBN} commands after connecting to @code{gdbserver}.
22857 @anchor{--multi Option in Types of Remote Connnections}
22858 You can start @code{gdbserver} without supplying an initial command to run
22859 or process ID to attach. To do this, use the @option{--multi} command line
22860 option. Then you can connect using @code{target extended-remote} and start
22861 the program you want to debug (see below for details on using the
22862 @code{run} command in this scenario). Note that the conditions under which
22863 @code{gdbserver} terminates depend on how @value{GDBN} connects to it
22864 (@code{target remote} or @code{target extended-remote}). The
22865 @option{--multi} option to @code{gdbserver} has no influence on that.
22867 @item The @code{run} command
22868 @strong{With target remote mode:} The @code{run} command is not
22869 supported. Once a connection has been established, you can use all
22870 the usual @value{GDBN} commands to examine and change data. The
22871 remote program is already running, so you can use commands like
22872 @kbd{step} and @kbd{continue}.
22874 @strong{With target extended-remote mode:} The @code{run} command is
22875 supported. The @code{run} command uses the value set by
22876 @code{set remote exec-file} (@pxref{set remote exec-file}) to select
22877 the program to run. Command line arguments are supported, except for
22878 wildcard expansion and I/O redirection (@pxref{Arguments}).
22880 If you specify the program to debug on the command line, then the
22881 @code{run} command is not required to start execution, and you can
22882 resume using commands like @kbd{step} and @kbd{continue} as with
22883 @code{target remote} mode.
22885 @anchor{Attaching in Types of Remote Connections}
22887 @strong{With target remote mode:} The @value{GDBN} command @code{attach} is
22888 not supported. To attach to a running program using @code{gdbserver}, you
22889 must use the @option{--attach} option (@pxref{Running gdbserver}).
22891 @strong{With target extended-remote mode:} To attach to a running program,
22892 you may use the @code{attach} command after the connection has been
22893 established. If you are using @code{gdbserver}, you may also invoke
22894 @code{gdbserver} using the @option{--attach} option
22895 (@pxref{Running gdbserver}).
22899 @anchor{Host and target files}
22900 @subsection Host and Target Files
22901 @cindex remote debugging, symbol files
22902 @cindex symbol files, remote debugging
22904 @value{GDBN}, running on the host, needs access to symbol and debugging
22905 information for your program running on the target. This requires
22906 access to an unstripped copy of your program, and possibly any associated
22907 symbol files. Note that this section applies equally to both @code{target
22908 remote} mode and @code{target extended-remote} mode.
22910 Some remote targets (@pxref{qXfer executable filename read}, and
22911 @pxref{Host I/O Packets}) allow @value{GDBN} to access program files over
22912 the same connection used to communicate with @value{GDBN}. With such a
22913 target, if the remote program is unstripped, the only command you need is
22914 @code{target remote} (or @code{target extended-remote}).
22916 If the remote program is stripped, or the target does not support remote
22917 program file access, start up @value{GDBN} using the name of the local
22918 unstripped copy of your program as the first argument, or use the
22919 @code{file} command. Use @code{set sysroot} to specify the location (on
22920 the host) of target libraries (unless your @value{GDBN} was compiled with
22921 the correct sysroot using @code{--with-sysroot}). Alternatively, you
22922 may use @code{set solib-search-path} to specify how @value{GDBN} locates
22925 The symbol file and target libraries must exactly match the executable
22926 and libraries on the target, with one exception: the files on the host
22927 system should not be stripped, even if the files on the target system
22928 are. Mismatched or missing files will lead to confusing results
22929 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
22930 files may also prevent @code{gdbserver} from debugging multi-threaded
22933 @subsection Remote Connection Commands
22934 @cindex remote connection commands
22935 @value{GDBN} can communicate with the target over a serial line, a
22936 local Unix domain socket, or
22937 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
22938 each case, @value{GDBN} uses the same protocol for debugging your
22939 program; only the medium carrying the debugging packets varies. The
22940 @code{target remote} and @code{target extended-remote} commands
22941 establish a connection to the target. Both commands accept the same
22942 arguments, which indicate the medium to use:
22946 @item target remote @var{serial-device}
22947 @itemx target extended-remote @var{serial-device}
22948 @cindex serial line, @code{target remote}
22949 Use @var{serial-device} to communicate with the target. For example,
22950 to use a serial line connected to the device named @file{/dev/ttyb}:
22953 target remote /dev/ttyb
22956 If you're using a serial line, you may want to give @value{GDBN} the
22957 @samp{--baud} option, or use the @code{set serial baud} command
22958 (@pxref{Remote Configuration, set serial baud}) before the
22959 @code{target} command.
22961 @item target remote @var{local-socket}
22962 @itemx target extended-remote @var{local-socket}
22963 @cindex local socket, @code{target remote}
22964 @cindex Unix domain socket
22965 Use @var{local-socket} to communicate with the target. For example,
22966 to use a local Unix domain socket bound to the file system entry @file{/tmp/gdb-socket0}:
22969 target remote /tmp/gdb-socket0
22972 Note that this command has the same form as the command to connect
22973 to a serial line. @value{GDBN} will automatically determine which
22974 kind of file you have specified and will make the appropriate kind
22976 This feature is not available if the host system does not support
22977 Unix domain sockets.
22979 @item target remote @code{@var{host}:@var{port}}
22980 @itemx target remote @code{@var{[host]}:@var{port}}
22981 @itemx target remote @code{tcp:@var{host}:@var{port}}
22982 @itemx target remote @code{tcp:@var{[host]}:@var{port}}
22983 @itemx target remote @code{tcp4:@var{host}:@var{port}}
22984 @itemx target remote @code{tcp6:@var{host}:@var{port}}
22985 @itemx target remote @code{tcp6:@var{[host]}:@var{port}}
22986 @itemx target extended-remote @code{@var{host}:@var{port}}
22987 @itemx target extended-remote @code{@var{[host]}:@var{port}}
22988 @itemx target extended-remote @code{tcp:@var{host}:@var{port}}
22989 @itemx target extended-remote @code{tcp:@var{[host]}:@var{port}}
22990 @itemx target extended-remote @code{tcp4:@var{host}:@var{port}}
22991 @itemx target extended-remote @code{tcp6:@var{host}:@var{port}}
22992 @itemx target extended-remote @code{tcp6:@var{[host]}:@var{port}}
22993 @cindex @acronym{TCP} port, @code{target remote}
22994 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
22995 The @var{host} may be either a host name, a numeric @acronym{IPv4}
22996 address, or a numeric @acronym{IPv6} address (with or without the
22997 square brackets to separate the address from the port); @var{port}
22998 must be a decimal number. The @var{host} could be the target machine
22999 itself, if it is directly connected to the net, or it might be a
23000 terminal server which in turn has a serial line to the target.
23002 For example, to connect to port 2828 on a terminal server named
23006 target remote manyfarms:2828
23009 To connect to port 2828 on a terminal server whose address is
23010 @code{2001:0db8:85a3:0000:0000:8a2e:0370:7334}, you can either use the
23011 square bracket syntax:
23014 target remote [2001:0db8:85a3:0000:0000:8a2e:0370:7334]:2828
23018 or explicitly specify the @acronym{IPv6} protocol:
23021 target remote tcp6:2001:0db8:85a3:0000:0000:8a2e:0370:7334:2828
23024 This last example may be confusing to the reader, because there is no
23025 visible separation between the hostname and the port number.
23026 Therefore, we recommend the user to provide @acronym{IPv6} addresses
23027 using square brackets for clarity. However, it is important to
23028 mention that for @value{GDBN} there is no ambiguity: the number after
23029 the last colon is considered to be the port number.
23031 If your remote target is actually running on the same machine as your
23032 debugger session (e.g.@: a simulator for your target running on the
23033 same host), you can omit the hostname. For example, to connect to
23034 port 1234 on your local machine:
23037 target remote :1234
23041 Note that the colon is still required here.
23043 @item target remote @code{udp:@var{host}:@var{port}}
23044 @itemx target remote @code{udp:@var{[host]}:@var{port}}
23045 @itemx target remote @code{udp4:@var{host}:@var{port}}
23046 @itemx target remote @code{udp6:@var{[host]}:@var{port}}
23047 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
23048 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
23049 @itemx target extended-remote @code{udp:@var{[host]}:@var{port}}
23050 @itemx target extended-remote @code{udp4:@var{host}:@var{port}}
23051 @itemx target extended-remote @code{udp6:@var{host}:@var{port}}
23052 @itemx target extended-remote @code{udp6:@var{[host]}:@var{port}}
23053 @cindex @acronym{UDP} port, @code{target remote}
23054 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
23055 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
23058 target remote udp:manyfarms:2828
23061 When using a @acronym{UDP} connection for remote debugging, you should
23062 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
23063 can silently drop packets on busy or unreliable networks, which will
23064 cause havoc with your debugging session.
23066 @item target remote | @var{command}
23067 @itemx target extended-remote | @var{command}
23068 @cindex pipe, @code{target remote} to
23069 Run @var{command} in the background and communicate with it using a
23070 pipe. The @var{command} is a shell command, to be parsed and expanded
23071 by the system's command shell, @code{/bin/sh}; it should expect remote
23072 protocol packets on its standard input, and send replies on its
23073 standard output. You could use this to run a stand-alone simulator
23074 that speaks the remote debugging protocol, to make net connections
23075 using programs like @code{ssh}, or for other similar tricks.
23077 If @var{command} closes its standard output (perhaps by exiting),
23078 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
23079 program has already exited, this will have no effect.)
23083 @cindex interrupting remote programs
23084 @cindex remote programs, interrupting
23085 Whenever @value{GDBN} is waiting for the remote program, if you type the
23086 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
23087 program. This may or may not succeed, depending in part on the hardware
23088 and the serial drivers the remote system uses. If you type the
23089 interrupt character once again, @value{GDBN} displays this prompt:
23092 Interrupted while waiting for the program.
23093 Give up (and stop debugging it)? (y or n)
23096 In @code{target remote} mode, if you type @kbd{y}, @value{GDBN} abandons
23097 the remote debugging session. (If you decide you want to try again later,
23098 you can use @kbd{target remote} again to connect once more.) If you type
23099 @kbd{n}, @value{GDBN} goes back to waiting.
23101 In @code{target extended-remote} mode, typing @kbd{n} will leave
23102 @value{GDBN} connected to the target.
23105 @kindex detach (remote)
23107 When you have finished debugging the remote program, you can use the
23108 @code{detach} command to release it from @value{GDBN} control.
23109 Detaching from the target normally resumes its execution, but the results
23110 will depend on your particular remote stub. After the @code{detach}
23111 command in @code{target remote} mode, @value{GDBN} is free to connect to
23112 another target. In @code{target extended-remote} mode, @value{GDBN} is
23113 still connected to the target.
23117 The @code{disconnect} command closes the connection to the target, and
23118 the target is generally not resumed. It will wait for @value{GDBN}
23119 (this instance or another one) to connect and continue debugging. After
23120 the @code{disconnect} command, @value{GDBN} is again free to connect to
23123 @cindex send command to remote monitor
23124 @cindex extend @value{GDBN} for remote targets
23125 @cindex add new commands for external monitor
23127 @item monitor @var{cmd}
23128 This command allows you to send arbitrary commands directly to the
23129 remote monitor. Since @value{GDBN} doesn't care about the commands it
23130 sends like this, this command is the way to extend @value{GDBN}---you
23131 can add new commands that only the external monitor will understand
23135 @node File Transfer
23136 @section Sending files to a remote system
23137 @cindex remote target, file transfer
23138 @cindex file transfer
23139 @cindex sending files to remote systems
23141 Some remote targets offer the ability to transfer files over the same
23142 connection used to communicate with @value{GDBN}. This is convenient
23143 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
23144 running @code{gdbserver} over a network interface. For other targets,
23145 e.g.@: embedded devices with only a single serial port, this may be
23146 the only way to upload or download files.
23148 Not all remote targets support these commands.
23152 @item remote put @var{hostfile} @var{targetfile}
23153 Copy file @var{hostfile} from the host system (the machine running
23154 @value{GDBN}) to @var{targetfile} on the target system.
23157 @item remote get @var{targetfile} @var{hostfile}
23158 Copy file @var{targetfile} from the target system to @var{hostfile}
23159 on the host system.
23161 @kindex remote delete
23162 @item remote delete @var{targetfile}
23163 Delete @var{targetfile} from the target system.
23168 @section Using the @code{gdbserver} Program
23171 @cindex remote connection without stubs
23172 @code{gdbserver} is a control program for Unix-like systems, which
23173 allows you to connect your program with a remote @value{GDBN} via
23174 @code{target remote} or @code{target extended-remote}---but without
23175 linking in the usual debugging stub.
23177 @code{gdbserver} is not a complete replacement for the debugging stubs,
23178 because it requires essentially the same operating-system facilities
23179 that @value{GDBN} itself does. In fact, a system that can run
23180 @code{gdbserver} to connect to a remote @value{GDBN} could also run
23181 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
23182 because it is a much smaller program than @value{GDBN} itself. It is
23183 also easier to port than all of @value{GDBN}, so you may be able to get
23184 started more quickly on a new system by using @code{gdbserver}.
23185 Finally, if you develop code for real-time systems, you may find that
23186 the tradeoffs involved in real-time operation make it more convenient to
23187 do as much development work as possible on another system, for example
23188 by cross-compiling. You can use @code{gdbserver} to make a similar
23189 choice for debugging.
23191 @value{GDBN} and @code{gdbserver} communicate via either a serial line
23192 or a TCP connection, using the standard @value{GDBN} remote serial
23196 @emph{Warning:} @code{gdbserver} does not have any built-in security.
23197 Do not run @code{gdbserver} connected to any public network; a
23198 @value{GDBN} connection to @code{gdbserver} provides access to the
23199 target system with the same privileges as the user running
23203 @anchor{Running gdbserver}
23204 @subsection Running @code{gdbserver}
23205 @cindex arguments, to @code{gdbserver}
23206 @cindex @code{gdbserver}, command-line arguments
23208 Run @code{gdbserver} on the target system. You need a copy of the
23209 program you want to debug, including any libraries it requires.
23210 @code{gdbserver} does not need your program's symbol table, so you can
23211 strip the program if necessary to save space. @value{GDBN} on the host
23212 system does all the symbol handling.
23214 To use the server, you must tell it how to communicate with @value{GDBN};
23215 the name of your program; and the arguments for your program. The usual
23219 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
23222 @var{comm} is either a device name (to use a serial line), or a TCP
23223 hostname and portnumber, or @code{-} or @code{stdio} to use
23224 stdin/stdout of @code{gdbserver}.
23225 For example, to debug Emacs with the argument
23226 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
23230 target> gdbserver /dev/com1 emacs foo.txt
23233 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
23236 To use a TCP connection instead of a serial line:
23239 target> gdbserver host:2345 emacs foo.txt
23242 The only difference from the previous example is the first argument,
23243 specifying that you are communicating with the host @value{GDBN} via
23244 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
23245 expect a TCP connection from machine @samp{host} to local TCP port 2345.
23246 (Currently, the @samp{host} part is ignored.) You can choose any number
23247 you want for the port number as long as it does not conflict with any
23248 TCP ports already in use on the target system (for example, @code{23} is
23249 reserved for @code{telnet}).@footnote{If you choose a port number that
23250 conflicts with another service, @code{gdbserver} prints an error message
23251 and exits.} You must use the same port number with the host @value{GDBN}
23252 @code{target remote} command.
23254 The @code{stdio} connection is useful when starting @code{gdbserver}
23258 (@value{GDBP}) target remote | ssh -T hostname gdbserver - hello
23261 The @samp{-T} option to ssh is provided because we don't need a remote pty,
23262 and we don't want escape-character handling. Ssh does this by default when
23263 a command is provided, the flag is provided to make it explicit.
23264 You could elide it if you want to.
23266 Programs started with stdio-connected gdbserver have @file{/dev/null} for
23267 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
23268 display through a pipe connected to gdbserver.
23269 Both @code{stdout} and @code{stderr} use the same pipe.
23271 @anchor{Attaching to a program}
23272 @subsubsection Attaching to a Running Program
23273 @cindex attach to a program, @code{gdbserver}
23274 @cindex @option{--attach}, @code{gdbserver} option
23276 On some targets, @code{gdbserver} can also attach to running programs.
23277 This is accomplished via the @code{--attach} argument. The syntax is:
23280 target> gdbserver --attach @var{comm} @var{pid}
23283 @var{pid} is the process ID of a currently running process. It isn't
23284 necessary to point @code{gdbserver} at a binary for the running process.
23286 In @code{target extended-remote} mode, you can also attach using the
23287 @value{GDBN} attach command
23288 (@pxref{Attaching in Types of Remote Connections}).
23291 You can debug processes by name instead of process ID if your target has the
23292 @code{pidof} utility:
23295 target> gdbserver --attach @var{comm} `pidof @var{program}`
23298 In case more than one copy of @var{program} is running, or @var{program}
23299 has multiple threads, most versions of @code{pidof} support the
23300 @code{-s} option to only return the first process ID.
23302 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
23304 This section applies only when @code{gdbserver} is run to listen on a TCP
23307 @code{gdbserver} normally terminates after all of its debugged processes have
23308 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
23309 extended-remote}, @code{gdbserver} stays running even with no processes left.
23310 @value{GDBN} normally terminates the spawned debugged process on its exit,
23311 which normally also terminates @code{gdbserver} in the @kbd{target remote}
23312 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
23313 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
23314 stays running even in the @kbd{target remote} mode.
23316 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
23317 Such reconnecting is useful for features like @ref{disconnected tracing}. For
23318 completeness, at most one @value{GDBN} can be connected at a time.
23320 @cindex @option{--once}, @code{gdbserver} option
23321 By default, @code{gdbserver} keeps the listening TCP port open, so that
23322 subsequent connections are possible. However, if you start @code{gdbserver}
23323 with the @option{--once} option, it will stop listening for any further
23324 connection attempts after connecting to the first @value{GDBN} session. This
23325 means no further connections to @code{gdbserver} will be possible after the
23326 first one. It also means @code{gdbserver} will terminate after the first
23327 connection with remote @value{GDBN} has closed, even for unexpectedly closed
23328 connections and even in the @kbd{target extended-remote} mode. The
23329 @option{--once} option allows reusing the same port number for connecting to
23330 multiple instances of @code{gdbserver} running on the same host, since each
23331 instance closes its port after the first connection.
23333 @anchor{Other Command-Line Arguments for gdbserver}
23334 @subsubsection Other Command-Line Arguments for @code{gdbserver}
23336 You can use the @option{--multi} option to start @code{gdbserver} without
23337 specifying a program to debug or a process to attach to. Then you can
23338 attach in @code{target extended-remote} mode and run or attach to a
23339 program. For more information,
23340 @pxref{--multi Option in Types of Remote Connnections}.
23342 @cindex @option{--debug}, @code{gdbserver} option
23343 The @option{--debug} option tells @code{gdbserver} to display extra
23344 status information about the debugging process.
23345 @cindex @option{--remote-debug}, @code{gdbserver} option
23346 The @option{--remote-debug} option tells @code{gdbserver} to display
23347 remote protocol debug output.
23348 @cindex @option{--debug-file}, @code{gdbserver} option
23349 @cindex @code{gdbserver}, send all debug output to a single file
23350 The @option{--debug-file=@var{filename}} option tells @code{gdbserver} to
23351 write any debug output to the given @var{filename}. These options are intended
23352 for @code{gdbserver} development and for bug reports to the developers.
23354 @cindex @option{--debug-format}, @code{gdbserver} option
23355 The @option{--debug-format=option1[,option2,...]} option tells
23356 @code{gdbserver} to include additional information in each output.
23357 Possible options are:
23361 Turn off all extra information in debugging output.
23363 Turn on all extra information in debugging output.
23365 Include a timestamp in each line of debugging output.
23368 Options are processed in order. Thus, for example, if @option{none}
23369 appears last then no additional information is added to debugging output.
23371 @cindex @option{--wrapper}, @code{gdbserver} option
23372 The @option{--wrapper} option specifies a wrapper to launch programs
23373 for debugging. The option should be followed by the name of the
23374 wrapper, then any command-line arguments to pass to the wrapper, then
23375 @kbd{--} indicating the end of the wrapper arguments.
23377 @code{gdbserver} runs the specified wrapper program with a combined
23378 command line including the wrapper arguments, then the name of the
23379 program to debug, then any arguments to the program. The wrapper
23380 runs until it executes your program, and then @value{GDBN} gains control.
23382 You can use any program that eventually calls @code{execve} with
23383 its arguments as a wrapper. Several standard Unix utilities do
23384 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
23385 with @code{exec "$@@"} will also work.
23387 For example, you can use @code{env} to pass an environment variable to
23388 the debugged program, without setting the variable in @code{gdbserver}'s
23392 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
23395 @cindex @option{--selftest}
23396 The @option{--selftest} option runs the self tests in @code{gdbserver}:
23399 $ gdbserver --selftest
23400 Ran 2 unit tests, 0 failed
23403 These tests are disabled in release.
23404 @subsection Connecting to @code{gdbserver}
23406 The basic procedure for connecting to the remote target is:
23410 Run @value{GDBN} on the host system.
23413 Make sure you have the necessary symbol files
23414 (@pxref{Host and target files}).
23415 Load symbols for your application using the @code{file} command before you
23416 connect. Use @code{set sysroot} to locate target libraries (unless your
23417 @value{GDBN} was compiled with the correct sysroot using
23418 @code{--with-sysroot}).
23421 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
23422 For TCP connections, you must start up @code{gdbserver} prior to using
23423 the @code{target} command. Otherwise you may get an error whose
23424 text depends on the host system, but which usually looks something like
23425 @samp{Connection refused}. Don't use the @code{load}
23426 command in @value{GDBN} when using @code{target remote} mode, since the
23427 program is already on the target.
23431 @anchor{Monitor Commands for gdbserver}
23432 @subsection Monitor Commands for @code{gdbserver}
23433 @cindex monitor commands, for @code{gdbserver}
23435 During a @value{GDBN} session using @code{gdbserver}, you can use the
23436 @code{monitor} command to send special requests to @code{gdbserver}.
23437 Here are the available commands.
23441 List the available monitor commands.
23443 @item monitor set debug 0
23444 @itemx monitor set debug 1
23445 Disable or enable general debugging messages.
23447 @item monitor set remote-debug 0
23448 @itemx monitor set remote-debug 1
23449 Disable or enable specific debugging messages associated with the remote
23450 protocol (@pxref{Remote Protocol}).
23452 @item monitor set debug-file filename
23453 @itemx monitor set debug-file
23454 Send any debug output to the given file, or to stderr.
23456 @item monitor set debug-format option1@r{[},option2,...@r{]}
23457 Specify additional text to add to debugging messages.
23458 Possible options are:
23462 Turn off all extra information in debugging output.
23464 Turn on all extra information in debugging output.
23466 Include a timestamp in each line of debugging output.
23469 Options are processed in order. Thus, for example, if @option{none}
23470 appears last then no additional information is added to debugging output.
23472 @item monitor set libthread-db-search-path [PATH]
23473 @cindex gdbserver, search path for @code{libthread_db}
23474 When this command is issued, @var{path} is a colon-separated list of
23475 directories to search for @code{libthread_db} (@pxref{Threads,,set
23476 libthread-db-search-path}). If you omit @var{path},
23477 @samp{libthread-db-search-path} will be reset to its default value.
23479 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
23480 not supported in @code{gdbserver}.
23483 Tell gdbserver to exit immediately. This command should be followed by
23484 @code{disconnect} to close the debugging session. @code{gdbserver} will
23485 detach from any attached processes and kill any processes it created.
23486 Use @code{monitor exit} to terminate @code{gdbserver} at the end
23487 of a multi-process mode debug session.
23491 @subsection Tracepoints support in @code{gdbserver}
23492 @cindex tracepoints support in @code{gdbserver}
23494 On some targets, @code{gdbserver} supports tracepoints, fast
23495 tracepoints and static tracepoints.
23497 For fast or static tracepoints to work, a special library called the
23498 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
23499 This library is built and distributed as an integral part of
23500 @code{gdbserver}. In addition, support for static tracepoints
23501 requires building the in-process agent library with static tracepoints
23502 support. At present, the UST (LTTng Userspace Tracer,
23503 @url{http://lttng.org/ust}) tracing engine is supported. This support
23504 is automatically available if UST development headers are found in the
23505 standard include path when @code{gdbserver} is built, or if
23506 @code{gdbserver} was explicitly configured using @option{--with-ust}
23507 to point at such headers. You can explicitly disable the support
23508 using @option{--with-ust=no}.
23510 There are several ways to load the in-process agent in your program:
23513 @item Specifying it as dependency at link time
23515 You can link your program dynamically with the in-process agent
23516 library. On most systems, this is accomplished by adding
23517 @code{-linproctrace} to the link command.
23519 @item Using the system's preloading mechanisms
23521 You can force loading the in-process agent at startup time by using
23522 your system's support for preloading shared libraries. Many Unixes
23523 support the concept of preloading user defined libraries. In most
23524 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
23525 in the environment. See also the description of @code{gdbserver}'s
23526 @option{--wrapper} command line option.
23528 @item Using @value{GDBN} to force loading the agent at run time
23530 On some systems, you can force the inferior to load a shared library,
23531 by calling a dynamic loader function in the inferior that takes care
23532 of dynamically looking up and loading a shared library. On most Unix
23533 systems, the function is @code{dlopen}. You'll use the @code{call}
23534 command for that. For example:
23537 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
23540 Note that on most Unix systems, for the @code{dlopen} function to be
23541 available, the program needs to be linked with @code{-ldl}.
23544 On systems that have a userspace dynamic loader, like most Unix
23545 systems, when you connect to @code{gdbserver} using @code{target
23546 remote}, you'll find that the program is stopped at the dynamic
23547 loader's entry point, and no shared library has been loaded in the
23548 program's address space yet, including the in-process agent. In that
23549 case, before being able to use any of the fast or static tracepoints
23550 features, you need to let the loader run and load the shared
23551 libraries. The simplest way to do that is to run the program to the
23552 main procedure. E.g., if debugging a C or C@t{++} program, start
23553 @code{gdbserver} like so:
23556 $ gdbserver :9999 myprogram
23559 Start GDB and connect to @code{gdbserver} like so, and run to main:
23563 (@value{GDBP}) target remote myhost:9999
23564 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
23565 (@value{GDBP}) b main
23566 (@value{GDBP}) continue
23569 The in-process tracing agent library should now be loaded into the
23570 process; you can confirm it with the @code{info sharedlibrary}
23571 command, which will list @file{libinproctrace.so} as loaded in the
23572 process. You are now ready to install fast tracepoints, list static
23573 tracepoint markers, probe static tracepoints markers, and start
23576 @node Remote Configuration
23577 @section Remote Configuration
23580 @kindex show remote
23581 This section documents the configuration options available when
23582 debugging remote programs. For the options related to the File I/O
23583 extensions of the remote protocol, see @ref{system,
23584 system-call-allowed}.
23587 @item set remoteaddresssize @var{bits}
23588 @cindex address size for remote targets
23589 @cindex bits in remote address
23590 Set the maximum size of address in a memory packet to the specified
23591 number of bits. @value{GDBN} will mask off the address bits above
23592 that number, when it passes addresses to the remote target. The
23593 default value is the number of bits in the target's address.
23595 @item show remoteaddresssize
23596 Show the current value of remote address size in bits.
23598 @item set serial baud @var{n}
23599 @cindex baud rate for remote targets
23600 Set the baud rate for the remote serial I/O to @var{n} baud. The
23601 value is used to set the speed of the serial port used for debugging
23604 @item show serial baud
23605 Show the current speed of the remote connection.
23607 @item set serial parity @var{parity}
23608 Set the parity for the remote serial I/O. Supported values of @var{parity} are:
23609 @code{even}, @code{none}, and @code{odd}. The default is @code{none}.
23611 @item show serial parity
23612 Show the current parity of the serial port.
23614 @item set remotebreak
23615 @cindex interrupt remote programs
23616 @cindex BREAK signal instead of Ctrl-C
23617 @anchor{set remotebreak}
23618 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
23619 when you type @kbd{Ctrl-c} to interrupt the program running
23620 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
23621 character instead. The default is off, since most remote systems
23622 expect to see @samp{Ctrl-C} as the interrupt signal.
23624 @item show remotebreak
23625 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
23626 interrupt the remote program.
23628 @item set remoteflow on
23629 @itemx set remoteflow off
23630 @kindex set remoteflow
23631 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
23632 on the serial port used to communicate to the remote target.
23634 @item show remoteflow
23635 @kindex show remoteflow
23636 Show the current setting of hardware flow control.
23638 @item set remotelogbase @var{base}
23639 Set the base (a.k.a.@: radix) of logging serial protocol
23640 communications to @var{base}. Supported values of @var{base} are:
23641 @code{ascii}, @code{octal}, and @code{hex}. The default is
23644 @item show remotelogbase
23645 Show the current setting of the radix for logging remote serial
23648 @item set remotelogfile @var{file}
23649 @cindex record serial communications on file
23650 Record remote serial communications on the named @var{file}. The
23651 default is not to record at all.
23653 @item show remotelogfile
23654 Show the current setting of the file name on which to record the
23655 serial communications.
23657 @item set remotetimeout @var{num}
23658 @cindex timeout for serial communications
23659 @cindex remote timeout
23660 Set the timeout limit to wait for the remote target to respond to
23661 @var{num} seconds. The default is 2 seconds.
23663 @item show remotetimeout
23664 Show the current number of seconds to wait for the remote target
23667 @cindex limit hardware breakpoints and watchpoints
23668 @cindex remote target, limit break- and watchpoints
23669 @anchor{set remote hardware-watchpoint-limit}
23670 @anchor{set remote hardware-breakpoint-limit}
23671 @item set remote hardware-watchpoint-limit @var{limit}
23672 @itemx set remote hardware-breakpoint-limit @var{limit}
23673 Restrict @value{GDBN} to using @var{limit} remote hardware watchpoints
23674 or breakpoints. The @var{limit} can be set to 0 to disable hardware
23675 watchpoints or breakpoints, and @code{unlimited} for unlimited
23676 watchpoints or breakpoints.
23678 @item show remote hardware-watchpoint-limit
23679 @itemx show remote hardware-breakpoint-limit
23680 Show the current limit for the number of hardware watchpoints or
23681 breakpoints that @value{GDBN} can use.
23683 @cindex limit hardware watchpoints length
23684 @cindex remote target, limit watchpoints length
23685 @anchor{set remote hardware-watchpoint-length-limit}
23686 @item set remote hardware-watchpoint-length-limit @var{limit}
23687 Restrict @value{GDBN} to using @var{limit} bytes for the maximum
23688 length of a remote hardware watchpoint. A @var{limit} of 0 disables
23689 hardware watchpoints and @code{unlimited} allows watchpoints of any
23692 @item show remote hardware-watchpoint-length-limit
23693 Show the current limit (in bytes) of the maximum length of
23694 a remote hardware watchpoint.
23696 @item set remote exec-file @var{filename}
23697 @itemx show remote exec-file
23698 @anchor{set remote exec-file}
23699 @cindex executable file, for remote target
23700 Select the file used for @code{run} with @code{target
23701 extended-remote}. This should be set to a filename valid on the
23702 target system. If it is not set, the target will use a default
23703 filename (e.g.@: the last program run).
23705 @item set remote interrupt-sequence
23706 @cindex interrupt remote programs
23707 @cindex select Ctrl-C, BREAK or BREAK-g
23708 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
23709 @samp{BREAK-g} as the
23710 sequence to the remote target in order to interrupt the execution.
23711 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
23712 is high level of serial line for some certain time.
23713 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
23714 It is @code{BREAK} signal followed by character @code{g}.
23716 @item show interrupt-sequence
23717 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
23718 is sent by @value{GDBN} to interrupt the remote program.
23719 @code{BREAK-g} is BREAK signal followed by @code{g} and
23720 also known as Magic SysRq g.
23722 @item set remote interrupt-on-connect
23723 @cindex send interrupt-sequence on start
23724 Specify whether interrupt-sequence is sent to remote target when
23725 @value{GDBN} connects to it. This is mostly needed when you debug
23726 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
23727 which is known as Magic SysRq g in order to connect @value{GDBN}.
23729 @item show interrupt-on-connect
23730 Show whether interrupt-sequence is sent
23731 to remote target when @value{GDBN} connects to it.
23735 @item set tcp auto-retry on
23736 @cindex auto-retry, for remote TCP target
23737 Enable auto-retry for remote TCP connections. This is useful if the remote
23738 debugging agent is launched in parallel with @value{GDBN}; there is a race
23739 condition because the agent may not become ready to accept the connection
23740 before @value{GDBN} attempts to connect. When auto-retry is
23741 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
23742 to establish the connection using the timeout specified by
23743 @code{set tcp connect-timeout}.
23745 @item set tcp auto-retry off
23746 Do not auto-retry failed TCP connections.
23748 @item show tcp auto-retry
23749 Show the current auto-retry setting.
23751 @item set tcp connect-timeout @var{seconds}
23752 @itemx set tcp connect-timeout unlimited
23753 @cindex connection timeout, for remote TCP target
23754 @cindex timeout, for remote target connection
23755 Set the timeout for establishing a TCP connection to the remote target to
23756 @var{seconds}. The timeout affects both polling to retry failed connections
23757 (enabled by @code{set tcp auto-retry on}) and waiting for connections
23758 that are merely slow to complete, and represents an approximate cumulative
23759 value. If @var{seconds} is @code{unlimited}, there is no timeout and
23760 @value{GDBN} will keep attempting to establish a connection forever,
23761 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
23763 @item show tcp connect-timeout
23764 Show the current connection timeout setting.
23767 @cindex remote packets, enabling and disabling
23768 The @value{GDBN} remote protocol autodetects the packets supported by
23769 your debugging stub. If you need to override the autodetection, you
23770 can use these commands to enable or disable individual packets. Each
23771 packet can be set to @samp{on} (the remote target supports this
23772 packet), @samp{off} (the remote target does not support this packet),
23773 or @samp{auto} (detect remote target support for this packet). They
23774 all default to @samp{auto}. For more information about each packet,
23775 see @ref{Remote Protocol}.
23777 During normal use, you should not have to use any of these commands.
23778 If you do, that may be a bug in your remote debugging stub, or a bug
23779 in @value{GDBN}. You may want to report the problem to the
23780 @value{GDBN} developers.
23782 For each packet @var{name}, the command to enable or disable the
23783 packet is @code{set remote @var{name}-packet}. The available settings
23786 @multitable @columnfractions 0.28 0.32 0.25
23789 @tab Related Features
23791 @item @code{fetch-register}
23793 @tab @code{info registers}
23795 @item @code{set-register}
23799 @item @code{binary-download}
23801 @tab @code{load}, @code{set}
23803 @item @code{read-aux-vector}
23804 @tab @code{qXfer:auxv:read}
23805 @tab @code{info auxv}
23807 @item @code{symbol-lookup}
23808 @tab @code{qSymbol}
23809 @tab Detecting multiple threads
23811 @item @code{attach}
23812 @tab @code{vAttach}
23815 @item @code{verbose-resume}
23817 @tab Stepping or resuming multiple threads
23823 @item @code{software-breakpoint}
23827 @item @code{hardware-breakpoint}
23831 @item @code{write-watchpoint}
23835 @item @code{read-watchpoint}
23839 @item @code{access-watchpoint}
23843 @item @code{pid-to-exec-file}
23844 @tab @code{qXfer:exec-file:read}
23845 @tab @code{attach}, @code{run}
23847 @item @code{target-features}
23848 @tab @code{qXfer:features:read}
23849 @tab @code{set architecture}
23851 @item @code{library-info}
23852 @tab @code{qXfer:libraries:read}
23853 @tab @code{info sharedlibrary}
23855 @item @code{memory-map}
23856 @tab @code{qXfer:memory-map:read}
23857 @tab @code{info mem}
23859 @item @code{read-sdata-object}
23860 @tab @code{qXfer:sdata:read}
23861 @tab @code{print $_sdata}
23863 @item @code{read-siginfo-object}
23864 @tab @code{qXfer:siginfo:read}
23865 @tab @code{print $_siginfo}
23867 @item @code{write-siginfo-object}
23868 @tab @code{qXfer:siginfo:write}
23869 @tab @code{set $_siginfo}
23871 @item @code{threads}
23872 @tab @code{qXfer:threads:read}
23873 @tab @code{info threads}
23875 @item @code{get-thread-local-@*storage-address}
23876 @tab @code{qGetTLSAddr}
23877 @tab Displaying @code{__thread} variables
23879 @item @code{get-thread-information-block-address}
23880 @tab @code{qGetTIBAddr}
23881 @tab Display MS-Windows Thread Information Block.
23883 @item @code{search-memory}
23884 @tab @code{qSearch:memory}
23887 @item @code{supported-packets}
23888 @tab @code{qSupported}
23889 @tab Remote communications parameters
23891 @item @code{catch-syscalls}
23892 @tab @code{QCatchSyscalls}
23893 @tab @code{catch syscall}
23895 @item @code{pass-signals}
23896 @tab @code{QPassSignals}
23897 @tab @code{handle @var{signal}}
23899 @item @code{program-signals}
23900 @tab @code{QProgramSignals}
23901 @tab @code{handle @var{signal}}
23903 @item @code{hostio-close-packet}
23904 @tab @code{vFile:close}
23905 @tab @code{remote get}, @code{remote put}
23907 @item @code{hostio-open-packet}
23908 @tab @code{vFile:open}
23909 @tab @code{remote get}, @code{remote put}
23911 @item @code{hostio-pread-packet}
23912 @tab @code{vFile:pread}
23913 @tab @code{remote get}, @code{remote put}
23915 @item @code{hostio-pwrite-packet}
23916 @tab @code{vFile:pwrite}
23917 @tab @code{remote get}, @code{remote put}
23919 @item @code{hostio-unlink-packet}
23920 @tab @code{vFile:unlink}
23921 @tab @code{remote delete}
23923 @item @code{hostio-readlink-packet}
23924 @tab @code{vFile:readlink}
23927 @item @code{hostio-fstat-packet}
23928 @tab @code{vFile:fstat}
23931 @item @code{hostio-setfs-packet}
23932 @tab @code{vFile:setfs}
23935 @item @code{noack-packet}
23936 @tab @code{QStartNoAckMode}
23937 @tab Packet acknowledgment
23939 @item @code{osdata}
23940 @tab @code{qXfer:osdata:read}
23941 @tab @code{info os}
23943 @item @code{query-attached}
23944 @tab @code{qAttached}
23945 @tab Querying remote process attach state.
23947 @item @code{trace-buffer-size}
23948 @tab @code{QTBuffer:size}
23949 @tab @code{set trace-buffer-size}
23951 @item @code{trace-status}
23952 @tab @code{qTStatus}
23953 @tab @code{tstatus}
23955 @item @code{traceframe-info}
23956 @tab @code{qXfer:traceframe-info:read}
23957 @tab Traceframe info
23959 @item @code{install-in-trace}
23960 @tab @code{InstallInTrace}
23961 @tab Install tracepoint in tracing
23963 @item @code{disable-randomization}
23964 @tab @code{QDisableRandomization}
23965 @tab @code{set disable-randomization}
23967 @item @code{startup-with-shell}
23968 @tab @code{QStartupWithShell}
23969 @tab @code{set startup-with-shell}
23971 @item @code{environment-hex-encoded}
23972 @tab @code{QEnvironmentHexEncoded}
23973 @tab @code{set environment}
23975 @item @code{environment-unset}
23976 @tab @code{QEnvironmentUnset}
23977 @tab @code{unset environment}
23979 @item @code{environment-reset}
23980 @tab @code{QEnvironmentReset}
23981 @tab @code{Reset the inferior environment (i.e., unset user-set variables)}
23983 @item @code{set-working-dir}
23984 @tab @code{QSetWorkingDir}
23985 @tab @code{set cwd}
23987 @item @code{conditional-breakpoints-packet}
23988 @tab @code{Z0 and Z1}
23989 @tab @code{Support for target-side breakpoint condition evaluation}
23991 @item @code{multiprocess-extensions}
23992 @tab @code{multiprocess extensions}
23993 @tab Debug multiple processes and remote process PID awareness
23995 @item @code{swbreak-feature}
23996 @tab @code{swbreak stop reason}
23999 @item @code{hwbreak-feature}
24000 @tab @code{hwbreak stop reason}
24003 @item @code{fork-event-feature}
24004 @tab @code{fork stop reason}
24007 @item @code{vfork-event-feature}
24008 @tab @code{vfork stop reason}
24011 @item @code{exec-event-feature}
24012 @tab @code{exec stop reason}
24015 @item @code{thread-events}
24016 @tab @code{QThreadEvents}
24017 @tab Tracking thread lifetime.
24019 @item @code{no-resumed-stop-reply}
24020 @tab @code{no resumed thread left stop reply}
24021 @tab Tracking thread lifetime.
24026 @section Implementing a Remote Stub
24028 @cindex debugging stub, example
24029 @cindex remote stub, example
24030 @cindex stub example, remote debugging
24031 The stub files provided with @value{GDBN} implement the target side of the
24032 communication protocol, and the @value{GDBN} side is implemented in the
24033 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
24034 these subroutines to communicate, and ignore the details. (If you're
24035 implementing your own stub file, you can still ignore the details: start
24036 with one of the existing stub files. @file{sparc-stub.c} is the best
24037 organized, and therefore the easiest to read.)
24039 @cindex remote serial debugging, overview
24040 To debug a program running on another machine (the debugging
24041 @dfn{target} machine), you must first arrange for all the usual
24042 prerequisites for the program to run by itself. For example, for a C
24047 A startup routine to set up the C runtime environment; these usually
24048 have a name like @file{crt0}. The startup routine may be supplied by
24049 your hardware supplier, or you may have to write your own.
24052 A C subroutine library to support your program's
24053 subroutine calls, notably managing input and output.
24056 A way of getting your program to the other machine---for example, a
24057 download program. These are often supplied by the hardware
24058 manufacturer, but you may have to write your own from hardware
24062 The next step is to arrange for your program to use a serial port to
24063 communicate with the machine where @value{GDBN} is running (the @dfn{host}
24064 machine). In general terms, the scheme looks like this:
24068 @value{GDBN} already understands how to use this protocol; when everything
24069 else is set up, you can simply use the @samp{target remote} command
24070 (@pxref{Targets,,Specifying a Debugging Target}).
24072 @item On the target,
24073 you must link with your program a few special-purpose subroutines that
24074 implement the @value{GDBN} remote serial protocol. The file containing these
24075 subroutines is called a @dfn{debugging stub}.
24077 On certain remote targets, you can use an auxiliary program
24078 @code{gdbserver} instead of linking a stub into your program.
24079 @xref{Server,,Using the @code{gdbserver} Program}, for details.
24082 The debugging stub is specific to the architecture of the remote
24083 machine; for example, use @file{sparc-stub.c} to debug programs on
24086 @cindex remote serial stub list
24087 These working remote stubs are distributed with @value{GDBN}:
24092 @cindex @file{i386-stub.c}
24095 For Intel 386 and compatible architectures.
24098 @cindex @file{m68k-stub.c}
24099 @cindex Motorola 680x0
24101 For Motorola 680x0 architectures.
24104 @cindex @file{sh-stub.c}
24107 For Renesas SH architectures.
24110 @cindex @file{sparc-stub.c}
24112 For @sc{sparc} architectures.
24114 @item sparcl-stub.c
24115 @cindex @file{sparcl-stub.c}
24118 For Fujitsu @sc{sparclite} architectures.
24122 The @file{README} file in the @value{GDBN} distribution may list other
24123 recently added stubs.
24126 * Stub Contents:: What the stub can do for you
24127 * Bootstrapping:: What you must do for the stub
24128 * Debug Session:: Putting it all together
24131 @node Stub Contents
24132 @subsection What the Stub Can Do for You
24134 @cindex remote serial stub
24135 The debugging stub for your architecture supplies these three
24139 @item set_debug_traps
24140 @findex set_debug_traps
24141 @cindex remote serial stub, initialization
24142 This routine arranges for @code{handle_exception} to run when your
24143 program stops. You must call this subroutine explicitly in your
24144 program's startup code.
24146 @item handle_exception
24147 @findex handle_exception
24148 @cindex remote serial stub, main routine
24149 This is the central workhorse, but your program never calls it
24150 explicitly---the setup code arranges for @code{handle_exception} to
24151 run when a trap is triggered.
24153 @code{handle_exception} takes control when your program stops during
24154 execution (for example, on a breakpoint), and mediates communications
24155 with @value{GDBN} on the host machine. This is where the communications
24156 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
24157 representative on the target machine. It begins by sending summary
24158 information on the state of your program, then continues to execute,
24159 retrieving and transmitting any information @value{GDBN} needs, until you
24160 execute a @value{GDBN} command that makes your program resume; at that point,
24161 @code{handle_exception} returns control to your own code on the target
24165 @cindex @code{breakpoint} subroutine, remote
24166 Use this auxiliary subroutine to make your program contain a
24167 breakpoint. Depending on the particular situation, this may be the only
24168 way for @value{GDBN} to get control. For instance, if your target
24169 machine has some sort of interrupt button, you won't need to call this;
24170 pressing the interrupt button transfers control to
24171 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
24172 simply receiving characters on the serial port may also trigger a trap;
24173 again, in that situation, you don't need to call @code{breakpoint} from
24174 your own program---simply running @samp{target remote} from the host
24175 @value{GDBN} session gets control.
24177 Call @code{breakpoint} if none of these is true, or if you simply want
24178 to make certain your program stops at a predetermined point for the
24179 start of your debugging session.
24182 @node Bootstrapping
24183 @subsection What You Must Do for the Stub
24185 @cindex remote stub, support routines
24186 The debugging stubs that come with @value{GDBN} are set up for a particular
24187 chip architecture, but they have no information about the rest of your
24188 debugging target machine.
24190 First of all you need to tell the stub how to communicate with the
24194 @item int getDebugChar()
24195 @findex getDebugChar
24196 Write this subroutine to read a single character from the serial port.
24197 It may be identical to @code{getchar} for your target system; a
24198 different name is used to allow you to distinguish the two if you wish.
24200 @item void putDebugChar(int)
24201 @findex putDebugChar
24202 Write this subroutine to write a single character to the serial port.
24203 It may be identical to @code{putchar} for your target system; a
24204 different name is used to allow you to distinguish the two if you wish.
24207 @cindex control C, and remote debugging
24208 @cindex interrupting remote targets
24209 If you want @value{GDBN} to be able to stop your program while it is
24210 running, you need to use an interrupt-driven serial driver, and arrange
24211 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
24212 character). That is the character which @value{GDBN} uses to tell the
24213 remote system to stop.
24215 Getting the debugging target to return the proper status to @value{GDBN}
24216 probably requires changes to the standard stub; one quick and dirty way
24217 is to just execute a breakpoint instruction (the ``dirty'' part is that
24218 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
24220 Other routines you need to supply are:
24223 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
24224 @findex exceptionHandler
24225 Write this function to install @var{exception_address} in the exception
24226 handling tables. You need to do this because the stub does not have any
24227 way of knowing what the exception handling tables on your target system
24228 are like (for example, the processor's table might be in @sc{rom},
24229 containing entries which point to a table in @sc{ram}).
24230 The @var{exception_number} specifies the exception which should be changed;
24231 its meaning is architecture-dependent (for example, different numbers
24232 might represent divide by zero, misaligned access, etc). When this
24233 exception occurs, control should be transferred directly to
24234 @var{exception_address}, and the processor state (stack, registers,
24235 and so on) should be just as it is when a processor exception occurs. So if
24236 you want to use a jump instruction to reach @var{exception_address}, it
24237 should be a simple jump, not a jump to subroutine.
24239 For the 386, @var{exception_address} should be installed as an interrupt
24240 gate so that interrupts are masked while the handler runs. The gate
24241 should be at privilege level 0 (the most privileged level). The
24242 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
24243 help from @code{exceptionHandler}.
24245 @item void flush_i_cache()
24246 @findex flush_i_cache
24247 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
24248 instruction cache, if any, on your target machine. If there is no
24249 instruction cache, this subroutine may be a no-op.
24251 On target machines that have instruction caches, @value{GDBN} requires this
24252 function to make certain that the state of your program is stable.
24256 You must also make sure this library routine is available:
24259 @item void *memset(void *, int, int)
24261 This is the standard library function @code{memset} that sets an area of
24262 memory to a known value. If you have one of the free versions of
24263 @code{libc.a}, @code{memset} can be found there; otherwise, you must
24264 either obtain it from your hardware manufacturer, or write your own.
24267 If you do not use the GNU C compiler, you may need other standard
24268 library subroutines as well; this varies from one stub to another,
24269 but in general the stubs are likely to use any of the common library
24270 subroutines which @code{@value{NGCC}} generates as inline code.
24273 @node Debug Session
24274 @subsection Putting it All Together
24276 @cindex remote serial debugging summary
24277 In summary, when your program is ready to debug, you must follow these
24282 Make sure you have defined the supporting low-level routines
24283 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
24285 @code{getDebugChar}, @code{putDebugChar},
24286 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
24290 Insert these lines in your program's startup code, before the main
24291 procedure is called:
24298 On some machines, when a breakpoint trap is raised, the hardware
24299 automatically makes the PC point to the instruction after the
24300 breakpoint. If your machine doesn't do that, you may need to adjust
24301 @code{handle_exception} to arrange for it to return to the instruction
24302 after the breakpoint on this first invocation, so that your program
24303 doesn't keep hitting the initial breakpoint instead of making
24307 For the 680x0 stub only, you need to provide a variable called
24308 @code{exceptionHook}. Normally you just use:
24311 void (*exceptionHook)() = 0;
24315 but if before calling @code{set_debug_traps}, you set it to point to a
24316 function in your program, that function is called when
24317 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
24318 error). The function indicated by @code{exceptionHook} is called with
24319 one parameter: an @code{int} which is the exception number.
24322 Compile and link together: your program, the @value{GDBN} debugging stub for
24323 your target architecture, and the supporting subroutines.
24326 Make sure you have a serial connection between your target machine and
24327 the @value{GDBN} host, and identify the serial port on the host.
24330 @c The "remote" target now provides a `load' command, so we should
24331 @c document that. FIXME.
24332 Download your program to your target machine (or get it there by
24333 whatever means the manufacturer provides), and start it.
24336 Start @value{GDBN} on the host, and connect to the target
24337 (@pxref{Connecting,,Connecting to a Remote Target}).
24341 @node Configurations
24342 @chapter Configuration-Specific Information
24344 While nearly all @value{GDBN} commands are available for all native and
24345 cross versions of the debugger, there are some exceptions. This chapter
24346 describes things that are only available in certain configurations.
24348 There are three major categories of configurations: native
24349 configurations, where the host and target are the same, embedded
24350 operating system configurations, which are usually the same for several
24351 different processor architectures, and bare embedded processors, which
24352 are quite different from each other.
24357 * Embedded Processors::
24364 This section describes details specific to particular native
24368 * BSD libkvm Interface:: Debugging BSD kernel memory images
24369 * Process Information:: Process information
24370 * DJGPP Native:: Features specific to the DJGPP port
24371 * Cygwin Native:: Features specific to the Cygwin port
24372 * Hurd Native:: Features specific to @sc{gnu} Hurd
24373 * Darwin:: Features specific to Darwin
24374 * FreeBSD:: Features specific to FreeBSD
24377 @node BSD libkvm Interface
24378 @subsection BSD libkvm Interface
24381 @cindex kernel memory image
24382 @cindex kernel crash dump
24384 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
24385 interface that provides a uniform interface for accessing kernel virtual
24386 memory images, including live systems and crash dumps. @value{GDBN}
24387 uses this interface to allow you to debug live kernels and kernel crash
24388 dumps on many native BSD configurations. This is implemented as a
24389 special @code{kvm} debugging target. For debugging a live system, load
24390 the currently running kernel into @value{GDBN} and connect to the
24394 (@value{GDBP}) @b{target kvm}
24397 For debugging crash dumps, provide the file name of the crash dump as an
24401 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
24404 Once connected to the @code{kvm} target, the following commands are
24410 Set current context from the @dfn{Process Control Block} (PCB) address.
24413 Set current context from proc address. This command isn't available on
24414 modern FreeBSD systems.
24417 @node Process Information
24418 @subsection Process Information
24420 @cindex examine process image
24421 @cindex process info via @file{/proc}
24423 Some operating systems provide interfaces to fetch additional
24424 information about running processes beyond memory and per-thread
24425 register state. If @value{GDBN} is configured for an operating system
24426 with a supported interface, the command @code{info proc} is available
24427 to report information about the process running your program, or about
24428 any process running on your system.
24430 One supported interface is a facility called @samp{/proc} that can be
24431 used to examine the image of a running process using file-system
24432 subroutines. This facility is supported on @sc{gnu}/Linux and Solaris
24435 On FreeBSD systems, system control nodes are used to query process
24438 In addition, some systems may provide additional process information
24439 in core files. Note that a core file may include a subset of the
24440 information available from a live process. Process information is
24441 currently available from cores created on @sc{gnu}/Linux and FreeBSD
24448 @itemx info proc @var{process-id}
24449 Summarize available information about a process. If a
24450 process ID is specified by @var{process-id}, display information about
24451 that process; otherwise display information about the program being
24452 debugged. The summary includes the debugged process ID, the command
24453 line used to invoke it, its current working directory, and its
24454 executable file's absolute file name.
24456 On some systems, @var{process-id} can be of the form
24457 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
24458 within a process. If the optional @var{pid} part is missing, it means
24459 a thread from the process being debugged (the leading @samp{/} still
24460 needs to be present, or else @value{GDBN} will interpret the number as
24461 a process ID rather than a thread ID).
24463 @item info proc cmdline
24464 @cindex info proc cmdline
24465 Show the original command line of the process. This command is
24466 supported on @sc{gnu}/Linux and FreeBSD.
24468 @item info proc cwd
24469 @cindex info proc cwd
24470 Show the current working directory of the process. This command is
24471 supported on @sc{gnu}/Linux and FreeBSD.
24473 @item info proc exe
24474 @cindex info proc exe
24475 Show the name of executable of the process. This command is supported
24476 on @sc{gnu}/Linux and FreeBSD.
24478 @item info proc files
24479 @cindex info proc files
24480 Show the file descriptors open by the process. For each open file
24481 descriptor, @value{GDBN} shows its number, type (file, directory,
24482 character device, socket), file pointer offset, and the name of the
24483 resource open on the descriptor. The resource name can be a file name
24484 (for files, directories, and devices) or a protocol followed by socket
24485 address (for network connections). This command is supported on
24488 This example shows the open file descriptors for a process using a
24489 tty for standard input and output as well as two network sockets:
24492 (@value{GDBP}) info proc files 22136
24496 FD Type Offset Flags Name
24497 text file - r-------- /usr/bin/ssh
24498 ctty chr - rw------- /dev/pts/20
24499 cwd dir - r-------- /usr/home/john
24500 root dir - r-------- /
24501 0 chr 0x32933a4 rw------- /dev/pts/20
24502 1 chr 0x32933a4 rw------- /dev/pts/20
24503 2 chr 0x32933a4 rw------- /dev/pts/20
24504 3 socket 0x0 rw----n-- tcp4 10.0.1.2:53014 -> 10.0.1.10:22
24505 4 socket 0x0 rw------- unix stream:/tmp/ssh-FIt89oAzOn5f/agent.2456
24508 @item info proc mappings
24509 @cindex memory address space mappings
24510 Report the memory address space ranges accessible in a process. On
24511 Solaris and FreeBSD systems, each memory range includes information on
24512 whether the process has read, write, or execute access rights to each
24513 range. On @sc{gnu}/Linux and FreeBSD systems, each memory range
24514 includes the object file which is mapped to that range.
24516 @item info proc stat
24517 @itemx info proc status
24518 @cindex process detailed status information
24519 Show additional process-related information, including the user ID and
24520 group ID; virtual memory usage; the signals that are pending, blocked,
24521 and ignored; its TTY; its consumption of system and user time; its
24522 stack size; its @samp{nice} value; etc. These commands are supported
24523 on @sc{gnu}/Linux and FreeBSD.
24525 For @sc{gnu}/Linux systems, see the @samp{proc} man page for more
24526 information (type @kbd{man 5 proc} from your shell prompt).
24528 For FreeBSD systems, @code{info proc stat} is an alias for @code{info
24531 @item info proc all
24532 Show all the information about the process described under all of the
24533 above @code{info proc} subcommands.
24536 @comment These sub-options of 'info proc' were not included when
24537 @comment procfs.c was re-written. Keep their descriptions around
24538 @comment against the day when someone finds the time to put them back in.
24539 @kindex info proc times
24540 @item info proc times
24541 Starting time, user CPU time, and system CPU time for your program and
24544 @kindex info proc id
24546 Report on the process IDs related to your program: its own process ID,
24547 the ID of its parent, the process group ID, and the session ID.
24550 @item set procfs-trace
24551 @kindex set procfs-trace
24552 @cindex @code{procfs} API calls
24553 This command enables and disables tracing of @code{procfs} API calls.
24555 @item show procfs-trace
24556 @kindex show procfs-trace
24557 Show the current state of @code{procfs} API call tracing.
24559 @item set procfs-file @var{file}
24560 @kindex set procfs-file
24561 Tell @value{GDBN} to write @code{procfs} API trace to the named
24562 @var{file}. @value{GDBN} appends the trace info to the previous
24563 contents of the file. The default is to display the trace on the
24566 @item show procfs-file
24567 @kindex show procfs-file
24568 Show the file to which @code{procfs} API trace is written.
24570 @item proc-trace-entry
24571 @itemx proc-trace-exit
24572 @itemx proc-untrace-entry
24573 @itemx proc-untrace-exit
24574 @kindex proc-trace-entry
24575 @kindex proc-trace-exit
24576 @kindex proc-untrace-entry
24577 @kindex proc-untrace-exit
24578 These commands enable and disable tracing of entries into and exits
24579 from the @code{syscall} interface.
24582 @kindex info pidlist
24583 @cindex process list, QNX Neutrino
24584 For QNX Neutrino only, this command displays the list of all the
24585 processes and all the threads within each process.
24588 @kindex info meminfo
24589 @cindex mapinfo list, QNX Neutrino
24590 For QNX Neutrino only, this command displays the list of all mapinfos.
24594 @subsection Features for Debugging @sc{djgpp} Programs
24595 @cindex @sc{djgpp} debugging
24596 @cindex native @sc{djgpp} debugging
24597 @cindex MS-DOS-specific commands
24600 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
24601 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
24602 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
24603 top of real-mode DOS systems and their emulations.
24605 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
24606 defines a few commands specific to the @sc{djgpp} port. This
24607 subsection describes those commands.
24612 This is a prefix of @sc{djgpp}-specific commands which print
24613 information about the target system and important OS structures.
24616 @cindex MS-DOS system info
24617 @cindex free memory information (MS-DOS)
24618 @item info dos sysinfo
24619 This command displays assorted information about the underlying
24620 platform: the CPU type and features, the OS version and flavor, the
24621 DPMI version, and the available conventional and DPMI memory.
24626 @cindex segment descriptor tables
24627 @cindex descriptor tables display
24629 @itemx info dos ldt
24630 @itemx info dos idt
24631 These 3 commands display entries from, respectively, Global, Local,
24632 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
24633 tables are data structures which store a descriptor for each segment
24634 that is currently in use. The segment's selector is an index into a
24635 descriptor table; the table entry for that index holds the
24636 descriptor's base address and limit, and its attributes and access
24639 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
24640 segment (used for both data and the stack), and a DOS segment (which
24641 allows access to DOS/BIOS data structures and absolute addresses in
24642 conventional memory). However, the DPMI host will usually define
24643 additional segments in order to support the DPMI environment.
24645 @cindex garbled pointers
24646 These commands allow to display entries from the descriptor tables.
24647 Without an argument, all entries from the specified table are
24648 displayed. An argument, which should be an integer expression, means
24649 display a single entry whose index is given by the argument. For
24650 example, here's a convenient way to display information about the
24651 debugged program's data segment:
24654 @exdent @code{(@value{GDBP}) info dos ldt $ds}
24655 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
24659 This comes in handy when you want to see whether a pointer is outside
24660 the data segment's limit (i.e.@: @dfn{garbled}).
24662 @cindex page tables display (MS-DOS)
24664 @itemx info dos pte
24665 These two commands display entries from, respectively, the Page
24666 Directory and the Page Tables. Page Directories and Page Tables are
24667 data structures which control how virtual memory addresses are mapped
24668 into physical addresses. A Page Table includes an entry for every
24669 page of memory that is mapped into the program's address space; there
24670 may be several Page Tables, each one holding up to 4096 entries. A
24671 Page Directory has up to 4096 entries, one each for every Page Table
24672 that is currently in use.
24674 Without an argument, @kbd{info dos pde} displays the entire Page
24675 Directory, and @kbd{info dos pte} displays all the entries in all of
24676 the Page Tables. An argument, an integer expression, given to the
24677 @kbd{info dos pde} command means display only that entry from the Page
24678 Directory table. An argument given to the @kbd{info dos pte} command
24679 means display entries from a single Page Table, the one pointed to by
24680 the specified entry in the Page Directory.
24682 @cindex direct memory access (DMA) on MS-DOS
24683 These commands are useful when your program uses @dfn{DMA} (Direct
24684 Memory Access), which needs physical addresses to program the DMA
24687 These commands are supported only with some DPMI servers.
24689 @cindex physical address from linear address
24690 @item info dos address-pte @var{addr}
24691 This command displays the Page Table entry for a specified linear
24692 address. The argument @var{addr} is a linear address which should
24693 already have the appropriate segment's base address added to it,
24694 because this command accepts addresses which may belong to @emph{any}
24695 segment. For example, here's how to display the Page Table entry for
24696 the page where a variable @code{i} is stored:
24699 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
24700 @exdent @code{Page Table entry for address 0x11a00d30:}
24701 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
24705 This says that @code{i} is stored at offset @code{0xd30} from the page
24706 whose physical base address is @code{0x02698000}, and shows all the
24707 attributes of that page.
24709 Note that you must cast the addresses of variables to a @code{char *},
24710 since otherwise the value of @code{__djgpp_base_address}, the base
24711 address of all variables and functions in a @sc{djgpp} program, will
24712 be added using the rules of C pointer arithmetics: if @code{i} is
24713 declared an @code{int}, @value{GDBN} will add 4 times the value of
24714 @code{__djgpp_base_address} to the address of @code{i}.
24716 Here's another example, it displays the Page Table entry for the
24720 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
24721 @exdent @code{Page Table entry for address 0x29110:}
24722 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
24726 (The @code{+ 3} offset is because the transfer buffer's address is the
24727 3rd member of the @code{_go32_info_block} structure.) The output
24728 clearly shows that this DPMI server maps the addresses in conventional
24729 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
24730 linear (@code{0x29110}) addresses are identical.
24732 This command is supported only with some DPMI servers.
24735 @cindex DOS serial data link, remote debugging
24736 In addition to native debugging, the DJGPP port supports remote
24737 debugging via a serial data link. The following commands are specific
24738 to remote serial debugging in the DJGPP port of @value{GDBN}.
24741 @kindex set com1base
24742 @kindex set com1irq
24743 @kindex set com2base
24744 @kindex set com2irq
24745 @kindex set com3base
24746 @kindex set com3irq
24747 @kindex set com4base
24748 @kindex set com4irq
24749 @item set com1base @var{addr}
24750 This command sets the base I/O port address of the @file{COM1} serial
24753 @item set com1irq @var{irq}
24754 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
24755 for the @file{COM1} serial port.
24757 There are similar commands @samp{set com2base}, @samp{set com3irq},
24758 etc.@: for setting the port address and the @code{IRQ} lines for the
24761 @kindex show com1base
24762 @kindex show com1irq
24763 @kindex show com2base
24764 @kindex show com2irq
24765 @kindex show com3base
24766 @kindex show com3irq
24767 @kindex show com4base
24768 @kindex show com4irq
24769 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
24770 display the current settings of the base address and the @code{IRQ}
24771 lines used by the COM ports.
24774 @kindex info serial
24775 @cindex DOS serial port status
24776 This command prints the status of the 4 DOS serial ports. For each
24777 port, it prints whether it's active or not, its I/O base address and
24778 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
24779 counts of various errors encountered so far.
24783 @node Cygwin Native
24784 @subsection Features for Debugging MS Windows PE Executables
24785 @cindex MS Windows debugging
24786 @cindex native Cygwin debugging
24787 @cindex Cygwin-specific commands
24789 @value{GDBN} supports native debugging of MS Windows programs, including
24790 DLLs with and without symbolic debugging information.
24792 @cindex Ctrl-BREAK, MS-Windows
24793 @cindex interrupt debuggee on MS-Windows
24794 MS-Windows programs that call @code{SetConsoleMode} to switch off the
24795 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
24796 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
24797 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
24798 sequence, which can be used to interrupt the debuggee even if it
24801 There are various additional Cygwin-specific commands, described in
24802 this section. Working with DLLs that have no debugging symbols is
24803 described in @ref{Non-debug DLL Symbols}.
24808 This is a prefix of MS Windows-specific commands which print
24809 information about the target system and important OS structures.
24811 @item info w32 selector
24812 This command displays information returned by
24813 the Win32 API @code{GetThreadSelectorEntry} function.
24814 It takes an optional argument that is evaluated to
24815 a long value to give the information about this given selector.
24816 Without argument, this command displays information
24817 about the six segment registers.
24819 @item info w32 thread-information-block
24820 This command displays thread specific information stored in the
24821 Thread Information Block (readable on the X86 CPU family using @code{$fs}
24822 selector for 32-bit programs and @code{$gs} for 64-bit programs).
24824 @kindex signal-event
24825 @item signal-event @var{id}
24826 This command signals an event with user-provided @var{id}. Used to resume
24827 crashing process when attached to it using MS-Windows JIT debugging (AeDebug).
24829 To use it, create or edit the following keys in
24830 @code{HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\AeDebug} and/or
24831 @code{HKLM\SOFTWARE\Wow6432Node\Microsoft\Windows NT\CurrentVersion\AeDebug}
24832 (for x86_64 versions):
24836 @code{Debugger} (REG_SZ) --- a command to launch the debugger.
24837 Suggested command is: @code{@var{fully-qualified-path-to-gdb.exe} -ex
24838 "attach %ld" -ex "signal-event %ld" -ex "continue"}.
24840 The first @code{%ld} will be replaced by the process ID of the
24841 crashing process, the second @code{%ld} will be replaced by the ID of
24842 the event that blocks the crashing process, waiting for @value{GDBN}
24846 @code{Auto} (REG_SZ) --- either @code{1} or @code{0}. @code{1} will
24847 make the system run debugger specified by the Debugger key
24848 automatically, @code{0} will cause a dialog box with ``OK'' and
24849 ``Cancel'' buttons to appear, which allows the user to either
24850 terminate the crashing process (OK) or debug it (Cancel).
24853 @kindex set cygwin-exceptions
24854 @cindex debugging the Cygwin DLL
24855 @cindex Cygwin DLL, debugging
24856 @item set cygwin-exceptions @var{mode}
24857 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
24858 happen inside the Cygwin DLL. If @var{mode} is @code{off},
24859 @value{GDBN} will delay recognition of exceptions, and may ignore some
24860 exceptions which seem to be caused by internal Cygwin DLL
24861 ``bookkeeping''. This option is meant primarily for debugging the
24862 Cygwin DLL itself; the default value is @code{off} to avoid annoying
24863 @value{GDBN} users with false @code{SIGSEGV} signals.
24865 @kindex show cygwin-exceptions
24866 @item show cygwin-exceptions
24867 Displays whether @value{GDBN} will break on exceptions that happen
24868 inside the Cygwin DLL itself.
24870 @kindex set new-console
24871 @item set new-console @var{mode}
24872 If @var{mode} is @code{on} the debuggee will
24873 be started in a new console on next start.
24874 If @var{mode} is @code{off}, the debuggee will
24875 be started in the same console as the debugger.
24877 @kindex show new-console
24878 @item show new-console
24879 Displays whether a new console is used
24880 when the debuggee is started.
24882 @kindex set new-group
24883 @item set new-group @var{mode}
24884 This boolean value controls whether the debuggee should
24885 start a new group or stay in the same group as the debugger.
24886 This affects the way the Windows OS handles
24889 @kindex show new-group
24890 @item show new-group
24891 Displays current value of new-group boolean.
24893 @kindex set debugevents
24894 @item set debugevents
24895 This boolean value adds debug output concerning kernel events related
24896 to the debuggee seen by the debugger. This includes events that
24897 signal thread and process creation and exit, DLL loading and
24898 unloading, console interrupts, and debugging messages produced by the
24899 Windows @code{OutputDebugString} API call.
24901 @kindex set debugexec
24902 @item set debugexec
24903 This boolean value adds debug output concerning execute events
24904 (such as resume thread) seen by the debugger.
24906 @kindex set debugexceptions
24907 @item set debugexceptions
24908 This boolean value adds debug output concerning exceptions in the
24909 debuggee seen by the debugger.
24911 @kindex set debugmemory
24912 @item set debugmemory
24913 This boolean value adds debug output concerning debuggee memory reads
24914 and writes by the debugger.
24918 This boolean values specifies whether the debuggee is called
24919 via a shell or directly (default value is on).
24923 Displays if the debuggee will be started with a shell.
24928 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
24931 @node Non-debug DLL Symbols
24932 @subsubsection Support for DLLs without Debugging Symbols
24933 @cindex DLLs with no debugging symbols
24934 @cindex Minimal symbols and DLLs
24936 Very often on windows, some of the DLLs that your program relies on do
24937 not include symbolic debugging information (for example,
24938 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
24939 symbols in a DLL, it relies on the minimal amount of symbolic
24940 information contained in the DLL's export table. This section
24941 describes working with such symbols, known internally to @value{GDBN} as
24942 ``minimal symbols''.
24944 Note that before the debugged program has started execution, no DLLs
24945 will have been loaded. The easiest way around this problem is simply to
24946 start the program --- either by setting a breakpoint or letting the
24947 program run once to completion.
24949 @subsubsection DLL Name Prefixes
24951 In keeping with the naming conventions used by the Microsoft debugging
24952 tools, DLL export symbols are made available with a prefix based on the
24953 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
24954 also entered into the symbol table, so @code{CreateFileA} is often
24955 sufficient. In some cases there will be name clashes within a program
24956 (particularly if the executable itself includes full debugging symbols)
24957 necessitating the use of the fully qualified name when referring to the
24958 contents of the DLL. Use single-quotes around the name to avoid the
24959 exclamation mark (``!'') being interpreted as a language operator.
24961 Note that the internal name of the DLL may be all upper-case, even
24962 though the file name of the DLL is lower-case, or vice-versa. Since
24963 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
24964 some confusion. If in doubt, try the @code{info functions} and
24965 @code{info variables} commands or even @code{maint print msymbols}
24966 (@pxref{Symbols}). Here's an example:
24969 (@value{GDBP}) info function CreateFileA
24970 All functions matching regular expression "CreateFileA":
24972 Non-debugging symbols:
24973 0x77e885f4 CreateFileA
24974 0x77e885f4 KERNEL32!CreateFileA
24978 (@value{GDBP}) info function !
24979 All functions matching regular expression "!":
24981 Non-debugging symbols:
24982 0x6100114c cygwin1!__assert
24983 0x61004034 cygwin1!_dll_crt0@@0
24984 0x61004240 cygwin1!dll_crt0(per_process *)
24988 @subsubsection Working with Minimal Symbols
24990 Symbols extracted from a DLL's export table do not contain very much
24991 type information. All that @value{GDBN} can do is guess whether a symbol
24992 refers to a function or variable depending on the linker section that
24993 contains the symbol. Also note that the actual contents of the memory
24994 contained in a DLL are not available unless the program is running. This
24995 means that you cannot examine the contents of a variable or disassemble
24996 a function within a DLL without a running program.
24998 Variables are generally treated as pointers and dereferenced
24999 automatically. For this reason, it is often necessary to prefix a
25000 variable name with the address-of operator (``&'') and provide explicit
25001 type information in the command. Here's an example of the type of
25005 (@value{GDBP}) print 'cygwin1!__argv'
25006 'cygwin1!__argv' has unknown type; cast it to its declared type
25010 (@value{GDBP}) x 'cygwin1!__argv'
25011 'cygwin1!__argv' has unknown type; cast it to its declared type
25014 And two possible solutions:
25017 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
25018 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
25022 (@value{GDBP}) x/2x &'cygwin1!__argv'
25023 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
25024 (@value{GDBP}) x/x 0x10021608
25025 0x10021608: 0x0022fd98
25026 (@value{GDBP}) x/s 0x0022fd98
25027 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
25030 Setting a break point within a DLL is possible even before the program
25031 starts execution. However, under these circumstances, @value{GDBN} can't
25032 examine the initial instructions of the function in order to skip the
25033 function's frame set-up code. You can work around this by using ``*&''
25034 to set the breakpoint at a raw memory address:
25037 (@value{GDBP}) break *&'python22!PyOS_Readline'
25038 Breakpoint 1 at 0x1e04eff0
25041 The author of these extensions is not entirely convinced that setting a
25042 break point within a shared DLL like @file{kernel32.dll} is completely
25046 @subsection Commands Specific to @sc{gnu} Hurd Systems
25047 @cindex @sc{gnu} Hurd debugging
25049 This subsection describes @value{GDBN} commands specific to the
25050 @sc{gnu} Hurd native debugging.
25055 @kindex set signals@r{, Hurd command}
25056 @kindex set sigs@r{, Hurd command}
25057 This command toggles the state of inferior signal interception by
25058 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
25059 affected by this command. @code{sigs} is a shorthand alias for
25064 @kindex show signals@r{, Hurd command}
25065 @kindex show sigs@r{, Hurd command}
25066 Show the current state of intercepting inferior's signals.
25068 @item set signal-thread
25069 @itemx set sigthread
25070 @kindex set signal-thread
25071 @kindex set sigthread
25072 This command tells @value{GDBN} which thread is the @code{libc} signal
25073 thread. That thread is run when a signal is delivered to a running
25074 process. @code{set sigthread} is the shorthand alias of @code{set
25077 @item show signal-thread
25078 @itemx show sigthread
25079 @kindex show signal-thread
25080 @kindex show sigthread
25081 These two commands show which thread will run when the inferior is
25082 delivered a signal.
25085 @kindex set stopped@r{, Hurd command}
25086 This commands tells @value{GDBN} that the inferior process is stopped,
25087 as with the @code{SIGSTOP} signal. The stopped process can be
25088 continued by delivering a signal to it.
25091 @kindex show stopped@r{, Hurd command}
25092 This command shows whether @value{GDBN} thinks the debuggee is
25095 @item set exceptions
25096 @kindex set exceptions@r{, Hurd command}
25097 Use this command to turn off trapping of exceptions in the inferior.
25098 When exception trapping is off, neither breakpoints nor
25099 single-stepping will work. To restore the default, set exception
25102 @item show exceptions
25103 @kindex show exceptions@r{, Hurd command}
25104 Show the current state of trapping exceptions in the inferior.
25106 @item set task pause
25107 @kindex set task@r{, Hurd commands}
25108 @cindex task attributes (@sc{gnu} Hurd)
25109 @cindex pause current task (@sc{gnu} Hurd)
25110 This command toggles task suspension when @value{GDBN} has control.
25111 Setting it to on takes effect immediately, and the task is suspended
25112 whenever @value{GDBN} gets control. Setting it to off will take
25113 effect the next time the inferior is continued. If this option is set
25114 to off, you can use @code{set thread default pause on} or @code{set
25115 thread pause on} (see below) to pause individual threads.
25117 @item show task pause
25118 @kindex show task@r{, Hurd commands}
25119 Show the current state of task suspension.
25121 @item set task detach-suspend-count
25122 @cindex task suspend count
25123 @cindex detach from task, @sc{gnu} Hurd
25124 This command sets the suspend count the task will be left with when
25125 @value{GDBN} detaches from it.
25127 @item show task detach-suspend-count
25128 Show the suspend count the task will be left with when detaching.
25130 @item set task exception-port
25131 @itemx set task excp
25132 @cindex task exception port, @sc{gnu} Hurd
25133 This command sets the task exception port to which @value{GDBN} will
25134 forward exceptions. The argument should be the value of the @dfn{send
25135 rights} of the task. @code{set task excp} is a shorthand alias.
25137 @item set noninvasive
25138 @cindex noninvasive task options
25139 This command switches @value{GDBN} to a mode that is the least
25140 invasive as far as interfering with the inferior is concerned. This
25141 is the same as using @code{set task pause}, @code{set exceptions}, and
25142 @code{set signals} to values opposite to the defaults.
25144 @item info send-rights
25145 @itemx info receive-rights
25146 @itemx info port-rights
25147 @itemx info port-sets
25148 @itemx info dead-names
25151 @cindex send rights, @sc{gnu} Hurd
25152 @cindex receive rights, @sc{gnu} Hurd
25153 @cindex port rights, @sc{gnu} Hurd
25154 @cindex port sets, @sc{gnu} Hurd
25155 @cindex dead names, @sc{gnu} Hurd
25156 These commands display information about, respectively, send rights,
25157 receive rights, port rights, port sets, and dead names of a task.
25158 There are also shorthand aliases: @code{info ports} for @code{info
25159 port-rights} and @code{info psets} for @code{info port-sets}.
25161 @item set thread pause
25162 @kindex set thread@r{, Hurd command}
25163 @cindex thread properties, @sc{gnu} Hurd
25164 @cindex pause current thread (@sc{gnu} Hurd)
25165 This command toggles current thread suspension when @value{GDBN} has
25166 control. Setting it to on takes effect immediately, and the current
25167 thread is suspended whenever @value{GDBN} gets control. Setting it to
25168 off will take effect the next time the inferior is continued.
25169 Normally, this command has no effect, since when @value{GDBN} has
25170 control, the whole task is suspended. However, if you used @code{set
25171 task pause off} (see above), this command comes in handy to suspend
25172 only the current thread.
25174 @item show thread pause
25175 @kindex show thread@r{, Hurd command}
25176 This command shows the state of current thread suspension.
25178 @item set thread run
25179 This command sets whether the current thread is allowed to run.
25181 @item show thread run
25182 Show whether the current thread is allowed to run.
25184 @item set thread detach-suspend-count
25185 @cindex thread suspend count, @sc{gnu} Hurd
25186 @cindex detach from thread, @sc{gnu} Hurd
25187 This command sets the suspend count @value{GDBN} will leave on a
25188 thread when detaching. This number is relative to the suspend count
25189 found by @value{GDBN} when it notices the thread; use @code{set thread
25190 takeover-suspend-count} to force it to an absolute value.
25192 @item show thread detach-suspend-count
25193 Show the suspend count @value{GDBN} will leave on the thread when
25196 @item set thread exception-port
25197 @itemx set thread excp
25198 Set the thread exception port to which to forward exceptions. This
25199 overrides the port set by @code{set task exception-port} (see above).
25200 @code{set thread excp} is the shorthand alias.
25202 @item set thread takeover-suspend-count
25203 Normally, @value{GDBN}'s thread suspend counts are relative to the
25204 value @value{GDBN} finds when it notices each thread. This command
25205 changes the suspend counts to be absolute instead.
25207 @item set thread default
25208 @itemx show thread default
25209 @cindex thread default settings, @sc{gnu} Hurd
25210 Each of the above @code{set thread} commands has a @code{set thread
25211 default} counterpart (e.g., @code{set thread default pause}, @code{set
25212 thread default exception-port}, etc.). The @code{thread default}
25213 variety of commands sets the default thread properties for all
25214 threads; you can then change the properties of individual threads with
25215 the non-default commands.
25222 @value{GDBN} provides the following commands specific to the Darwin target:
25225 @item set debug darwin @var{num}
25226 @kindex set debug darwin
25227 When set to a non zero value, enables debugging messages specific to
25228 the Darwin support. Higher values produce more verbose output.
25230 @item show debug darwin
25231 @kindex show debug darwin
25232 Show the current state of Darwin messages.
25234 @item set debug mach-o @var{num}
25235 @kindex set debug mach-o
25236 When set to a non zero value, enables debugging messages while
25237 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
25238 file format used on Darwin for object and executable files.) Higher
25239 values produce more verbose output. This is a command to diagnose
25240 problems internal to @value{GDBN} and should not be needed in normal
25243 @item show debug mach-o
25244 @kindex show debug mach-o
25245 Show the current state of Mach-O file messages.
25247 @item set mach-exceptions on
25248 @itemx set mach-exceptions off
25249 @kindex set mach-exceptions
25250 On Darwin, faults are first reported as a Mach exception and are then
25251 mapped to a Posix signal. Use this command to turn on trapping of
25252 Mach exceptions in the inferior. This might be sometimes useful to
25253 better understand the cause of a fault. The default is off.
25255 @item show mach-exceptions
25256 @kindex show mach-exceptions
25257 Show the current state of exceptions trapping.
25261 @subsection FreeBSD
25264 When the ABI of a system call is changed in the FreeBSD kernel, this
25265 is implemented by leaving a compatibility system call using the old
25266 ABI at the existing number and allocating a new system call number for
25267 the version using the new ABI. As a convenience, when a system call
25268 is caught by name (@pxref{catch syscall}), compatibility system calls
25271 For example, FreeBSD 12 introduced a new variant of the @code{kevent}
25272 system call and catching the @code{kevent} system call by name catches
25276 (@value{GDBP}) catch syscall kevent
25277 Catchpoint 1 (syscalls 'freebsd11_kevent' [363] 'kevent' [560])
25283 @section Embedded Operating Systems
25285 This section describes configurations involving the debugging of
25286 embedded operating systems that are available for several different
25289 @value{GDBN} includes the ability to debug programs running on
25290 various real-time operating systems.
25292 @node Embedded Processors
25293 @section Embedded Processors
25295 This section goes into details specific to particular embedded
25298 @cindex send command to simulator
25299 Whenever a specific embedded processor has a simulator, @value{GDBN}
25300 allows to send an arbitrary command to the simulator.
25303 @item sim @var{command}
25304 @kindex sim@r{, a command}
25305 Send an arbitrary @var{command} string to the simulator. Consult the
25306 documentation for the specific simulator in use for information about
25307 acceptable commands.
25312 * ARC:: Synopsys ARC
25314 * M68K:: Motorola M68K
25315 * MicroBlaze:: Xilinx MicroBlaze
25316 * MIPS Embedded:: MIPS Embedded
25317 * OpenRISC 1000:: OpenRISC 1000 (or1k)
25318 * PowerPC Embedded:: PowerPC Embedded
25321 * Super-H:: Renesas Super-H
25325 @subsection Synopsys ARC
25326 @cindex Synopsys ARC
25327 @cindex ARC specific commands
25333 @value{GDBN} provides the following ARC-specific commands:
25336 @item set debug arc
25337 @kindex set debug arc
25338 Control the level of ARC specific debug messages. Use 0 for no messages (the
25339 default), 1 for debug messages, and 2 for even more debug messages.
25341 @item show debug arc
25342 @kindex show debug arc
25343 Show the level of ARC specific debugging in operation.
25345 @item maint print arc arc-instruction @var{address}
25346 @kindex maint print arc arc-instruction
25347 Print internal disassembler information about instruction at a given address.
25354 @value{GDBN} provides the following ARM-specific commands:
25357 @item set arm disassembler
25359 This commands selects from a list of disassembly styles. The
25360 @code{"std"} style is the standard style.
25362 @item show arm disassembler
25364 Show the current disassembly style.
25366 @item set arm apcs32
25367 @cindex ARM 32-bit mode
25368 This command toggles ARM operation mode between 32-bit and 26-bit.
25370 @item show arm apcs32
25371 Display the current usage of the ARM 32-bit mode.
25373 @item set arm fpu @var{fputype}
25374 This command sets the ARM floating-point unit (FPU) type. The
25375 argument @var{fputype} can be one of these:
25379 Determine the FPU type by querying the OS ABI.
25381 Software FPU, with mixed-endian doubles on little-endian ARM
25384 GCC-compiled FPA co-processor.
25386 Software FPU with pure-endian doubles.
25392 Show the current type of the FPU.
25395 This command forces @value{GDBN} to use the specified ABI.
25398 Show the currently used ABI.
25400 @item set arm fallback-mode (arm|thumb|auto)
25401 @value{GDBN} uses the symbol table, when available, to determine
25402 whether instructions are ARM or Thumb. This command controls
25403 @value{GDBN}'s default behavior when the symbol table is not
25404 available. The default is @samp{auto}, which causes @value{GDBN} to
25405 use the current execution mode (from the @code{T} bit in the @code{CPSR}
25408 @item show arm fallback-mode
25409 Show the current fallback instruction mode.
25411 @item set arm force-mode (arm|thumb|auto)
25412 This command overrides use of the symbol table to determine whether
25413 instructions are ARM or Thumb. The default is @samp{auto}, which
25414 causes @value{GDBN} to use the symbol table and then the setting
25415 of @samp{set arm fallback-mode}.
25417 @item show arm force-mode
25418 Show the current forced instruction mode.
25420 @item set debug arm
25421 Toggle whether to display ARM-specific debugging messages from the ARM
25422 target support subsystem.
25424 @item show debug arm
25425 Show whether ARM-specific debugging messages are enabled.
25429 @item target sim @r{[}@var{simargs}@r{]} @dots{}
25430 The @value{GDBN} ARM simulator accepts the following optional arguments.
25433 @item --swi-support=@var{type}
25434 Tell the simulator which SWI interfaces to support. The argument
25435 @var{type} may be a comma separated list of the following values.
25436 The default value is @code{all}.
25451 The Motorola m68k configuration includes ColdFire support.
25454 @subsection MicroBlaze
25455 @cindex Xilinx MicroBlaze
25456 @cindex XMD, Xilinx Microprocessor Debugger
25458 The MicroBlaze is a soft-core processor supported on various Xilinx
25459 FPGAs, such as Spartan or Virtex series. Boards with these processors
25460 usually have JTAG ports which connect to a host system running the Xilinx
25461 Embedded Development Kit (EDK) or Software Development Kit (SDK).
25462 This host system is used to download the configuration bitstream to
25463 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
25464 communicates with the target board using the JTAG interface and
25465 presents a @code{gdbserver} interface to the board. By default
25466 @code{xmd} uses port @code{1234}. (While it is possible to change
25467 this default port, it requires the use of undocumented @code{xmd}
25468 commands. Contact Xilinx support if you need to do this.)
25470 Use these GDB commands to connect to the MicroBlaze target processor.
25473 @item target remote :1234
25474 Use this command to connect to the target if you are running @value{GDBN}
25475 on the same system as @code{xmd}.
25477 @item target remote @var{xmd-host}:1234
25478 Use this command to connect to the target if it is connected to @code{xmd}
25479 running on a different system named @var{xmd-host}.
25482 Use this command to download a program to the MicroBlaze target.
25484 @item set debug microblaze @var{n}
25485 Enable MicroBlaze-specific debugging messages if non-zero.
25487 @item show debug microblaze @var{n}
25488 Show MicroBlaze-specific debugging level.
25491 @node MIPS Embedded
25492 @subsection @acronym{MIPS} Embedded
25495 @value{GDBN} supports these special commands for @acronym{MIPS} targets:
25498 @item set mipsfpu double
25499 @itemx set mipsfpu single
25500 @itemx set mipsfpu none
25501 @itemx set mipsfpu auto
25502 @itemx show mipsfpu
25503 @kindex set mipsfpu
25504 @kindex show mipsfpu
25505 @cindex @acronym{MIPS} remote floating point
25506 @cindex floating point, @acronym{MIPS} remote
25507 If your target board does not support the @acronym{MIPS} floating point
25508 coprocessor, you should use the command @samp{set mipsfpu none} (if you
25509 need this, you may wish to put the command in your @value{GDBN} init
25510 file). This tells @value{GDBN} how to find the return value of
25511 functions which return floating point values. It also allows
25512 @value{GDBN} to avoid saving the floating point registers when calling
25513 functions on the board. If you are using a floating point coprocessor
25514 with only single precision floating point support, as on the @sc{r4650}
25515 processor, use the command @samp{set mipsfpu single}. The default
25516 double precision floating point coprocessor may be selected using
25517 @samp{set mipsfpu double}.
25519 In previous versions the only choices were double precision or no
25520 floating point, so @samp{set mipsfpu on} will select double precision
25521 and @samp{set mipsfpu off} will select no floating point.
25523 As usual, you can inquire about the @code{mipsfpu} variable with
25524 @samp{show mipsfpu}.
25527 @node OpenRISC 1000
25528 @subsection OpenRISC 1000
25529 @cindex OpenRISC 1000
25532 The OpenRISC 1000 provides a free RISC instruction set architecture. It is
25533 mainly provided as a soft-core which can run on Xilinx, Altera and other
25536 @value{GDBN} for OpenRISC supports the below commands when connecting to
25544 Runs the builtin CPU simulator which can run very basic
25545 programs but does not support most hardware functions like MMU.
25546 For more complex use cases the user is advised to run an external
25547 target, and connect using @samp{target remote}.
25549 Example: @code{target sim}
25551 @item set debug or1k
25552 Toggle whether to display OpenRISC-specific debugging messages from the
25553 OpenRISC target support subsystem.
25555 @item show debug or1k
25556 Show whether OpenRISC-specific debugging messages are enabled.
25559 @node PowerPC Embedded
25560 @subsection PowerPC Embedded
25562 @cindex DVC register
25563 @value{GDBN} supports using the DVC (Data Value Compare) register to
25564 implement in hardware simple hardware watchpoint conditions of the form:
25567 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
25568 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
25571 The DVC register will be automatically used when @value{GDBN} detects
25572 such pattern in a condition expression, and the created watchpoint uses one
25573 debug register (either the @code{exact-watchpoints} option is on and the
25574 variable is scalar, or the variable has a length of one byte). This feature
25575 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
25578 When running on PowerPC embedded processors, @value{GDBN} automatically uses
25579 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
25580 in which case watchpoints using only one debug register are created when
25581 watching variables of scalar types.
25583 You can create an artificial array to watch an arbitrary memory
25584 region using one of the following commands (@pxref{Expressions}):
25587 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
25588 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
25591 PowerPC embedded processors support masked watchpoints. See the discussion
25592 about the @code{mask} argument in @ref{Set Watchpoints}.
25594 @cindex ranged breakpoint
25595 PowerPC embedded processors support hardware accelerated
25596 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
25597 the inferior whenever it executes an instruction at any address within
25598 the range it specifies. To set a ranged breakpoint in @value{GDBN},
25599 use the @code{break-range} command.
25601 @value{GDBN} provides the following PowerPC-specific commands:
25604 @kindex break-range
25605 @item break-range @var{start-location}, @var{end-location}
25606 Set a breakpoint for an address range given by
25607 @var{start-location} and @var{end-location}, which can specify a function name,
25608 a line number, an offset of lines from the current line or from the start
25609 location, or an address of an instruction (see @ref{Specify Location},
25610 for a list of all the possible ways to specify a @var{location}.)
25611 The breakpoint will stop execution of the inferior whenever it
25612 executes an instruction at any address within the specified range,
25613 (including @var{start-location} and @var{end-location}.)
25615 @kindex set powerpc
25616 @item set powerpc soft-float
25617 @itemx show powerpc soft-float
25618 Force @value{GDBN} to use (or not use) a software floating point calling
25619 convention. By default, @value{GDBN} selects the calling convention based
25620 on the selected architecture and the provided executable file.
25622 @item set powerpc vector-abi
25623 @itemx show powerpc vector-abi
25624 Force @value{GDBN} to use the specified calling convention for vector
25625 arguments and return values. The valid options are @samp{auto};
25626 @samp{generic}, to avoid vector registers even if they are present;
25627 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
25628 registers. By default, @value{GDBN} selects the calling convention
25629 based on the selected architecture and the provided executable file.
25631 @item set powerpc exact-watchpoints
25632 @itemx show powerpc exact-watchpoints
25633 Allow @value{GDBN} to use only one debug register when watching a variable
25634 of scalar type, thus assuming that the variable is accessed through the
25635 address of its first byte.
25640 @subsection Atmel AVR
25643 When configured for debugging the Atmel AVR, @value{GDBN} supports the
25644 following AVR-specific commands:
25647 @item info io_registers
25648 @kindex info io_registers@r{, AVR}
25649 @cindex I/O registers (Atmel AVR)
25650 This command displays information about the AVR I/O registers. For
25651 each register, @value{GDBN} prints its number and value.
25658 When configured for debugging CRIS, @value{GDBN} provides the
25659 following CRIS-specific commands:
25662 @item set cris-version @var{ver}
25663 @cindex CRIS version
25664 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
25665 The CRIS version affects register names and sizes. This command is useful in
25666 case autodetection of the CRIS version fails.
25668 @item show cris-version
25669 Show the current CRIS version.
25671 @item set cris-dwarf2-cfi
25672 @cindex DWARF-2 CFI and CRIS
25673 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
25674 Change to @samp{off} when using @code{gcc-cris} whose version is below
25677 @item show cris-dwarf2-cfi
25678 Show the current state of using DWARF-2 CFI.
25680 @item set cris-mode @var{mode}
25682 Set the current CRIS mode to @var{mode}. It should only be changed when
25683 debugging in guru mode, in which case it should be set to
25684 @samp{guru} (the default is @samp{normal}).
25686 @item show cris-mode
25687 Show the current CRIS mode.
25691 @subsection Renesas Super-H
25694 For the Renesas Super-H processor, @value{GDBN} provides these
25698 @item set sh calling-convention @var{convention}
25699 @kindex set sh calling-convention
25700 Set the calling-convention used when calling functions from @value{GDBN}.
25701 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
25702 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
25703 convention. If the DWARF-2 information of the called function specifies
25704 that the function follows the Renesas calling convention, the function
25705 is called using the Renesas calling convention. If the calling convention
25706 is set to @samp{renesas}, the Renesas calling convention is always used,
25707 regardless of the DWARF-2 information. This can be used to override the
25708 default of @samp{gcc} if debug information is missing, or the compiler
25709 does not emit the DWARF-2 calling convention entry for a function.
25711 @item show sh calling-convention
25712 @kindex show sh calling-convention
25713 Show the current calling convention setting.
25718 @node Architectures
25719 @section Architectures
25721 This section describes characteristics of architectures that affect
25722 all uses of @value{GDBN} with the architecture, both native and cross.
25729 * HPPA:: HP PA architecture
25734 * AMD GPU:: @acronym{AMD GPU} architectures
25738 @subsection AArch64
25739 @cindex AArch64 support
25741 When @value{GDBN} is debugging the AArch64 architecture, it provides the
25742 following special commands:
25745 @item set debug aarch64
25746 @kindex set debug aarch64
25747 This command determines whether AArch64 architecture-specific debugging
25748 messages are to be displayed.
25750 @item show debug aarch64
25751 Show whether AArch64 debugging messages are displayed.
25755 @subsubsection AArch64 SVE.
25756 @cindex AArch64 SVE.
25758 When @value{GDBN} is debugging the AArch64 architecture, if the Scalable Vector
25759 Extension (SVE) is present, then @value{GDBN} will provide the vector registers
25760 @code{$z0} through @code{$z31}, vector predicate registers @code{$p0} through
25761 @code{$p15}, and the @code{$ffr} register. In addition, the pseudo register
25762 @code{$vg} will be provided. This is the vector granule for the current thread
25763 and represents the number of 64-bit chunks in an SVE @code{z} register.
25765 If the vector length changes, then the @code{$vg} register will be updated,
25766 but the lengths of the @code{z} and @code{p} registers will not change. This
25767 is a known limitation of @value{GDBN} and does not affect the execution of the
25770 @subsubsection AArch64 Pointer Authentication.
25771 @cindex AArch64 Pointer Authentication.
25773 When @value{GDBN} is debugging the AArch64 architecture, and the program is
25774 using the v8.3-A feature Pointer Authentication (PAC), then whenever the link
25775 register @code{$lr} is pointing to an PAC function its value will be masked.
25776 When GDB prints a backtrace, any addresses that required unmasking will be
25777 postfixed with the marker [PAC]. When using the MI, this is printed as part
25778 of the @code{addr_flags} field.
25781 @subsection x86 Architecture-specific Issues
25784 @item set struct-convention @var{mode}
25785 @kindex set struct-convention
25786 @cindex struct return convention
25787 @cindex struct/union returned in registers
25788 Set the convention used by the inferior to return @code{struct}s and
25789 @code{union}s from functions to @var{mode}. Possible values of
25790 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
25791 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
25792 are returned on the stack, while @code{"reg"} means that a
25793 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
25794 be returned in a register.
25796 @item show struct-convention
25797 @kindex show struct-convention
25798 Show the current setting of the convention to return @code{struct}s
25803 @subsubsection Intel @dfn{Memory Protection Extensions} (MPX).
25804 @cindex Intel Memory Protection Extensions (MPX).
25806 Memory Protection Extension (MPX) adds the bound registers @samp{BND0}
25807 @footnote{The register named with capital letters represent the architecture
25808 registers.} through @samp{BND3}. Bound registers store a pair of 64-bit values
25809 which are the lower bound and upper bound. Bounds are effective addresses or
25810 memory locations. The upper bounds are architecturally represented in 1's
25811 complement form. A bound having lower bound = 0, and upper bound = 0
25812 (1's complement of all bits set) will allow access to the entire address space.
25814 @samp{BND0} through @samp{BND3} are represented in @value{GDBN} as @samp{bnd0raw}
25815 through @samp{bnd3raw}. Pseudo registers @samp{bnd0} through @samp{bnd3}
25816 display the upper bound performing the complement of one operation on the
25817 upper bound value, i.e.@ when upper bound in @samp{bnd0raw} is 0 in the
25818 @value{GDBN} @samp{bnd0} it will be @code{0xfff@dots{}}. In this sense it
25819 can also be noted that the upper bounds are inclusive.
25821 As an example, assume that the register BND0 holds bounds for a pointer having
25822 access allowed for the range between 0x32 and 0x71. The values present on
25823 bnd0raw and bnd registers are presented as follows:
25826 bnd0raw = @{0x32, 0xffffffff8e@}
25827 bnd0 = @{lbound = 0x32, ubound = 0x71@} : size 64
25830 This way the raw value can be accessed via bnd0raw@dots{}bnd3raw. Any
25831 change on bnd0@dots{}bnd3 or bnd0raw@dots{}bnd3raw is reflect on its
25832 counterpart. When the bnd0@dots{}bnd3 registers are displayed via
25833 Python, the display includes the memory size, in bits, accessible to
25836 Bounds can also be stored in bounds tables, which are stored in
25837 application memory. These tables store bounds for pointers by specifying
25838 the bounds pointer's value along with its bounds. Evaluating and changing
25839 bounds located in bound tables is therefore interesting while investigating
25840 bugs on MPX context. @value{GDBN} provides commands for this purpose:
25843 @item show mpx bound @var{pointer}
25844 @kindex show mpx bound
25845 Display bounds of the given @var{pointer}.
25847 @item set mpx bound @var{pointer}, @var{lbound}, @var{ubound}
25848 @kindex set mpx bound
25849 Set the bounds of a pointer in the bound table.
25850 This command takes three parameters: @var{pointer} is the pointers
25851 whose bounds are to be changed, @var{lbound} and @var{ubound} are new values
25852 for lower and upper bounds respectively.
25855 When you call an inferior function on an Intel MPX enabled program,
25856 GDB sets the inferior's bound registers to the init (disabled) state
25857 before calling the function. As a consequence, bounds checks for the
25858 pointer arguments passed to the function will always pass.
25860 This is necessary because when you call an inferior function, the
25861 program is usually in the middle of the execution of other function.
25862 Since at that point bound registers are in an arbitrary state, not
25863 clearing them would lead to random bound violations in the called
25866 You can still examine the influence of the bound registers on the
25867 execution of the called function by stopping the execution of the
25868 called function at its prologue, setting bound registers, and
25869 continuing the execution. For example:
25873 Breakpoint 2 at 0x4009de: file i386-mpx-call.c, line 47.
25874 $ print upper (a, b, c, d, 1)
25875 Breakpoint 2, upper (a=0x0, b=0x6e0000005b, c=0x0, d=0x0, len=48)....
25877 @{lbound = 0x0, ubound = ffffffff@} : size -1
25880 At this last step the value of bnd0 can be changed for investigation of bound
25881 violations caused along the execution of the call. In order to know how to
25882 set the bound registers or bound table for the call consult the ABI.
25887 See the following section.
25890 @subsection @acronym{MIPS}
25892 @cindex stack on Alpha
25893 @cindex stack on @acronym{MIPS}
25894 @cindex Alpha stack
25895 @cindex @acronym{MIPS} stack
25896 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
25897 sometimes requires @value{GDBN} to search backward in the object code to
25898 find the beginning of a function.
25900 @cindex response time, @acronym{MIPS} debugging
25901 To improve response time (especially for embedded applications, where
25902 @value{GDBN} may be restricted to a slow serial line for this search)
25903 you may want to limit the size of this search, using one of these
25907 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
25908 @item set heuristic-fence-post @var{limit}
25909 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
25910 search for the beginning of a function. A value of @var{0} (the
25911 default) means there is no limit. However, except for @var{0}, the
25912 larger the limit the more bytes @code{heuristic-fence-post} must search
25913 and therefore the longer it takes to run. You should only need to use
25914 this command when debugging a stripped executable.
25916 @item show heuristic-fence-post
25917 Display the current limit.
25921 These commands are available @emph{only} when @value{GDBN} is configured
25922 for debugging programs on Alpha or @acronym{MIPS} processors.
25924 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
25928 @item set mips abi @var{arg}
25929 @kindex set mips abi
25930 @cindex set ABI for @acronym{MIPS}
25931 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
25932 values of @var{arg} are:
25936 The default ABI associated with the current binary (this is the
25946 @item show mips abi
25947 @kindex show mips abi
25948 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
25950 @item set mips compression @var{arg}
25951 @kindex set mips compression
25952 @cindex code compression, @acronym{MIPS}
25953 Tell @value{GDBN} which @acronym{MIPS} compressed
25954 @acronym{ISA, Instruction Set Architecture} encoding is used by the
25955 inferior. @value{GDBN} uses this for code disassembly and other
25956 internal interpretation purposes. This setting is only referred to
25957 when no executable has been associated with the debugging session or
25958 the executable does not provide information about the encoding it uses.
25959 Otherwise this setting is automatically updated from information
25960 provided by the executable.
25962 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
25963 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
25964 executables containing @acronym{MIPS16} code frequently are not
25965 identified as such.
25967 This setting is ``sticky''; that is, it retains its value across
25968 debugging sessions until reset either explicitly with this command or
25969 implicitly from an executable.
25971 The compiler and/or assembler typically add symbol table annotations to
25972 identify functions compiled for the @acronym{MIPS16} or
25973 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
25974 are present, @value{GDBN} uses them in preference to the global
25975 compressed @acronym{ISA} encoding setting.
25977 @item show mips compression
25978 @kindex show mips compression
25979 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
25980 @value{GDBN} to debug the inferior.
25983 @itemx show mipsfpu
25984 @xref{MIPS Embedded, set mipsfpu}.
25986 @item set mips mask-address @var{arg}
25987 @kindex set mips mask-address
25988 @cindex @acronym{MIPS} addresses, masking
25989 This command determines whether the most-significant 32 bits of 64-bit
25990 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
25991 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
25992 setting, which lets @value{GDBN} determine the correct value.
25994 @item show mips mask-address
25995 @kindex show mips mask-address
25996 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
25999 @item set remote-mips64-transfers-32bit-regs
26000 @kindex set remote-mips64-transfers-32bit-regs
26001 This command controls compatibility with 64-bit @acronym{MIPS} targets that
26002 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
26003 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
26004 and 64 bits for other registers, set this option to @samp{on}.
26006 @item show remote-mips64-transfers-32bit-regs
26007 @kindex show remote-mips64-transfers-32bit-regs
26008 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
26010 @item set debug mips
26011 @kindex set debug mips
26012 This command turns on and off debugging messages for the @acronym{MIPS}-specific
26013 target code in @value{GDBN}.
26015 @item show debug mips
26016 @kindex show debug mips
26017 Show the current setting of @acronym{MIPS} debugging messages.
26023 @cindex HPPA support
26025 When @value{GDBN} is debugging the HP PA architecture, it provides the
26026 following special commands:
26029 @item set debug hppa
26030 @kindex set debug hppa
26031 This command determines whether HPPA architecture-specific debugging
26032 messages are to be displayed.
26034 @item show debug hppa
26035 Show whether HPPA debugging messages are displayed.
26037 @item maint print unwind @var{address}
26038 @kindex maint print unwind@r{, HPPA}
26039 This command displays the contents of the unwind table entry at the
26040 given @var{address}.
26046 @subsection PowerPC
26047 @cindex PowerPC architecture
26049 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
26050 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
26051 numbers stored in the floating point registers. These values must be stored
26052 in two consecutive registers, always starting at an even register like
26053 @code{f0} or @code{f2}.
26055 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
26056 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
26057 @code{f2} and @code{f3} for @code{$dl1} and so on.
26059 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
26060 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
26063 @subsection Nios II
26064 @cindex Nios II architecture
26066 When @value{GDBN} is debugging the Nios II architecture,
26067 it provides the following special commands:
26071 @item set debug nios2
26072 @kindex set debug nios2
26073 This command turns on and off debugging messages for the Nios II
26074 target code in @value{GDBN}.
26076 @item show debug nios2
26077 @kindex show debug nios2
26078 Show the current setting of Nios II debugging messages.
26082 @subsection Sparc64
26083 @cindex Sparc64 support
26084 @cindex Application Data Integrity
26085 @subsubsection ADI Support
26087 The M7 processor supports an Application Data Integrity (ADI) feature that
26088 detects invalid data accesses. When software allocates memory and enables
26089 ADI on the allocated memory, it chooses a 4-bit version number, sets the
26090 version in the upper 4 bits of the 64-bit pointer to that data, and stores
26091 the 4-bit version in every cacheline of that data. Hardware saves the latter
26092 in spare bits in the cache and memory hierarchy. On each load and store,
26093 the processor compares the upper 4 VA (virtual address) bits to the
26094 cacheline's version. If there is a mismatch, the processor generates a
26095 version mismatch trap which can be either precise or disrupting. The trap
26096 is an error condition which the kernel delivers to the process as a SIGSEGV
26099 Note that only 64-bit applications can use ADI and need to be built with
26102 Values of the ADI version tags, which are in granularity of a
26103 cacheline (64 bytes), can be viewed or modified.
26107 @kindex adi examine
26108 @item adi (examine | x) [ / @var{n} ] @var{addr}
26110 The @code{adi examine} command displays the value of one ADI version tag per
26113 @var{n} is a decimal integer specifying the number in bytes; the default
26114 is 1. It specifies how much ADI version information, at the ratio of 1:ADI
26115 block size, to display.
26117 @var{addr} is the address in user address space where you want @value{GDBN}
26118 to begin displaying the ADI version tags.
26120 Below is an example of displaying ADI versions of variable "shmaddr".
26123 (@value{GDBP}) adi x/100 shmaddr
26124 0xfff800010002c000: 0 0
26128 @item adi (assign | a) [ / @var{n} ] @var{addr} = @var{tag}
26130 The @code{adi assign} command is used to assign new ADI version tag
26133 @var{n} is a decimal integer specifying the number in bytes;
26134 the default is 1. It specifies how much ADI version information, at the
26135 ratio of 1:ADI block size, to modify.
26137 @var{addr} is the address in user address space where you want @value{GDBN}
26138 to begin modifying the ADI version tags.
26140 @var{tag} is the new ADI version tag.
26142 For example, do the following to modify then verify ADI versions of
26143 variable "shmaddr":
26146 (@value{GDBP}) adi a/100 shmaddr = 7
26147 (@value{GDBP}) adi x/100 shmaddr
26148 0xfff800010002c000: 7 7
26155 @cindex S12Z support
26157 When @value{GDBN} is debugging the S12Z architecture,
26158 it provides the following special command:
26161 @item maint info bdccsr
26162 @kindex maint info bdccsr@r{, S12Z}
26163 This command displays the current value of the microprocessor's
26168 @subsection @acronym{AMD GPU}
26169 @cindex @acronym{AMD GPU} support
26171 @value{GDBN} provides support for systems that have heterogeneous
26172 agents associated with @acronym{AMD GPU} devices (@pxref{Heterogeneous
26173 Debugging}) when @acronym{AMD}'s
26174 @url{https://rocm-documentation.readthedocs.io/, @acronym{ROCm, Radeon
26175 Open Compute platforM}} for HIP-Clang is installed.
26177 The following AMD GPU chips are supported:
26182 ``Vega 10'' which is displayed as @samp{vega10} by @value{GDBN} and
26183 denoted as @samp{gfx900} by the compiler.
26186 ``Vega 7nm'' which is displayed as @samp{vega20} by @value{GDBN} and
26187 denoted as @samp{gfx906} by the compiler.
26190 ``Arcturus'' which is displayed as @samp{arcturus} by @value{GDBN} and
26191 denoted as @samp{gfx908} by the compiler.
26195 @value{GDBN} supports the following source languages:
26201 @url{https://github.com/ROCm-Developer-Tools/HIP/blob/master/docs/markdown/hip_kernel_language.md,
26202 HIP Programming Language} is supported.
26204 When compiling, the @w{@option{-ggdb}} option should be used to
26205 produce debugging information suitable for use by @value{GDBN}. The
26206 @w{@option{--amdgpu-target}} option is used to specify the AMD GPUs
26207 that the executable is required to support. For example, to compile a
26208 HIP program that can utilize ``Vega 10'', ``Vega 7nm'', and
26209 ``Arcturus'' AMD GPU chips, with no optimization:
26212 hipcc -O0 -ggdb --amdgpu-target=gfx900 --amdgpu-target=gfx906 \
26213 --amdgpu-target=gfx908 bit_extract.cpp -o bit_extract
26216 The AMD GPU ROCm for HIP-Clang release compiler maps HIP source
26217 language work-items to the lanes of an AMD GPU wavefront, which are
26218 represented in @value{GDBN} as heterogeneous lanes.
26220 @item Assembly Code
26221 Assembly code kernels are supported.
26223 @item Other Languages
26224 Other languages, including OpenCL and Fortran, are currently supported
26225 as the minimal pseudo-language, provided they are compiled specifying
26226 the AMD GPU Code Object V3 and DWARF 4 formats. @xref{Unsupported
26231 The @code{info agents} command (@pxref{Heterogeneous Debugging}) lists
26232 the following information for each @acronym{AMD GPU} heterogeneous
26233 agent (in this order):
26237 the per-inferior heterogeneous agent number assigned by @value{GDBN}
26240 the global heterogeneous agent number assigned by @value{GDBN}, if the
26241 @w{@option{-gid}} option was specified
26244 the @acronym{PCIe} slot number in @acronym{BDF, Bus:Device.Function}
26251 the number of shader engines
26254 the number of @acronym{CU, Compute Unit}
26257 the number of @acronym{SIMD, Single Instruction Multiple Data} units per @acronym{CU}
26260 the number of wavefronts per @acronym{SIMD}
26267 (@value{GDBP}) info agents
26268 Id PCI Slot Device Name Shader Engines Compute Units SIMD/CU Wavefronts/SIMD
26269 1 43:00.0 vega10 4 56 4 10
26272 @acronym{AMD GPU} heterogeneous agents are not listed until the
26273 inferior has started executing the program.
26275 The @code{info queues}, @code{info dispatches}, and @code{info
26276 packets} commands are not yet supported by @acronym{AMD GPU}.
26278 An AMD GPU wavefront is represented in @value{GDBN} as a thread.
26280 @acronym{AMD GPU} supports the following @var{reggroup} values for the
26281 @samp{info registers @var{reggroup} @dots{}} command:
26299 The number of scalar and vector registers is configured when a
26300 wavefront is created. Only allocated registers are displayed. Scalar
26301 registers are reported as 32-bit signed integer values. Vector
26302 registers are reported as a wavefront size vector of signed 32-bit
26303 values. The @code{pc} is reported as a function pointer value. The
26304 @code{exec} register is reported as a wavefront size-bit unsigned
26305 integer value. The @code{vcc} and @code{xnack_mask} pseudo registers
26306 are reported as a wavefront size-bit unsigned integer value. The
26307 @code{flat_scratch} pseudo register is reported as a 64-bit unsigned
26310 AMD GPU code objects are loaded into each AMD GPU device separately.
26311 The @code{info sharedlibrary} command will therefore show the same
26312 code object loaded multiple times. As a consequence, setting a
26313 breakpoint in AMD GPU code will result in multiple breakpoints if
26314 there are multiple AMD GPU devices.
26316 If the source language runtime defers loading code objects until
26317 kernels are launched, then setting breakpoints may result in pending
26318 breakpoints that will be set when the code object is finally loaded.
26320 Threads created on @acronym{AMD GPU} heterogeneous agents have the
26321 following identifier formats:
26326 The target system's thread identifier (@var{systag}) string has the
26330 ROCm process @var{process-num} agent @var{agent-num} queue @var{queue-num} dispatch @var{dispatch-num} work-group(@var{work-group-x},@var{work-group-y},@var{work-group-z})/@var{work-group-thread-index}
26333 @c TODO: What order should coordinates be: x,y,z or z,y,x?
26335 @item @var{lane_systag}
26336 The target system's heterogeneous lane identifier (@var{lane_systag})
26337 string has the following format:
26340 ROCm process @var{process-num} agent @var{agent-num} queue @var{queue-num} dispatch @var{dispatch-num} work-group(@var{work-group-x},@var{work-group-y},@var{work-group-z}) work-item(@var{work-item-x},@var{work-item-y},@var{work-item-z})
26343 @item @code{$_dispatch_pos}
26344 The string returned by the @code{$_dispatch_pos} debugger convenience
26345 variable has the following format:
26348 (@var{work-group-x},@var{work-group-y},@var{work-group-z})/@var{work-group-thread-index}
26351 @item @code{$_thread_workgroup_pos}
26352 The string returned by the @code{$_thread_workgroup_pos} debugger
26353 convenience variable has the following format:
26356 @var{work-group-thread-index}
26359 @item @code{$_lane_workgroup_pos}
26360 The string returned by the @code{$_lane_workgroup_pos} debugger
26361 convenience variable has the following format:
26364 (@var{work-item-x},@var{work-item-y},@var{work-item-z})
26375 the inferior process LWP number
26379 @itemx dispatch-num
26380 the per-inferior heterogeneous agent number, the per-inferior
26381 heterogeneous queue number, and the per-inferior heterogeneous
26382 dispatch number associated with the thread respectively
26385 @itemx work-group-y
26386 @itemx work-group-z
26387 the grid position of the thread's work-group within the heterogeneous
26390 @item work-group-thread-index
26391 the threads's number within the heterogeneous work-group
26396 the position of the heterogeneous lane's work-item within the
26397 heterogeneous work-group
26401 @acronym{AMD GPU} heterogeneous agents support the following address
26407 the default global virtual address space
26410 the per heterogeneous work-group shared address space (@acronym{LDS,
26414 the per heterogeneous lane private address space (Scratch)
26417 the generic address space that can access the @var{global},
26418 @var{group}, or @var{private} address spaces (Flat)
26422 The @code{set debug amd-dbgapi log-level @var{level}} command can be
26423 used to enable diagnostic messages for the AMD GPU target, where
26424 @var{level} can be:
26429 no logging is enabled
26432 fatal errors are reported
26435 fatal errors and warnings are reported
26438 fatal errors, warnings, and info messages are reported
26441 all messages are reported
26445 The @code{show debug amd-dbgapi log-level} command displays the
26446 current AMD GPU target log level.
26448 For example, the following will enable information messages and send
26449 the log to a new file:
26452 (@value{GDBP}) set debug amd-dbgapi log-level info
26453 (@value{GDBP}) set logging overwrite
26454 (@value{GDBP}) set logging file log.out
26455 (@value{GDBP}) set logging debugredirect on
26456 (@value{GDBP}) set logging on
26459 If you want to print the log to both the console and a file, ommit the
26460 @code{set the logging debugredirect} command. @xref{Logging Output}.
26462 @c TODO: Add when support available:
26464 @c The @var{AMD_???} environment variable can be set to disable the kernel
26465 @c driver from ensuring that all AMD GPU wavefronts created will fully
26466 @c support the @value{GDBN} if it attached. If AMD GPU wavefronts are
26467 @c created when support is disabled, @value{GDBN} will be unable to
26468 @c report the heterogeneous dispatch associated with the wavefront, or the
26469 @c wavefront's heterogeneous work-group position. The default is enabled.
26470 @c Disabling may very marginally improve wavefront launch latency.
26472 @value{GDBN} @acronym{AMD GPU} support is currently a prototype and
26473 has the following restrictions. Future releases may remove these
26479 The debugger convenience variables, convenience functions, and
26480 commands described in @ref{Heterogeneous Debugging} are not yet
26481 implemented. The exception is the @code{info agents} command, which
26482 only currently supports the textual MI interface and does not have a
26485 However, the debugger convenience variable @code{$_wave_id} is
26486 available which returns a string that has the format:
26489 (@var{work-group-z},@var{work-group-y},@var{work-group-x})/@var{work-group-thread-index}
26497 @itemx work-group-y
26498 @itemx work-group-z
26499 the grid position of the thread's work-group within the heterogeneous
26502 @item work-group-thread-index
26503 the threads's number within the heterogeneous work-group
26507 The AMD GPU system's thread identifier (@var{systag}) string format
26508 differs from that described above, and currently has the following
26512 AMDGPU Thread @var{dispatch-num}.@var{wave-num} (@var{work-group-z},@var{work-group-y},@var{work-group-x})/@var{work-group-thread-index}
26520 the thread's per-heterogeneous queue ROCm AQL packet number of the
26521 associated dispatch packet
26524 the thread's per-inferior AMD GPU target wavefront number
26527 @itemx work-group-y
26528 @itemx work-group-z
26529 the grid position of the thread's work-group within the heterogeneous
26532 @item work-group-thread-index
26533 the threads's number within the heterogeneous work-group
26537 Only the @code{global} address space is implemented. Memory cannot be
26538 read or written in the @code{group} or @code{private} address spaces.
26539 The address space qualification of addresses described in
26540 @ref{Heterogeneous Debugging} is not implemented.
26543 The AMD GPU ROCm for HIP-Clang release compiler currently does not yet
26544 support generating valid DWARF information for symbolic variables and
26545 call frame information. As a consequence:
26550 Source variables or expressions cannot be specified in any command,
26551 such as the @code{print} command and breakpoint conditions. This
26552 includes static variables, local variables, function arguments, and
26553 any language types. However, global symbols for functions and
26554 variables can be specified, and source line information is available.
26557 The @code{backtrace} command can only show the current frame and
26558 parent frames that are fully inlined. Function or kernel arguments
26559 will not be displayed and instead an empty formal argument list may be
26563 The @code{next} command may not step over function calls, but instead
26564 stop at the first statement of the called function.
26567 Breakpoints are only reported for wavefronts. There is no support for
26568 HIP work-items that are mapped to heterogeneous lanes. The HIP
26569 work-item ID of a heterogeneous lane is not available.
26573 The AMD GPU ROCm for HIP-Clang release compiler currently adds the
26574 @w{@option{-gline-tables-only}} @w{@option{-disable-O0-noinline}}
26575 @w{@option{-disable-O0-optnone}} options when the @w{@option{-ggdb}}
26576 option is specified. These ensure source line information is
26577 generated, but not invalid DWARF, and full inlining is performed, even
26578 at @w{@option{-O0}}, so the backtrace will be available even without
26579 CFI information. If these options are not used the invalid DWARF may
26580 cause @value{GDBN} to report that it is unable to read memory (such as
26581 when reading arguments in a backtrace), and may limit the backtrace to
26582 only the top frame.
26584 Note that even with @w{@option{-ggdb}}, functions marked
26585 @code{noinline} may result in function call frames which will prevent
26586 a full backtrace. If function calls are not inlined, the @code{next}
26587 command may report errors inserting breakpoints when stepping over
26588 calls due to the invalid CFI information.
26591 Only AMD GPU Code Object V3 is supported. This is the default for the
26592 AMD GPU ROCm for HIP-Clang release compiler. The following error will
26593 be reported for incompatible code objects:
26596 warning: `ROCm-supplied DSO [loaded from memory 0x2361160..0x236d9b8]': ELF file ABI version (o) is not supported.
26597 warning: Could not load shared library symbols for ROCm-supplied DSO [loaded from memory 0x2361160..0x236d9b8].
26601 DWARF 5 is not yet supported. There is no support for compressed or split
26604 DWARF 4 is the default for the AMD GPU ROCm for HIP-Clang release
26608 No support yet for AMD GPU core dumps.
26611 The @code{watch} command is not yet support on AMD GPU devices.
26614 When in all-stop mode, AMD GPU does not currently prevent new
26615 wavefronts from being created, which may report breakpoints being hit.
26616 However, @value{GDBN} is configured by default to not remove
26617 breakpoints when at the command line in all-stop mode. This prevents
26618 breakpoints being missed by wavefronts created after at the command
26619 line in all-stop mode. The @code{set breakpoint always-inserted on}
26620 command can be used to change the default to remove breakpoints when
26621 at the command line in all-stop mode, but this may result in new
26622 wavefronts missing breakpoints.
26625 The performance of resuming from a breakpoint when a large number of
26626 threads have hit a breakpoint can currently take up to 25 seconds on a
26627 fully occupied single AMD GPU device. The techniques described in
26628 @xref{Heterogeneous Debugging} can be used to mitigate this. Once
26629 continued from the first breakpoint hit, the responsiveness of
26630 commands normally is better. Other techniques that can improve
26631 responsiveness are:
26636 Try to avoid having a lot of threads stopping at a breakpoint. For
26637 example, by placing breakpoints in conditional paths only executed by
26641 Use of @code{tbreak} so only one thread reports the breakpoint and the
26642 other threads hitting the breakpoint will be continued. A similar
26643 effect can be achieved by deleting the breakpoint manually when it is
26647 Reduce the number of wavefronts when debugging if practical.
26652 Currently each AMD GPU device can only be in use by one process that
26653 is being debugged by @value{GDBN}. The Linux @emph{cgroups} facility
26654 can be used to limit which AMD GPU devices are used by a process. In
26655 order for a @value{GDBN} process to access the AMD GPU devices of the
26656 process it is debugging, the AMD GPU devices must be included in the
26657 @value{GDBN} process @emph{cgroup}.
26659 Therefore, multiple @value{GDBN} processes can each debug a process
26660 provided the @emph{cgroups} specify disjoint sets of AMD GPU devices.
26661 However, a single @value{GDBN} process cannot debug multiple inferiors
26662 that use AMD GPU devices even if those inferiors have @emph{cgroups}
26663 that specify disjoint AMD GPU devices. This is because the
26664 @value{GDBN} process must have all the AMD GPU devices in its
26665 @emph{cgroups} and so will attempt to enable debugging for all AMD GPU
26666 devices for all inferiors it is debugging.
26668 The @code{HIP_VISIBLE_DEVICES} environment variable can also be used
26669 to limit the visible GPUs used by the HIP-Clang VDI runtime. For
26673 export HIP_VISIBLE_DEVICES=0
26677 Currently the @code{flat_scratch}, @code{vcc}, and @code{xnack_mask}
26678 special scalar registers are only accessible using their scalar
26679 register numbers and not by their register names. This will not match
26680 the assembly source text which uses register names.
26683 The @code{until} command does not work when multiple AMD GPUs are
26684 present as @value{GDBN} has limitations when there are multiple code
26685 objects that have the same breakpoint set. The work around is to use
26686 @samp{tbreak @var{line}; continue}.
26689 Restarting a program in @value{GDBN} may result in the followig error
26690 message when setting breakpoints:
26693 warning: Can't read data for section '.debug_ranges' in file 'ROCm-supplied DSO [loaded from memory 0xbe5c00..0xbe98a8]'
26696 This is due to the ROCm runtime not finalizing the loader code object
26697 list. Performing the @code{info sharedlibrary} command before setting
26698 the breakpoint ensures the code object list is updated and avoids the
26702 Currently when debugging on a ``Arcturus'' AMD GPU, @value{GDBN} may
26703 randomly report it is unable to halt a thread and report a fatal error
26704 in the @emph{dmesg} log resulting in the AMD GPU hanging.
26707 @value{GDBN} does not support following a forked process.
26710 The @code{gdbserver} is not supported.
26713 No language specific support for Fortran or OpenCL. No OpenMP
26714 language extension support for C, C++, or Fortran.
26717 Does not support the AMD GPU ROCm for HIP-HCC release compiler or
26721 AMD GPU does not currently support the compiler address, memory, or
26726 @node Controlling GDB
26727 @chapter Controlling @value{GDBN}
26729 You can alter the way @value{GDBN} interacts with you by using the
26730 @code{set} command. For commands controlling how @value{GDBN} displays
26731 data, see @ref{Print Settings, ,Print Settings}. Other settings are
26736 * Editing:: Command editing
26737 * Command History:: Command history
26738 * Screen Size:: Screen size
26739 * Output Styling:: Output styling
26740 * Numbers:: Numbers
26741 * ABI:: Configuring the current ABI
26742 * Auto-loading:: Automatically loading associated files
26743 * Messages/Warnings:: Optional warnings and messages
26744 * Debugging Output:: Optional messages about internal happenings
26745 * Other Misc Settings:: Other Miscellaneous Settings
26753 @value{GDBN} indicates its readiness to read a command by printing a string
26754 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
26755 can change the prompt string with the @code{set prompt} command. For
26756 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
26757 the prompt in one of the @value{GDBN} sessions so that you can always tell
26758 which one you are talking to.
26760 @emph{Note:} @code{set prompt} does not add a space for you after the
26761 prompt you set. This allows you to set a prompt which ends in a space
26762 or a prompt that does not.
26766 @item set prompt @var{newprompt}
26767 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
26769 @kindex show prompt
26771 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
26774 Versions of @value{GDBN} that ship with Python scripting enabled have
26775 prompt extensions. The commands for interacting with these extensions
26779 @kindex set extended-prompt
26780 @item set extended-prompt @var{prompt}
26781 Set an extended prompt that allows for substitutions.
26782 @xref{gdb.prompt}, for a list of escape sequences that can be used for
26783 substitution. Any escape sequences specified as part of the prompt
26784 string are replaced with the corresponding strings each time the prompt
26790 set extended-prompt Current working directory: \w (@value{GDBP})
26793 Note that when an extended-prompt is set, it takes control of the
26794 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
26796 @kindex show extended-prompt
26797 @item show extended-prompt
26798 Prints the extended prompt. Any escape sequences specified as part of
26799 the prompt string with @code{set extended-prompt}, are replaced with the
26800 corresponding strings each time the prompt is displayed.
26804 @section Command Editing
26806 @cindex command line editing
26808 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
26809 @sc{gnu} library provides consistent behavior for programs which provide a
26810 command line interface to the user. Advantages are @sc{gnu} Emacs-style
26811 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
26812 substitution, and a storage and recall of command history across
26813 debugging sessions.
26815 You may control the behavior of command line editing in @value{GDBN} with the
26816 command @code{set}.
26819 @kindex set editing
26822 @itemx set editing on
26823 Enable command line editing (enabled by default).
26825 @item set editing off
26826 Disable command line editing.
26828 @kindex show editing
26830 Show whether command line editing is enabled.
26833 @ifset SYSTEM_READLINE
26834 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
26836 @ifclear SYSTEM_READLINE
26837 @xref{Command Line Editing},
26839 for more details about the Readline
26840 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
26841 encouraged to read that chapter.
26843 @cindex Readline application name
26844 @value{GDBN} sets the Readline application name to @samp{gdb}. This
26845 is useful for conditions in @file{.inputrc}.
26847 @cindex operate-and-get-next
26848 @value{GDBN} defines a bindable Readline command,
26849 @code{operate-and-get-next}. This is bound to @kbd{C-o} by default.
26850 This command accepts the current line for execution and fetches the
26851 next line relative to the current line from the history for editing.
26852 Any argument is ignored.
26854 @node Command History
26855 @section Command History
26856 @cindex command history
26858 @value{GDBN} can keep track of the commands you type during your
26859 debugging sessions, so that you can be certain of precisely what
26860 happened. Use these commands to manage the @value{GDBN} command
26863 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
26864 package, to provide the history facility.
26865 @ifset SYSTEM_READLINE
26866 @xref{Using History Interactively, , , history, GNU History Library},
26868 @ifclear SYSTEM_READLINE
26869 @xref{Using History Interactively},
26871 for the detailed description of the History library.
26873 To issue a command to @value{GDBN} without affecting certain aspects of
26874 the state which is seen by users, prefix it with @samp{server }
26875 (@pxref{Server Prefix}). This
26876 means that this command will not affect the command history, nor will it
26877 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
26878 pressed on a line by itself.
26880 @cindex @code{server}, command prefix
26881 The server prefix does not affect the recording of values into the value
26882 history; to print a value without recording it into the value history,
26883 use the @code{output} command instead of the @code{print} command.
26885 Here is the description of @value{GDBN} commands related to command
26889 @cindex history substitution
26890 @cindex history file
26891 @kindex set history filename
26892 @cindex @env{GDBHISTFILE}, environment variable
26893 @item set history filename @var{fname}
26894 Set the name of the @value{GDBN} command history file to @var{fname}.
26895 This is the file where @value{GDBN} reads an initial command history
26896 list, and where it writes the command history from this session when it
26897 exits. You can access this list through history expansion or through
26898 the history command editing characters listed below. This file defaults
26899 to the value of the environment variable @code{GDBHISTFILE}, or to
26900 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
26903 @cindex save command history
26904 @kindex set history save
26905 @item set history save
26906 @itemx set history save on
26907 Record command history in a file, whose name may be specified with the
26908 @code{set history filename} command. By default, this option is disabled.
26910 @item set history save off
26911 Stop recording command history in a file.
26913 @cindex history size
26914 @kindex set history size
26915 @cindex @env{GDBHISTSIZE}, environment variable
26916 @item set history size @var{size}
26917 @itemx set history size unlimited
26918 Set the number of commands which @value{GDBN} keeps in its history list.
26919 This defaults to the value of the environment variable @env{GDBHISTSIZE}, or
26920 to 256 if this variable is not set. Non-numeric values of @env{GDBHISTSIZE}
26921 are ignored. If @var{size} is @code{unlimited} or if @env{GDBHISTSIZE} is
26922 either a negative number or the empty string, then the number of commands
26923 @value{GDBN} keeps in the history list is unlimited.
26925 @cindex remove duplicate history
26926 @kindex set history remove-duplicates
26927 @item set history remove-duplicates @var{count}
26928 @itemx set history remove-duplicates unlimited
26929 Control the removal of duplicate history entries in the command history list.
26930 If @var{count} is non-zero, @value{GDBN} will look back at the last @var{count}
26931 history entries and remove the first entry that is a duplicate of the current
26932 entry being added to the command history list. If @var{count} is
26933 @code{unlimited} then this lookbehind is unbounded. If @var{count} is 0, then
26934 removal of duplicate history entries is disabled.
26936 Only history entries added during the current session are considered for
26937 removal. This option is set to 0 by default.
26941 History expansion assigns special meaning to the character @kbd{!}.
26942 @ifset SYSTEM_READLINE
26943 @xref{Event Designators, , , history, GNU History Library},
26945 @ifclear SYSTEM_READLINE
26946 @xref{Event Designators},
26950 @cindex history expansion, turn on/off
26951 Since @kbd{!} is also the logical not operator in C, history expansion
26952 is off by default. If you decide to enable history expansion with the
26953 @code{set history expansion on} command, you may sometimes need to
26954 follow @kbd{!} (when it is used as logical not, in an expression) with
26955 a space or a tab to prevent it from being expanded. The readline
26956 history facilities do not attempt substitution on the strings
26957 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
26959 The commands to control history expansion are:
26962 @item set history expansion on
26963 @itemx set history expansion
26964 @kindex set history expansion
26965 Enable history expansion. History expansion is off by default.
26967 @item set history expansion off
26968 Disable history expansion.
26971 @kindex show history
26973 @itemx show history filename
26974 @itemx show history save
26975 @itemx show history size
26976 @itemx show history expansion
26977 These commands display the state of the @value{GDBN} history parameters.
26978 @code{show history} by itself displays all four states.
26983 @kindex show commands
26984 @cindex show last commands
26985 @cindex display command history
26986 @item show commands
26987 Display the last ten commands in the command history.
26989 @item show commands @var{n}
26990 Print ten commands centered on command number @var{n}.
26992 @item show commands +
26993 Print ten commands just after the commands last printed.
26997 @section Screen Size
26998 @cindex size of screen
26999 @cindex screen size
27002 @cindex pauses in output
27004 Certain commands to @value{GDBN} may produce large amounts of
27005 information output to the screen. To help you read all of it,
27006 @value{GDBN} pauses and asks you for input at the end of each page of
27007 output. Type @key{RET} when you want to see one more page of output,
27008 @kbd{q} to discard the remaining output, or @kbd{c} to continue
27009 without paging for the rest of the current command. Also, the screen
27010 width setting determines when to wrap lines of output. Depending on
27011 what is being printed, @value{GDBN} tries to break the line at a
27012 readable place, rather than simply letting it overflow onto the
27015 Normally @value{GDBN} knows the size of the screen from the terminal
27016 driver software. For example, on Unix @value{GDBN} uses the termcap data base
27017 together with the value of the @code{TERM} environment variable and the
27018 @code{stty rows} and @code{stty cols} settings. If this is not correct,
27019 you can override it with the @code{set height} and @code{set
27026 @kindex show height
27027 @item set height @var{lpp}
27028 @itemx set height unlimited
27030 @itemx set width @var{cpl}
27031 @itemx set width unlimited
27033 These @code{set} commands specify a screen height of @var{lpp} lines and
27034 a screen width of @var{cpl} characters. The associated @code{show}
27035 commands display the current settings.
27037 If you specify a height of either @code{unlimited} or zero lines,
27038 @value{GDBN} does not pause during output no matter how long the
27039 output is. This is useful if output is to a file or to an editor
27042 Likewise, you can specify @samp{set width unlimited} or @samp{set
27043 width 0} to prevent @value{GDBN} from wrapping its output.
27045 @item set pagination on
27046 @itemx set pagination off
27047 @kindex set pagination
27048 Turn the output pagination on or off; the default is on. Turning
27049 pagination off is the alternative to @code{set height unlimited}. Note that
27050 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
27051 Options, -batch}) also automatically disables pagination.
27053 @item show pagination
27054 @kindex show pagination
27055 Show the current pagination mode.
27058 @node Output Styling
27059 @section Output Styling
27065 @value{GDBN} can style its output on a capable terminal. This is
27066 enabled by default on most systems, but disabled by default when in
27067 batch mode (@pxref{Mode Options}). Various style settings are available;
27068 and styles can also be disabled entirely.
27071 @item set style enabled @samp{on|off}
27072 Enable or disable all styling. The default is host-dependent, with
27073 most hosts defaulting to @samp{on}.
27075 @item show style enabled
27076 Show the current state of styling.
27078 @item set style sources @samp{on|off}
27079 Enable or disable source code styling. This affects whether source
27080 code, such as the output of the @code{list} command, is styled. Note
27081 that source styling only works if styling in general is enabled, and
27082 if @value{GDBN} was linked with the GNU Source Highlight library. The
27083 default is @samp{on}.
27085 @item show style sources
27086 Show the current state of source code styling.
27089 Subcommands of @code{set style} control specific forms of styling.
27090 These subcommands all follow the same pattern: each style-able object
27091 can be styled with a foreground color, a background color, and an
27094 For example, the style of file names can be controlled using the
27095 @code{set style filename} group of commands:
27098 @item set style filename background @var{color}
27099 Set the background to @var{color}. Valid colors are @samp{none}
27100 (meaning the terminal's default color), @samp{black}, @samp{red},
27101 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
27104 @item set style filename foreground @var{color}
27105 Set the foreground to @var{color}. Valid colors are @samp{none}
27106 (meaning the terminal's default color), @samp{black}, @samp{red},
27107 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
27110 @item set style filename intensity @var{value}
27111 Set the intensity to @var{value}. Valid intensities are @samp{normal}
27112 (the default), @samp{bold}, and @samp{dim}.
27115 The @code{show style} command and its subcommands are styling
27116 a style name in their output using its own style.
27117 So, use @command{show style} to see the complete list of styles,
27118 their characteristics and the visual aspect of each style.
27120 The style-able objects are:
27123 Control the styling of file names. By default, this style's
27124 foreground color is green.
27127 Control the styling of function names. These are managed with the
27128 @code{set style function} family of commands. By default, this
27129 style's foreground color is yellow.
27132 Control the styling of variable names. These are managed with the
27133 @code{set style variable} family of commands. By default, this style's
27134 foreground color is cyan.
27137 Control the styling of addresses. These are managed with the
27138 @code{set style address} family of commands. By default, this style's
27139 foreground color is blue.
27142 Control the styling of titles. These are managed with the
27143 @code{set style title} family of commands. By default, this style's
27144 intensity is bold. Commands are using the title style to improve
27145 the readability of large output. For example, the commands
27146 @command{apropos} and @command{help} are using the title style
27147 for the command names.
27150 Control the styling of highlightings. These are managed with the
27151 @code{set style highlight} family of commands. By default, this style's
27152 foreground color is red. Commands are using the highlight style to draw
27153 the user attention to some specific parts of their output. For example,
27154 the command @command{apropos -v REGEXP} uses the highlight style to
27155 mark the documentation parts matching @var{regexp}.
27158 Control the styling of the TUI border. Note that, unlike other
27159 styling options, only the color of the border can be controlled via
27160 @code{set style}. This was done for compatibility reasons, as TUI
27161 controls to set the border's intensity predated the addition of
27162 general styling to @value{GDBN}. @xref{TUI Configuration}.
27164 @item tui-active-border
27165 Control the styling of the active TUI border; that is, the TUI window
27166 that has the focus.
27172 @cindex number representation
27173 @cindex entering numbers
27175 You can always enter numbers in octal, decimal, or hexadecimal in
27176 @value{GDBN} by the usual conventions: octal numbers begin with
27177 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
27178 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
27179 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
27180 10; likewise, the default display for numbers---when no particular
27181 format is specified---is base 10. You can change the default base for
27182 both input and output with the commands described below.
27185 @kindex set input-radix
27186 @item set input-radix @var{base}
27187 Set the default base for numeric input. Supported choices
27188 for @var{base} are decimal 8, 10, or 16. The base must itself be
27189 specified either unambiguously or using the current input radix; for
27193 set input-radix 012
27194 set input-radix 10.
27195 set input-radix 0xa
27199 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
27200 leaves the input radix unchanged, no matter what it was, since
27201 @samp{10}, being without any leading or trailing signs of its base, is
27202 interpreted in the current radix. Thus, if the current radix is 16,
27203 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
27206 @kindex set output-radix
27207 @item set output-radix @var{base}
27208 Set the default base for numeric display. Supported choices
27209 for @var{base} are decimal 8, 10, or 16. The base must itself be
27210 specified either unambiguously or using the current input radix.
27212 @kindex show input-radix
27213 @item show input-radix
27214 Display the current default base for numeric input.
27216 @kindex show output-radix
27217 @item show output-radix
27218 Display the current default base for numeric display.
27220 @item set radix @r{[}@var{base}@r{]}
27224 These commands set and show the default base for both input and output
27225 of numbers. @code{set radix} sets the radix of input and output to
27226 the same base; without an argument, it resets the radix back to its
27227 default value of 10.
27232 @section Configuring the Current ABI
27234 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
27235 application automatically. However, sometimes you need to override its
27236 conclusions. Use these commands to manage @value{GDBN}'s view of the
27242 @cindex Newlib OS ABI and its influence on the longjmp handling
27244 One @value{GDBN} configuration can debug binaries for multiple operating
27245 system targets, either via remote debugging or native emulation.
27246 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
27247 but you can override its conclusion using the @code{set osabi} command.
27248 One example where this is useful is in debugging of binaries which use
27249 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
27250 not have the same identifying marks that the standard C library for your
27253 When @value{GDBN} is debugging the AArch64 architecture, it provides a
27254 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
27255 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
27256 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
27260 Show the OS ABI currently in use.
27263 With no argument, show the list of registered available OS ABI's.
27265 @item set osabi @var{abi}
27266 Set the current OS ABI to @var{abi}.
27269 @cindex float promotion
27271 Generally, the way that an argument of type @code{float} is passed to a
27272 function depends on whether the function is prototyped. For a prototyped
27273 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
27274 according to the architecture's convention for @code{float}. For unprototyped
27275 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
27276 @code{double} and then passed.
27278 Unfortunately, some forms of debug information do not reliably indicate whether
27279 a function is prototyped. If @value{GDBN} calls a function that is not marked
27280 as prototyped, it consults @kbd{set coerce-float-to-double}.
27283 @kindex set coerce-float-to-double
27284 @item set coerce-float-to-double
27285 @itemx set coerce-float-to-double on
27286 Arguments of type @code{float} will be promoted to @code{double} when passed
27287 to an unprototyped function. This is the default setting.
27289 @item set coerce-float-to-double off
27290 Arguments of type @code{float} will be passed directly to unprototyped
27293 @kindex show coerce-float-to-double
27294 @item show coerce-float-to-double
27295 Show the current setting of promoting @code{float} to @code{double}.
27299 @kindex show cp-abi
27300 @value{GDBN} needs to know the ABI used for your program's C@t{++}
27301 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
27302 used to build your application. @value{GDBN} only fully supports
27303 programs with a single C@t{++} ABI; if your program contains code using
27304 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
27305 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
27306 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
27307 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
27308 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
27309 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
27314 Show the C@t{++} ABI currently in use.
27317 With no argument, show the list of supported C@t{++} ABI's.
27319 @item set cp-abi @var{abi}
27320 @itemx set cp-abi auto
27321 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
27325 @section Automatically loading associated files
27326 @cindex auto-loading
27328 @value{GDBN} sometimes reads files with commands and settings automatically,
27329 without being explicitly told so by the user. We call this feature
27330 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
27331 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
27332 results or introduce security risks (e.g., if the file comes from untrusted
27336 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
27337 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
27339 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
27340 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
27343 There are various kinds of files @value{GDBN} can automatically load.
27344 In addition to these files, @value{GDBN} supports auto-loading code written
27345 in various extension languages. @xref{Auto-loading extensions}.
27347 Note that loading of these associated files (including the local @file{.gdbinit}
27348 file) requires accordingly configured @code{auto-load safe-path}
27349 (@pxref{Auto-loading safe path}).
27351 For these reasons, @value{GDBN} includes commands and options to let you
27352 control when to auto-load files and which files should be auto-loaded.
27355 @anchor{set auto-load off}
27356 @kindex set auto-load off
27357 @item set auto-load off
27358 Globally disable loading of all auto-loaded files.
27359 You may want to use this command with the @samp{-iex} option
27360 (@pxref{Option -init-eval-command}) such as:
27362 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
27365 Be aware that system init file (@pxref{System-wide configuration})
27366 and init files from your home directory (@pxref{Home Directory Init File})
27367 still get read (as they come from generally trusted directories).
27368 To prevent @value{GDBN} from auto-loading even those init files, use the
27369 @option{-nx} option (@pxref{Mode Options}), in addition to
27370 @code{set auto-load no}.
27372 @anchor{show auto-load}
27373 @kindex show auto-load
27374 @item show auto-load
27375 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
27379 (@value{GDBP}) show auto-load
27380 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
27381 libthread-db: Auto-loading of inferior specific libthread_db is on.
27382 local-gdbinit: Auto-loading of .gdbinit script from current directory
27384 python-scripts: Auto-loading of Python scripts is on.
27385 safe-path: List of directories from which it is safe to auto-load files
27386 is $debugdir:$datadir/auto-load.
27387 scripts-directory: List of directories from which to load auto-loaded scripts
27388 is $debugdir:$datadir/auto-load.
27391 @anchor{info auto-load}
27392 @kindex info auto-load
27393 @item info auto-load
27394 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
27398 (@value{GDBP}) info auto-load
27401 Yes /home/user/gdb/gdb-gdb.gdb
27402 libthread-db: No auto-loaded libthread-db.
27403 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
27407 Yes /home/user/gdb/gdb-gdb.py
27411 These are @value{GDBN} control commands for the auto-loading:
27413 @multitable @columnfractions .5 .5
27414 @item @xref{set auto-load off}.
27415 @tab Disable auto-loading globally.
27416 @item @xref{show auto-load}.
27417 @tab Show setting of all kinds of files.
27418 @item @xref{info auto-load}.
27419 @tab Show state of all kinds of files.
27420 @item @xref{set auto-load gdb-scripts}.
27421 @tab Control for @value{GDBN} command scripts.
27422 @item @xref{show auto-load gdb-scripts}.
27423 @tab Show setting of @value{GDBN} command scripts.
27424 @item @xref{info auto-load gdb-scripts}.
27425 @tab Show state of @value{GDBN} command scripts.
27426 @item @xref{set auto-load python-scripts}.
27427 @tab Control for @value{GDBN} Python scripts.
27428 @item @xref{show auto-load python-scripts}.
27429 @tab Show setting of @value{GDBN} Python scripts.
27430 @item @xref{info auto-load python-scripts}.
27431 @tab Show state of @value{GDBN} Python scripts.
27432 @item @xref{set auto-load guile-scripts}.
27433 @tab Control for @value{GDBN} Guile scripts.
27434 @item @xref{show auto-load guile-scripts}.
27435 @tab Show setting of @value{GDBN} Guile scripts.
27436 @item @xref{info auto-load guile-scripts}.
27437 @tab Show state of @value{GDBN} Guile scripts.
27438 @item @xref{set auto-load scripts-directory}.
27439 @tab Control for @value{GDBN} auto-loaded scripts location.
27440 @item @xref{show auto-load scripts-directory}.
27441 @tab Show @value{GDBN} auto-loaded scripts location.
27442 @item @xref{add-auto-load-scripts-directory}.
27443 @tab Add directory for auto-loaded scripts location list.
27444 @item @xref{set auto-load local-gdbinit}.
27445 @tab Control for init file in the current directory.
27446 @item @xref{show auto-load local-gdbinit}.
27447 @tab Show setting of init file in the current directory.
27448 @item @xref{info auto-load local-gdbinit}.
27449 @tab Show state of init file in the current directory.
27450 @item @xref{set auto-load libthread-db}.
27451 @tab Control for thread debugging library.
27452 @item @xref{show auto-load libthread-db}.
27453 @tab Show setting of thread debugging library.
27454 @item @xref{info auto-load libthread-db}.
27455 @tab Show state of thread debugging library.
27456 @item @xref{set auto-load safe-path}.
27457 @tab Control directories trusted for automatic loading.
27458 @item @xref{show auto-load safe-path}.
27459 @tab Show directories trusted for automatic loading.
27460 @item @xref{add-auto-load-safe-path}.
27461 @tab Add directory trusted for automatic loading.
27464 @node Init File in the Current Directory
27465 @subsection Automatically loading init file in the current directory
27466 @cindex auto-loading init file in the current directory
27468 By default, @value{GDBN} reads and executes the canned sequences of commands
27469 from init file (if any) in the current working directory,
27470 see @ref{Init File in the Current Directory during Startup}.
27472 Note that loading of this local @file{.gdbinit} file also requires accordingly
27473 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
27476 @anchor{set auto-load local-gdbinit}
27477 @kindex set auto-load local-gdbinit
27478 @item set auto-load local-gdbinit [on|off]
27479 Enable or disable the auto-loading of canned sequences of commands
27480 (@pxref{Sequences}) found in init file in the current directory.
27482 @anchor{show auto-load local-gdbinit}
27483 @kindex show auto-load local-gdbinit
27484 @item show auto-load local-gdbinit
27485 Show whether auto-loading of canned sequences of commands from init file in the
27486 current directory is enabled or disabled.
27488 @anchor{info auto-load local-gdbinit}
27489 @kindex info auto-load local-gdbinit
27490 @item info auto-load local-gdbinit
27491 Print whether canned sequences of commands from init file in the
27492 current directory have been auto-loaded.
27495 @node libthread_db.so.1 file
27496 @subsection Automatically loading thread debugging library
27497 @cindex auto-loading libthread_db.so.1
27499 This feature is currently present only on @sc{gnu}/Linux native hosts.
27501 @value{GDBN} reads in some cases thread debugging library from places specific
27502 to the inferior (@pxref{set libthread-db-search-path}).
27504 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
27505 without checking this @samp{set auto-load libthread-db} switch as system
27506 libraries have to be trusted in general. In all other cases of
27507 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
27508 auto-load libthread-db} is enabled before trying to open such thread debugging
27511 Note that loading of this debugging library also requires accordingly configured
27512 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
27515 @anchor{set auto-load libthread-db}
27516 @kindex set auto-load libthread-db
27517 @item set auto-load libthread-db [on|off]
27518 Enable or disable the auto-loading of inferior specific thread debugging library.
27520 @anchor{show auto-load libthread-db}
27521 @kindex show auto-load libthread-db
27522 @item show auto-load libthread-db
27523 Show whether auto-loading of inferior specific thread debugging library is
27524 enabled or disabled.
27526 @anchor{info auto-load libthread-db}
27527 @kindex info auto-load libthread-db
27528 @item info auto-load libthread-db
27529 Print the list of all loaded inferior specific thread debugging libraries and
27530 for each such library print list of inferior @var{pid}s using it.
27533 @node Auto-loading safe path
27534 @subsection Security restriction for auto-loading
27535 @cindex auto-loading safe-path
27537 As the files of inferior can come from untrusted source (such as submitted by
27538 an application user) @value{GDBN} does not always load any files automatically.
27539 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
27540 directories trusted for loading files not explicitly requested by user.
27541 Each directory can also be a shell wildcard pattern.
27543 If the path is not set properly you will see a warning and the file will not
27548 Reading symbols from /home/user/gdb/gdb...done.
27549 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
27550 declined by your `auto-load safe-path' set
27551 to "$debugdir:$datadir/auto-load".
27552 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
27553 declined by your `auto-load safe-path' set
27554 to "$debugdir:$datadir/auto-load".
27558 To instruct @value{GDBN} to go ahead and use the init files anyway,
27559 invoke @value{GDBN} like this:
27562 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
27565 The list of trusted directories is controlled by the following commands:
27568 @anchor{set auto-load safe-path}
27569 @kindex set auto-load safe-path
27570 @item set auto-load safe-path @r{[}@var{directories}@r{]}
27571 Set the list of directories (and their subdirectories) trusted for automatic
27572 loading and execution of scripts. You can also enter a specific trusted file.
27573 Each directory can also be a shell wildcard pattern; wildcards do not match
27574 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
27575 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
27576 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
27577 its default value as specified during @value{GDBN} compilation.
27579 The list of directories uses path separator (@samp{:} on GNU and Unix
27580 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
27581 to the @env{PATH} environment variable.
27583 @anchor{show auto-load safe-path}
27584 @kindex show auto-load safe-path
27585 @item show auto-load safe-path
27586 Show the list of directories trusted for automatic loading and execution of
27589 @anchor{add-auto-load-safe-path}
27590 @kindex add-auto-load-safe-path
27591 @item add-auto-load-safe-path
27592 Add an entry (or list of entries) to the list of directories trusted for
27593 automatic loading and execution of scripts. Multiple entries may be delimited
27594 by the host platform path separator in use.
27597 This variable defaults to what @code{--with-auto-load-dir} has been configured
27598 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
27599 substitution applies the same as for @ref{set auto-load scripts-directory}.
27600 The default @code{set auto-load safe-path} value can be also overriden by
27601 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
27603 Setting this variable to @file{/} disables this security protection,
27604 corresponding @value{GDBN} configuration option is
27605 @option{--without-auto-load-safe-path}.
27606 This variable is supposed to be set to the system directories writable by the
27607 system superuser only. Users can add their source directories in init files in
27608 their home directories (@pxref{Home Directory Init File}). See also deprecated
27609 init file in the current directory
27610 (@pxref{Init File in the Current Directory during Startup}).
27612 To force @value{GDBN} to load the files it declined to load in the previous
27613 example, you could use one of the following ways:
27616 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
27617 Specify this trusted directory (or a file) as additional component of the list.
27618 You have to specify also any existing directories displayed by
27619 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
27621 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
27622 Specify this directory as in the previous case but just for a single
27623 @value{GDBN} session.
27625 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
27626 Disable auto-loading safety for a single @value{GDBN} session.
27627 This assumes all the files you debug during this @value{GDBN} session will come
27628 from trusted sources.
27630 @item @kbd{./configure --without-auto-load-safe-path}
27631 During compilation of @value{GDBN} you may disable any auto-loading safety.
27632 This assumes all the files you will ever debug with this @value{GDBN} come from
27636 On the other hand you can also explicitly forbid automatic files loading which
27637 also suppresses any such warning messages:
27640 @item @kbd{gdb -iex "set auto-load no" @dots{}}
27641 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
27643 @item @file{~/.gdbinit}: @samp{set auto-load no}
27644 Disable auto-loading globally for the user
27645 (@pxref{Home Directory Init File}). While it is improbable, you could also
27646 use system init file instead (@pxref{System-wide configuration}).
27649 This setting applies to the file names as entered by user. If no entry matches
27650 @value{GDBN} tries as a last resort to also resolve all the file names into
27651 their canonical form (typically resolving symbolic links) and compare the
27652 entries again. @value{GDBN} already canonicalizes most of the filenames on its
27653 own before starting the comparison so a canonical form of directories is
27654 recommended to be entered.
27656 @node Auto-loading verbose mode
27657 @subsection Displaying files tried for auto-load
27658 @cindex auto-loading verbose mode
27660 For better visibility of all the file locations where you can place scripts to
27661 be auto-loaded with inferior --- or to protect yourself against accidental
27662 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
27663 all the files attempted to be loaded. Both existing and non-existing files may
27666 For example the list of directories from which it is safe to auto-load files
27667 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
27668 may not be too obvious while setting it up.
27671 (@value{GDBP}) set debug auto-load on
27672 (@value{GDBP}) file ~/src/t/true
27673 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
27674 for objfile "/tmp/true".
27675 auto-load: Updating directories of "/usr:/opt".
27676 auto-load: Using directory "/usr".
27677 auto-load: Using directory "/opt".
27678 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
27679 by your `auto-load safe-path' set to "/usr:/opt".
27683 @anchor{set debug auto-load}
27684 @kindex set debug auto-load
27685 @item set debug auto-load [on|off]
27686 Set whether to print the filenames attempted to be auto-loaded.
27688 @anchor{show debug auto-load}
27689 @kindex show debug auto-load
27690 @item show debug auto-load
27691 Show whether printing of the filenames attempted to be auto-loaded is turned
27695 @node Messages/Warnings
27696 @section Optional Warnings and Messages
27698 @cindex verbose operation
27699 @cindex optional warnings
27700 By default, @value{GDBN} is silent about its inner workings. If you are
27701 running on a slow machine, you may want to use the @code{set verbose}
27702 command. This makes @value{GDBN} tell you when it does a lengthy
27703 internal operation, so you will not think it has crashed.
27705 Currently, the messages controlled by @code{set verbose} are those
27706 which announce that the symbol table for a source file is being read;
27707 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
27710 @kindex set verbose
27711 @item set verbose on
27712 Enables @value{GDBN} output of certain informational messages.
27714 @item set verbose off
27715 Disables @value{GDBN} output of certain informational messages.
27717 @kindex show verbose
27719 Displays whether @code{set verbose} is on or off.
27722 By default, if @value{GDBN} encounters bugs in the symbol table of an
27723 object file, it is silent; but if you are debugging a compiler, you may
27724 find this information useful (@pxref{Symbol Errors, ,Errors Reading
27729 @kindex set complaints
27730 @item set complaints @var{limit}
27731 Permits @value{GDBN} to output @var{limit} complaints about each type of
27732 unusual symbols before becoming silent about the problem. Set
27733 @var{limit} to zero to suppress all complaints; set it to a large number
27734 to prevent complaints from being suppressed.
27736 @kindex show complaints
27737 @item show complaints
27738 Displays how many symbol complaints @value{GDBN} is permitted to produce.
27742 @anchor{confirmation requests}
27743 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
27744 lot of stupid questions to confirm certain commands. For example, if
27745 you try to run a program which is already running:
27749 The program being debugged has been started already.
27750 Start it from the beginning? (y or n)
27753 If you are willing to unflinchingly face the consequences of your own
27754 commands, you can disable this ``feature'':
27758 @kindex set confirm
27760 @cindex confirmation
27761 @cindex stupid questions
27762 @item set confirm off
27763 Disables confirmation requests. Note that running @value{GDBN} with
27764 the @option{--batch} option (@pxref{Mode Options, -batch}) also
27765 automatically disables confirmation requests.
27767 @item set confirm on
27768 Enables confirmation requests (the default).
27770 @kindex show confirm
27772 Displays state of confirmation requests.
27776 @cindex command tracing
27777 If you need to debug user-defined commands or sourced files you may find it
27778 useful to enable @dfn{command tracing}. In this mode each command will be
27779 printed as it is executed, prefixed with one or more @samp{+} symbols, the
27780 quantity denoting the call depth of each command.
27783 @kindex set trace-commands
27784 @cindex command scripts, debugging
27785 @item set trace-commands on
27786 Enable command tracing.
27787 @item set trace-commands off
27788 Disable command tracing.
27789 @item show trace-commands
27790 Display the current state of command tracing.
27793 @node Debugging Output
27794 @section Optional Messages about Internal Happenings
27795 @cindex optional debugging messages
27797 @value{GDBN} has commands that enable optional debugging messages from
27798 various @value{GDBN} subsystems; normally these commands are of
27799 interest to @value{GDBN} maintainers, or when reporting a bug. This
27800 section documents those commands.
27803 @kindex set exec-done-display
27804 @item set exec-done-display
27805 Turns on or off the notification of asynchronous commands'
27806 completion. When on, @value{GDBN} will print a message when an
27807 asynchronous command finishes its execution. The default is off.
27808 @kindex show exec-done-display
27809 @item show exec-done-display
27810 Displays the current setting of asynchronous command completion
27813 @cindex ARM AArch64
27814 @item set debug aarch64
27815 Turns on or off display of debugging messages related to ARM AArch64.
27816 The default is off.
27818 @item show debug aarch64
27819 Displays the current state of displaying debugging messages related to
27821 @cindex gdbarch debugging info
27822 @cindex architecture debugging info
27823 @item set debug arch
27824 Turns on or off display of gdbarch debugging info. The default is off
27825 @item show debug arch
27826 Displays the current state of displaying gdbarch debugging info.
27827 @item set debug aix-solib
27828 @cindex AIX shared library debugging
27829 Control display of debugging messages from the AIX shared library
27830 support module. The default is off.
27831 @item show debug aix-thread
27832 Show the current state of displaying AIX shared library debugging messages.
27833 @item set debug aix-thread
27834 @cindex AIX threads
27835 Display debugging messages about inner workings of the AIX thread
27837 @item show debug aix-thread
27838 Show the current state of AIX thread debugging info display.
27839 @item set debug check-physname
27841 Check the results of the ``physname'' computation. When reading DWARF
27842 debugging information for C@t{++}, @value{GDBN} attempts to compute
27843 each entity's name. @value{GDBN} can do this computation in two
27844 different ways, depending on exactly what information is present.
27845 When enabled, this setting causes @value{GDBN} to compute the names
27846 both ways and display any discrepancies.
27847 @item show debug check-physname
27848 Show the current state of ``physname'' checking.
27849 @item set debug coff-pe-read
27850 @cindex COFF/PE exported symbols
27851 Control display of debugging messages related to reading of COFF/PE
27852 exported symbols. The default is off.
27853 @item show debug coff-pe-read
27854 Displays the current state of displaying debugging messages related to
27855 reading of COFF/PE exported symbols.
27856 @item set debug dwarf-die
27858 Dump DWARF DIEs after they are read in.
27859 The value is the number of nesting levels to print.
27860 A value of zero turns off the display.
27861 @item show debug dwarf-die
27862 Show the current state of DWARF DIE debugging.
27863 @item set debug dwarf-line
27864 @cindex DWARF Line Tables
27865 Turns on or off display of debugging messages related to reading
27866 DWARF line tables. The default is 0 (off).
27867 A value of 1 provides basic information.
27868 A value greater than 1 provides more verbose information.
27869 @item show debug dwarf-line
27870 Show the current state of DWARF line table debugging.
27871 @item set debug dwarf-read
27872 @cindex DWARF Reading
27873 Turns on or off display of debugging messages related to reading
27874 DWARF debug info. The default is 0 (off).
27875 A value of 1 provides basic information.
27876 A value greater than 1 provides more verbose information.
27877 @item show debug dwarf-read
27878 Show the current state of DWARF reader debugging.
27879 @item set debug displaced
27880 @cindex displaced stepping debugging info
27881 Turns on or off display of @value{GDBN} debugging info for the
27882 displaced stepping support. The default is off.
27883 @item show debug displaced
27884 Displays the current state of displaying @value{GDBN} debugging info
27885 related to displaced stepping.
27886 @item set debug event
27887 @cindex event debugging info
27888 Turns on or off display of @value{GDBN} event debugging info. The
27890 @item show debug event
27891 Displays the current state of displaying @value{GDBN} event debugging
27893 @item set debug expression
27894 @cindex expression debugging info
27895 Turns on or off display of debugging info about @value{GDBN}
27896 expression parsing. The default is off.
27897 @item show debug expression
27898 Displays the current state of displaying debugging info about
27899 @value{GDBN} expression parsing.
27900 @item set debug fbsd-lwp
27901 @cindex FreeBSD LWP debug messages
27902 Turns on or off debugging messages from the FreeBSD LWP debug support.
27903 @item show debug fbsd-lwp
27904 Show the current state of FreeBSD LWP debugging messages.
27905 @item set debug fbsd-nat
27906 @cindex FreeBSD native target debug messages
27907 Turns on or off debugging messages from the FreeBSD native target.
27908 @item show debug fbsd-nat
27909 Show the current state of FreeBSD native target debugging messages.
27910 @item set debug frame
27911 @cindex frame debugging info
27912 Turns on or off display of @value{GDBN} frame debugging info. The
27914 @item show debug frame
27915 Displays the current state of displaying @value{GDBN} frame debugging
27917 @item set debug gnu-nat
27918 @cindex @sc{gnu}/Hurd debug messages
27919 Turn on or off debugging messages from the @sc{gnu}/Hurd debug support.
27920 @item show debug gnu-nat
27921 Show the current state of @sc{gnu}/Hurd debugging messages.
27922 @item set debug infrun
27923 @cindex inferior debugging info
27924 Turns on or off display of @value{GDBN} debugging info for running the inferior.
27925 The default is off. @file{infrun.c} contains GDB's runtime state machine used
27926 for implementing operations such as single-stepping the inferior.
27927 @item show debug infrun
27928 Displays the current state of @value{GDBN} inferior debugging.
27929 @item set debug jit
27930 @cindex just-in-time compilation, debugging messages
27931 Turn on or off debugging messages from JIT debug support.
27932 @item show debug jit
27933 Displays the current state of @value{GDBN} JIT debugging.
27934 @item set debug lin-lwp
27935 @cindex @sc{gnu}/Linux LWP debug messages
27936 @cindex Linux lightweight processes
27937 Turn on or off debugging messages from the Linux LWP debug support.
27938 @item show debug lin-lwp
27939 Show the current state of Linux LWP debugging messages.
27940 @item set debug linux-namespaces
27941 @cindex @sc{gnu}/Linux namespaces debug messages
27942 Turn on or off debugging messages from the Linux namespaces debug support.
27943 @item show debug linux-namespaces
27944 Show the current state of Linux namespaces debugging messages.
27945 @item set debug mach-o
27946 @cindex Mach-O symbols processing
27947 Control display of debugging messages related to Mach-O symbols
27948 processing. The default is off.
27949 @item show debug mach-o
27950 Displays the current state of displaying debugging messages related to
27951 reading of COFF/PE exported symbols.
27952 @item set debug notification
27953 @cindex remote async notification debugging info
27954 Turn on or off debugging messages about remote async notification.
27955 The default is off.
27956 @item show debug notification
27957 Displays the current state of remote async notification debugging messages.
27958 @item set debug observer
27959 @cindex observer debugging info
27960 Turns on or off display of @value{GDBN} observer debugging. This
27961 includes info such as the notification of observable events.
27962 @item show debug observer
27963 Displays the current state of observer debugging.
27964 @item set debug overload
27965 @cindex C@t{++} overload debugging info
27966 Turns on or off display of @value{GDBN} C@t{++} overload debugging
27967 info. This includes info such as ranking of functions, etc. The default
27969 @item show debug overload
27970 Displays the current state of displaying @value{GDBN} C@t{++} overload
27972 @cindex expression parser, debugging info
27973 @cindex debug expression parser
27974 @item set debug parser
27975 Turns on or off the display of expression parser debugging output.
27976 Internally, this sets the @code{yydebug} variable in the expression
27977 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
27978 details. The default is off.
27979 @item show debug parser
27980 Show the current state of expression parser debugging.
27981 @cindex packets, reporting on stdout
27982 @cindex serial connections, debugging
27983 @cindex debug remote protocol
27984 @cindex remote protocol debugging
27985 @cindex display remote packets
27986 @item set debug remote
27987 Turns on or off display of reports on all packets sent back and forth across
27988 the serial line to the remote machine. The info is printed on the
27989 @value{GDBN} standard output stream. The default is off.
27990 @item show debug remote
27991 Displays the state of display of remote packets.
27993 @item set debug remote-packet-max-chars
27994 Sets the maximum number of characters to display for each remote packet when
27995 @code{set debug remote} is on. This is useful to prevent @value{GDBN} from
27996 displaying lengthy remote packets and polluting the console.
27998 The default value is @code{512}, which means @value{GDBN} will truncate each
27999 remote packet after 512 bytes.
28001 Setting this option to @code{unlimited} will disable truncation and will output
28002 the full length of the remote packets.
28003 @item show debug remote-packet-max-chars
28004 Displays the number of bytes to output for remote packet debugging.
28006 @item set debug separate-debug-file
28007 Turns on or off display of debug output about separate debug file search.
28008 @item show debug separate-debug-file
28009 Displays the state of separate debug file search debug output.
28011 @item set debug serial
28012 Turns on or off display of @value{GDBN} serial debugging info. The
28014 @item show debug serial
28015 Displays the current state of displaying @value{GDBN} serial debugging
28017 @item set debug solib-frv
28018 @cindex FR-V shared-library debugging
28019 Turn on or off debugging messages for FR-V shared-library code.
28020 @item show debug solib-frv
28021 Display the current state of FR-V shared-library code debugging
28023 @item set debug symbol-lookup
28024 @cindex symbol lookup
28025 Turns on or off display of debugging messages related to symbol lookup.
28026 The default is 0 (off).
28027 A value of 1 provides basic information.
28028 A value greater than 1 provides more verbose information.
28029 @item show debug symbol-lookup
28030 Show the current state of symbol lookup debugging messages.
28031 @item set debug symfile
28032 @cindex symbol file functions
28033 Turns on or off display of debugging messages related to symbol file functions.
28034 The default is off. @xref{Files}.
28035 @item show debug symfile
28036 Show the current state of symbol file debugging messages.
28037 @item set debug symtab-create
28038 @cindex symbol table creation
28039 Turns on or off display of debugging messages related to symbol table creation.
28040 The default is 0 (off).
28041 A value of 1 provides basic information.
28042 A value greater than 1 provides more verbose information.
28043 @item show debug symtab-create
28044 Show the current state of symbol table creation debugging.
28045 @item set debug target
28046 @cindex target debugging info
28047 Turns on or off display of @value{GDBN} target debugging info. This info
28048 includes what is going on at the target level of GDB, as it happens. The
28049 default is 0. Set it to 1 to track events, and to 2 to also track the
28050 value of large memory transfers.
28051 @item show debug target
28052 Displays the current state of displaying @value{GDBN} target debugging
28054 @item set debug timestamp
28055 @cindex timestamping debugging info
28056 Turns on or off display of timestamps with @value{GDBN} debugging info.
28057 When enabled, seconds and microseconds are displayed before each debugging
28059 @item show debug timestamp
28060 Displays the current state of displaying timestamps with @value{GDBN}
28062 @item set debug varobj
28063 @cindex variable object debugging info
28064 Turns on or off display of @value{GDBN} variable object debugging
28065 info. The default is off.
28066 @item show debug varobj
28067 Displays the current state of displaying @value{GDBN} variable object
28069 @item set debug xml
28070 @cindex XML parser debugging
28071 Turn on or off debugging messages for built-in XML parsers.
28072 @item show debug xml
28073 Displays the current state of XML debugging messages.
28076 @node Other Misc Settings
28077 @section Other Miscellaneous Settings
28078 @cindex miscellaneous settings
28081 @kindex set interactive-mode
28082 @item set interactive-mode
28083 If @code{on}, forces @value{GDBN} to assume that GDB was started
28084 in a terminal. In practice, this means that @value{GDBN} should wait
28085 for the user to answer queries generated by commands entered at
28086 the command prompt. If @code{off}, forces @value{GDBN} to operate
28087 in the opposite mode, and it uses the default answers to all queries.
28088 If @code{auto} (the default), @value{GDBN} tries to determine whether
28089 its standard input is a terminal, and works in interactive-mode if it
28090 is, non-interactively otherwise.
28092 In the vast majority of cases, the debugger should be able to guess
28093 correctly which mode should be used. But this setting can be useful
28094 in certain specific cases, such as running a MinGW @value{GDBN}
28095 inside a cygwin window.
28097 @kindex show interactive-mode
28098 @item show interactive-mode
28099 Displays whether the debugger is operating in interactive mode or not.
28102 @node Extending GDB
28103 @chapter Extending @value{GDBN}
28104 @cindex extending GDB
28106 @value{GDBN} provides several mechanisms for extension.
28107 @value{GDBN} also provides the ability to automatically load
28108 extensions when it reads a file for debugging. This allows the
28109 user to automatically customize @value{GDBN} for the program
28113 * Sequences:: Canned Sequences of @value{GDBN} Commands
28114 * Python:: Extending @value{GDBN} using Python
28115 * Guile:: Extending @value{GDBN} using Guile
28116 * Auto-loading extensions:: Automatically loading extensions
28117 * Multiple Extension Languages:: Working with multiple extension languages
28118 * Aliases:: Creating new spellings of existing commands
28121 To facilitate the use of extension languages, @value{GDBN} is capable
28122 of evaluating the contents of a file. When doing so, @value{GDBN}
28123 can recognize which extension language is being used by looking at
28124 the filename extension. Files with an unrecognized filename extension
28125 are always treated as a @value{GDBN} Command Files.
28126 @xref{Command Files,, Command files}.
28128 You can control how @value{GDBN} evaluates these files with the following
28132 @kindex set script-extension
28133 @kindex show script-extension
28134 @item set script-extension off
28135 All scripts are always evaluated as @value{GDBN} Command Files.
28137 @item set script-extension soft
28138 The debugger determines the scripting language based on filename
28139 extension. If this scripting language is supported, @value{GDBN}
28140 evaluates the script using that language. Otherwise, it evaluates
28141 the file as a @value{GDBN} Command File.
28143 @item set script-extension strict
28144 The debugger determines the scripting language based on filename
28145 extension, and evaluates the script using that language. If the
28146 language is not supported, then the evaluation fails.
28148 @item show script-extension
28149 Display the current value of the @code{script-extension} option.
28153 @ifset SYSTEM_GDBINIT_DIR
28154 This setting is not used for files in the system-wide gdbinit directory.
28155 Files in that directory must have an extension matching their language,
28156 or have a @file{.gdb} extension to be interpreted as regular @value{GDBN}
28157 commands. @xref{Startup}.
28161 @section Canned Sequences of Commands
28163 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
28164 Command Lists}), @value{GDBN} provides two ways to store sequences of
28165 commands for execution as a unit: user-defined commands and command
28169 * Define:: How to define your own commands
28170 * Hooks:: Hooks for user-defined commands
28171 * Command Files:: How to write scripts of commands to be stored in a file
28172 * Output:: Commands for controlled output
28173 * Auto-loading sequences:: Controlling auto-loaded command files
28177 @subsection User-defined Commands
28179 @cindex user-defined command
28180 @cindex arguments, to user-defined commands
28181 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
28182 which you assign a new name as a command. This is done with the
28183 @code{define} command. User commands may accept an unlimited number of arguments
28184 separated by whitespace. Arguments are accessed within the user command
28185 via @code{$arg0@dots{}$argN}. A trivial example:
28189 print $arg0 + $arg1 + $arg2
28194 To execute the command use:
28201 This defines the command @code{adder}, which prints the sum of
28202 its three arguments. Note the arguments are text substitutions, so they may
28203 reference variables, use complex expressions, or even perform inferior
28206 @cindex argument count in user-defined commands
28207 @cindex how many arguments (user-defined commands)
28208 In addition, @code{$argc} may be used to find out how many arguments have
28214 print $arg0 + $arg1
28217 print $arg0 + $arg1 + $arg2
28222 Combining with the @code{eval} command (@pxref{eval}) makes it easier
28223 to process a variable number of arguments:
28230 eval "set $sum = $sum + $arg%d", $i
28240 @item define @var{commandname}
28241 Define a command named @var{commandname}. If there is already a command
28242 by that name, you are asked to confirm that you want to redefine it.
28243 The argument @var{commandname} may be a bare command name consisting of letters,
28244 numbers, dashes, dots, and underscores. It may also start with any
28245 predefined or user-defined prefix command.
28246 For example, @samp{define target my-target} creates
28247 a user-defined @samp{target my-target} command.
28249 The definition of the command is made up of other @value{GDBN} command lines,
28250 which are given following the @code{define} command. The end of these
28251 commands is marked by a line containing @code{end}.
28254 @kindex end@r{ (user-defined commands)}
28255 @item document @var{commandname}
28256 Document the user-defined command @var{commandname}, so that it can be
28257 accessed by @code{help}. The command @var{commandname} must already be
28258 defined. This command reads lines of documentation just as @code{define}
28259 reads the lines of the command definition, ending with @code{end}.
28260 After the @code{document} command is finished, @code{help} on command
28261 @var{commandname} displays the documentation you have written.
28263 You may use the @code{document} command again to change the
28264 documentation of a command. Redefining the command with @code{define}
28265 does not change the documentation.
28267 @kindex define-prefix
28268 @item define-prefix @var{commandname}
28269 Define or mark the command @var{commandname} as a user-defined prefix
28270 command. Once marked, @var{commandname} can be used as prefix command
28271 by the @code{define} command.
28272 Note that @code{define-prefix} can be used with a not yet defined
28273 @var{commandname}. In such a case, @var{commandname} is defined as
28274 an empty user-defined command.
28275 In case you redefine a command that was marked as a user-defined
28276 prefix command, the subcommands of the redefined command are kept
28277 (and @value{GDBN} indicates so to the user).
28281 (@value{GDBP}) define-prefix abc
28282 (@value{GDBP}) define-prefix abc def
28283 (@value{GDBP}) define abc def
28284 Type commands for definition of "abc def".
28285 End with a line saying just "end".
28286 >echo command initial def\n
28288 (@value{GDBP}) define abc def ghi
28289 Type commands for definition of "abc def ghi".
28290 End with a line saying just "end".
28291 >echo command ghi\n
28293 (@value{GDBP}) define abc def
28294 Keeping subcommands of prefix command "def".
28295 Redefine command "def"? (y or n) y
28296 Type commands for definition of "abc def".
28297 End with a line saying just "end".
28298 >echo command def\n
28300 (@value{GDBP}) abc def ghi
28302 (@value{GDBP}) abc def
28307 @kindex dont-repeat
28308 @cindex don't repeat command
28310 Used inside a user-defined command, this tells @value{GDBN} that this
28311 command should not be repeated when the user hits @key{RET}
28312 (@pxref{Command Syntax, repeat last command}).
28314 @kindex help user-defined
28315 @item help user-defined
28316 List all user-defined commands and all python commands defined in class
28317 COMMAND_USER. The first line of the documentation or docstring is
28322 @itemx show user @var{commandname}
28323 Display the @value{GDBN} commands used to define @var{commandname} (but
28324 not its documentation). If no @var{commandname} is given, display the
28325 definitions for all user-defined commands.
28326 This does not work for user-defined python commands.
28328 @cindex infinite recursion in user-defined commands
28329 @kindex show max-user-call-depth
28330 @kindex set max-user-call-depth
28331 @item show max-user-call-depth
28332 @itemx set max-user-call-depth
28333 The value of @code{max-user-call-depth} controls how many recursion
28334 levels are allowed in user-defined commands before @value{GDBN} suspects an
28335 infinite recursion and aborts the command.
28336 This does not apply to user-defined python commands.
28339 In addition to the above commands, user-defined commands frequently
28340 use control flow commands, described in @ref{Command Files}.
28342 When user-defined commands are executed, the
28343 commands of the definition are not printed. An error in any command
28344 stops execution of the user-defined command.
28346 If used interactively, commands that would ask for confirmation proceed
28347 without asking when used inside a user-defined command. Many @value{GDBN}
28348 commands that normally print messages to say what they are doing omit the
28349 messages when used in a user-defined command.
28352 @subsection User-defined Command Hooks
28353 @cindex command hooks
28354 @cindex hooks, for commands
28355 @cindex hooks, pre-command
28358 You may define @dfn{hooks}, which are a special kind of user-defined
28359 command. Whenever you run the command @samp{foo}, if the user-defined
28360 command @samp{hook-foo} exists, it is executed (with no arguments)
28361 before that command.
28363 @cindex hooks, post-command
28365 A hook may also be defined which is run after the command you executed.
28366 Whenever you run the command @samp{foo}, if the user-defined command
28367 @samp{hookpost-foo} exists, it is executed (with no arguments) after
28368 that command. Post-execution hooks may exist simultaneously with
28369 pre-execution hooks, for the same command.
28371 It is valid for a hook to call the command which it hooks. If this
28372 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
28374 @c It would be nice if hookpost could be passed a parameter indicating
28375 @c if the command it hooks executed properly or not. FIXME!
28377 @kindex stop@r{, a pseudo-command}
28378 In addition, a pseudo-command, @samp{stop} exists. Defining
28379 (@samp{hook-stop}) makes the associated commands execute every time
28380 execution stops in your program: before breakpoint commands are run,
28381 displays are printed, or the stack frame is printed.
28383 For example, to ignore @code{SIGALRM} signals while
28384 single-stepping, but treat them normally during normal execution,
28389 handle SIGALRM nopass
28393 handle SIGALRM pass
28396 define hook-continue
28397 handle SIGALRM pass
28401 As a further example, to hook at the beginning and end of the @code{echo}
28402 command, and to add extra text to the beginning and end of the message,
28410 define hookpost-echo
28414 (@value{GDBP}) echo Hello World
28415 <<<---Hello World--->>>
28420 You can define a hook for any single-word command in @value{GDBN}, but
28421 not for command aliases; you should define a hook for the basic command
28422 name, e.g.@: @code{backtrace} rather than @code{bt}.
28423 @c FIXME! So how does Joe User discover whether a command is an alias
28425 You can hook a multi-word command by adding @code{hook-} or
28426 @code{hookpost-} to the last word of the command, e.g.@:
28427 @samp{define target hook-remote} to add a hook to @samp{target remote}.
28429 If an error occurs during the execution of your hook, execution of
28430 @value{GDBN} commands stops and @value{GDBN} issues a prompt
28431 (before the command that you actually typed had a chance to run).
28433 If you try to define a hook which does not match any known command, you
28434 get a warning from the @code{define} command.
28436 @node Command Files
28437 @subsection Command Files
28439 @cindex command files
28440 @cindex scripting commands
28441 A command file for @value{GDBN} is a text file made of lines that are
28442 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
28443 also be included. An empty line in a command file does nothing; it
28444 does not mean to repeat the last command, as it would from the
28447 You can request the execution of a command file with the @code{source}
28448 command. Note that the @code{source} command is also used to evaluate
28449 scripts that are not Command Files. The exact behavior can be configured
28450 using the @code{script-extension} setting.
28451 @xref{Extending GDB,, Extending GDB}.
28455 @cindex execute commands from a file
28456 @item source [-s] [-v] @var{filename}
28457 Execute the command file @var{filename}.
28460 The lines in a command file are generally executed sequentially,
28461 unless the order of execution is changed by one of the
28462 @emph{flow-control commands} described below. The commands are not
28463 printed as they are executed. An error in any command terminates
28464 execution of the command file and control is returned to the console.
28466 @value{GDBN} first searches for @var{filename} in the current directory.
28467 If the file is not found there, and @var{filename} does not specify a
28468 directory, then @value{GDBN} also looks for the file on the source search path
28469 (specified with the @samp{directory} command);
28470 except that @file{$cdir} is not searched because the compilation directory
28471 is not relevant to scripts.
28473 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
28474 on the search path even if @var{filename} specifies a directory.
28475 The search is done by appending @var{filename} to each element of the
28476 search path. So, for example, if @var{filename} is @file{mylib/myscript}
28477 and the search path contains @file{/home/user} then @value{GDBN} will
28478 look for the script @file{/home/user/mylib/myscript}.
28479 The search is also done if @var{filename} is an absolute path.
28480 For example, if @var{filename} is @file{/tmp/myscript} and
28481 the search path contains @file{/home/user} then @value{GDBN} will
28482 look for the script @file{/home/user/tmp/myscript}.
28483 For DOS-like systems, if @var{filename} contains a drive specification,
28484 it is stripped before concatenation. For example, if @var{filename} is
28485 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
28486 will look for the script @file{c:/tmp/myscript}.
28488 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
28489 each command as it is executed. The option must be given before
28490 @var{filename}, and is interpreted as part of the filename anywhere else.
28492 Commands that would ask for confirmation if used interactively proceed
28493 without asking when used in a command file. Many @value{GDBN} commands that
28494 normally print messages to say what they are doing omit the messages
28495 when called from command files.
28497 @value{GDBN} also accepts command input from standard input. In this
28498 mode, normal output goes to standard output and error output goes to
28499 standard error. Errors in a command file supplied on standard input do
28500 not terminate execution of the command file---execution continues with
28504 gdb < cmds > log 2>&1
28507 (The syntax above will vary depending on the shell used.) This example
28508 will execute commands from the file @file{cmds}. All output and errors
28509 would be directed to @file{log}.
28511 Since commands stored on command files tend to be more general than
28512 commands typed interactively, they frequently need to deal with
28513 complicated situations, such as different or unexpected values of
28514 variables and symbols, changes in how the program being debugged is
28515 built, etc. @value{GDBN} provides a set of flow-control commands to
28516 deal with these complexities. Using these commands, you can write
28517 complex scripts that loop over data structures, execute commands
28518 conditionally, etc.
28525 This command allows to include in your script conditionally executed
28526 commands. The @code{if} command takes a single argument, which is an
28527 expression to evaluate. It is followed by a series of commands that
28528 are executed only if the expression is true (its value is nonzero).
28529 There can then optionally be an @code{else} line, followed by a series
28530 of commands that are only executed if the expression was false. The
28531 end of the list is marked by a line containing @code{end}.
28535 This command allows to write loops. Its syntax is similar to
28536 @code{if}: the command takes a single argument, which is an expression
28537 to evaluate, and must be followed by the commands to execute, one per
28538 line, terminated by an @code{end}. These commands are called the
28539 @dfn{body} of the loop. The commands in the body of @code{while} are
28540 executed repeatedly as long as the expression evaluates to true.
28544 This command exits the @code{while} loop in whose body it is included.
28545 Execution of the script continues after that @code{while}s @code{end}
28548 @kindex loop_continue
28549 @item loop_continue
28550 This command skips the execution of the rest of the body of commands
28551 in the @code{while} loop in whose body it is included. Execution
28552 branches to the beginning of the @code{while} loop, where it evaluates
28553 the controlling expression.
28555 @kindex end@r{ (if/else/while commands)}
28557 Terminate the block of commands that are the body of @code{if},
28558 @code{else}, or @code{while} flow-control commands.
28563 @subsection Commands for Controlled Output
28565 During the execution of a command file or a user-defined command, normal
28566 @value{GDBN} output is suppressed; the only output that appears is what is
28567 explicitly printed by the commands in the definition. This section
28568 describes three commands useful for generating exactly the output you
28573 @item echo @var{text}
28574 @c I do not consider backslash-space a standard C escape sequence
28575 @c because it is not in ANSI.
28576 Print @var{text}. Nonprinting characters can be included in
28577 @var{text} using C escape sequences, such as @samp{\n} to print a
28578 newline. @strong{No newline is printed unless you specify one.}
28579 In addition to the standard C escape sequences, a backslash followed
28580 by a space stands for a space. This is useful for displaying a
28581 string with spaces at the beginning or the end, since leading and
28582 trailing spaces are otherwise trimmed from all arguments.
28583 To print @samp{@w{ }and foo =@w{ }}, use the command
28584 @samp{echo \@w{ }and foo = \@w{ }}.
28586 A backslash at the end of @var{text} can be used, as in C, to continue
28587 the command onto subsequent lines. For example,
28590 echo This is some text\n\
28591 which is continued\n\
28592 onto several lines.\n
28595 produces the same output as
28598 echo This is some text\n
28599 echo which is continued\n
28600 echo onto several lines.\n
28604 @item output @var{expression}
28605 Print the value of @var{expression} and nothing but that value: no
28606 newlines, no @samp{$@var{nn} = }. The value is not entered in the
28607 value history either. @xref{Expressions, ,Expressions}, for more information
28610 @item output/@var{fmt} @var{expression}
28611 Print the value of @var{expression} in format @var{fmt}. You can use
28612 the same formats as for @code{print}. @xref{Output Formats,,Output
28613 Formats}, for more information.
28616 @item printf @var{template}, @var{expressions}@dots{}
28617 Print the values of one or more @var{expressions} under the control of
28618 the string @var{template}. To print several values, make
28619 @var{expressions} be a comma-separated list of individual expressions,
28620 which may be either numbers or pointers. Their values are printed as
28621 specified by @var{template}, exactly as a C program would do by
28622 executing the code below:
28625 printf (@var{template}, @var{expressions}@dots{});
28628 As in @code{C} @code{printf}, ordinary characters in @var{template}
28629 are printed verbatim, while @dfn{conversion specification} introduced
28630 by the @samp{%} character cause subsequent @var{expressions} to be
28631 evaluated, their values converted and formatted according to type and
28632 style information encoded in the conversion specifications, and then
28635 For example, you can print two values in hex like this:
28638 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
28641 @code{printf} supports all the standard @code{C} conversion
28642 specifications, including the flags and modifiers between the @samp{%}
28643 character and the conversion letter, with the following exceptions:
28647 The argument-ordering modifiers, such as @samp{2$}, are not supported.
28650 The modifier @samp{*} is not supported for specifying precision or
28654 The @samp{'} flag (for separation of digits into groups according to
28655 @code{LC_NUMERIC'}) is not supported.
28658 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
28662 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
28665 The conversion letters @samp{a} and @samp{A} are not supported.
28669 Note that the @samp{ll} type modifier is supported only if the
28670 underlying @code{C} implementation used to build @value{GDBN} supports
28671 the @code{long long int} type, and the @samp{L} type modifier is
28672 supported only if @code{long double} type is available.
28674 As in @code{C}, @code{printf} supports simple backslash-escape
28675 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
28676 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
28677 single character. Octal and hexadecimal escape sequences are not
28680 Additionally, @code{printf} supports conversion specifications for DFP
28681 (@dfn{Decimal Floating Point}) types using the following length modifiers
28682 together with a floating point specifier.
28687 @samp{H} for printing @code{Decimal32} types.
28690 @samp{D} for printing @code{Decimal64} types.
28693 @samp{DD} for printing @code{Decimal128} types.
28696 If the underlying @code{C} implementation used to build @value{GDBN} has
28697 support for the three length modifiers for DFP types, other modifiers
28698 such as width and precision will also be available for @value{GDBN} to use.
28700 In case there is no such @code{C} support, no additional modifiers will be
28701 available and the value will be printed in the standard way.
28703 Here's an example of printing DFP types using the above conversion letters:
28705 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
28710 @item eval @var{template}, @var{expressions}@dots{}
28711 Convert the values of one or more @var{expressions} under the control of
28712 the string @var{template} to a command line, and call it.
28716 @node Auto-loading sequences
28717 @subsection Controlling auto-loading native @value{GDBN} scripts
28718 @cindex native script auto-loading
28720 When a new object file is read (for example, due to the @code{file}
28721 command, or because the inferior has loaded a shared library),
28722 @value{GDBN} will look for the command file @file{@var{objfile}-gdb.gdb}.
28723 @xref{Auto-loading extensions}.
28725 Auto-loading can be enabled or disabled,
28726 and the list of auto-loaded scripts can be printed.
28729 @anchor{set auto-load gdb-scripts}
28730 @kindex set auto-load gdb-scripts
28731 @item set auto-load gdb-scripts [on|off]
28732 Enable or disable the auto-loading of canned sequences of commands scripts.
28734 @anchor{show auto-load gdb-scripts}
28735 @kindex show auto-load gdb-scripts
28736 @item show auto-load gdb-scripts
28737 Show whether auto-loading of canned sequences of commands scripts is enabled or
28740 @anchor{info auto-load gdb-scripts}
28741 @kindex info auto-load gdb-scripts
28742 @cindex print list of auto-loaded canned sequences of commands scripts
28743 @item info auto-load gdb-scripts [@var{regexp}]
28744 Print the list of all canned sequences of commands scripts that @value{GDBN}
28748 If @var{regexp} is supplied only canned sequences of commands scripts with
28749 matching names are printed.
28751 @c Python docs live in a separate file.
28752 @include python.texi
28754 @c Guile docs live in a separate file.
28755 @include guile.texi
28757 @node Auto-loading extensions
28758 @section Auto-loading extensions
28759 @cindex auto-loading extensions
28761 @value{GDBN} provides two mechanisms for automatically loading extensions
28762 when a new object file is read (for example, due to the @code{file}
28763 command, or because the inferior has loaded a shared library):
28764 @file{@var{objfile}-gdb.@var{ext}} and the @code{.debug_gdb_scripts}
28765 section of modern file formats like ELF.
28768 * objfile-gdb.ext file: objfile-gdbdotext file. The @file{@var{objfile}-gdb.@var{ext}} file
28769 * .debug_gdb_scripts section: dotdebug_gdb_scripts section. The @code{.debug_gdb_scripts} section
28770 * Which flavor to choose?::
28773 The auto-loading feature is useful for supplying application-specific
28774 debugging commands and features.
28776 Auto-loading can be enabled or disabled,
28777 and the list of auto-loaded scripts can be printed.
28778 See the @samp{auto-loading} section of each extension language
28779 for more information.
28780 For @value{GDBN} command files see @ref{Auto-loading sequences}.
28781 For Python files see @ref{Python Auto-loading}.
28783 Note that loading of this script file also requires accordingly configured
28784 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
28786 @node objfile-gdbdotext file
28787 @subsection The @file{@var{objfile}-gdb.@var{ext}} file
28788 @cindex @file{@var{objfile}-gdb.gdb}
28789 @cindex @file{@var{objfile}-gdb.py}
28790 @cindex @file{@var{objfile}-gdb.scm}
28792 When a new object file is read, @value{GDBN} looks for a file named
28793 @file{@var{objfile}-gdb.@var{ext}} (we call it @var{script-name} below),
28794 where @var{objfile} is the object file's name and
28795 where @var{ext} is the file extension for the extension language:
28798 @item @file{@var{objfile}-gdb.gdb}
28799 GDB's own command language
28800 @item @file{@var{objfile}-gdb.py}
28802 @item @file{@var{objfile}-gdb.scm}
28806 @var{script-name} is formed by ensuring that the file name of @var{objfile}
28807 is absolute, following all symlinks, and resolving @code{.} and @code{..}
28808 components, and appending the @file{-gdb.@var{ext}} suffix.
28809 If this file exists and is readable, @value{GDBN} will evaluate it as a
28810 script in the specified extension language.
28812 If this file does not exist, then @value{GDBN} will look for
28813 @var{script-name} file in all of the directories as specified below.
28815 Note that loading of these files requires an accordingly configured
28816 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
28818 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
28819 scripts normally according to its @file{.exe} filename. But if no scripts are
28820 found @value{GDBN} also tries script filenames matching the object file without
28821 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
28822 is attempted on any platform. This makes the script filenames compatible
28823 between Unix and MS-Windows hosts.
28826 @anchor{set auto-load scripts-directory}
28827 @kindex set auto-load scripts-directory
28828 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
28829 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
28830 may be delimited by the host platform path separator in use
28831 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
28833 Each entry here needs to be covered also by the security setting
28834 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
28836 @anchor{with-auto-load-dir}
28837 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
28838 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
28839 configuration option @option{--with-auto-load-dir}.
28841 Any reference to @file{$debugdir} will get replaced by
28842 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
28843 reference to @file{$datadir} will get replaced by @var{data-directory} which is
28844 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
28845 @file{$datadir} must be placed as a directory component --- either alone or
28846 delimited by @file{/} or @file{\} directory separators, depending on the host
28849 The list of directories uses path separator (@samp{:} on GNU and Unix
28850 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
28851 to the @env{PATH} environment variable.
28853 @anchor{show auto-load scripts-directory}
28854 @kindex show auto-load scripts-directory
28855 @item show auto-load scripts-directory
28856 Show @value{GDBN} auto-loaded scripts location.
28858 @anchor{add-auto-load-scripts-directory}
28859 @kindex add-auto-load-scripts-directory
28860 @item add-auto-load-scripts-directory @r{[}@var{directories}@dots{}@r{]}
28861 Add an entry (or list of entries) to the list of auto-loaded scripts locations.
28862 Multiple entries may be delimited by the host platform path separator in use.
28865 @value{GDBN} does not track which files it has already auto-loaded this way.
28866 @value{GDBN} will load the associated script every time the corresponding
28867 @var{objfile} is opened.
28868 So your @file{-gdb.@var{ext}} file should be careful to avoid errors if it
28869 is evaluated more than once.
28871 @node dotdebug_gdb_scripts section
28872 @subsection The @code{.debug_gdb_scripts} section
28873 @cindex @code{.debug_gdb_scripts} section
28875 For systems using file formats like ELF and COFF,
28876 when @value{GDBN} loads a new object file
28877 it will look for a special section named @code{.debug_gdb_scripts}.
28878 If this section exists, its contents is a list of null-terminated entries
28879 specifying scripts to load. Each entry begins with a non-null prefix byte that
28880 specifies the kind of entry, typically the extension language and whether the
28881 script is in a file or inlined in @code{.debug_gdb_scripts}.
28883 The following entries are supported:
28886 @item SECTION_SCRIPT_ID_PYTHON_FILE = 1
28887 @item SECTION_SCRIPT_ID_SCHEME_FILE = 3
28888 @item SECTION_SCRIPT_ID_PYTHON_TEXT = 4
28889 @item SECTION_SCRIPT_ID_SCHEME_TEXT = 6
28892 @subsubsection Script File Entries
28894 If the entry specifies a file, @value{GDBN} will look for the file first
28895 in the current directory and then along the source search path
28896 (@pxref{Source Path, ,Specifying Source Directories}),
28897 except that @file{$cdir} is not searched, since the compilation
28898 directory is not relevant to scripts.
28900 File entries can be placed in section @code{.debug_gdb_scripts} with,
28901 for example, this GCC macro for Python scripts.
28904 /* Note: The "MS" section flags are to remove duplicates. */
28905 #define DEFINE_GDB_PY_SCRIPT(script_name) \
28907 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
28908 .byte 1 /* Python */\n\
28909 .asciz \"" script_name "\"\n\
28915 For Guile scripts, replace @code{.byte 1} with @code{.byte 3}.
28916 Then one can reference the macro in a header or source file like this:
28919 DEFINE_GDB_PY_SCRIPT ("my-app-scripts.py")
28922 The script name may include directories if desired.
28924 Note that loading of this script file also requires accordingly configured
28925 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
28927 If the macro invocation is put in a header, any application or library
28928 using this header will get a reference to the specified script,
28929 and with the use of @code{"MS"} attributes on the section, the linker
28930 will remove duplicates.
28932 @subsubsection Script Text Entries
28934 Script text entries allow to put the executable script in the entry
28935 itself instead of loading it from a file.
28936 The first line of the entry, everything after the prefix byte and up to
28937 the first newline (@code{0xa}) character, is the script name, and must not
28938 contain any kind of space character, e.g., spaces or tabs.
28939 The rest of the entry, up to the trailing null byte, is the script to
28940 execute in the specified language. The name needs to be unique among
28941 all script names, as @value{GDBN} executes each script only once based
28944 Here is an example from file @file{py-section-script.c} in the @value{GDBN}
28948 #include "symcat.h"
28949 #include "gdb/section-scripts.h"
28951 ".pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n"
28952 ".byte " XSTRING (SECTION_SCRIPT_ID_PYTHON_TEXT) "\n"
28953 ".ascii \"gdb.inlined-script\\n\"\n"
28954 ".ascii \"class test_cmd (gdb.Command):\\n\"\n"
28955 ".ascii \" def __init__ (self):\\n\"\n"
28956 ".ascii \" super (test_cmd, self).__init__ ("
28957 "\\\"test-cmd\\\", gdb.COMMAND_OBSCURE)\\n\"\n"
28958 ".ascii \" def invoke (self, arg, from_tty):\\n\"\n"
28959 ".ascii \" print (\\\"test-cmd output, arg = %s\\\" % arg)\\n\"\n"
28960 ".ascii \"test_cmd ()\\n\"\n"
28966 Loading of inlined scripts requires a properly configured
28967 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
28968 The path to specify in @code{auto-load safe-path} is the path of the file
28969 containing the @code{.debug_gdb_scripts} section.
28971 @node Which flavor to choose?
28972 @subsection Which flavor to choose?
28974 Given the multiple ways of auto-loading extensions, it might not always
28975 be clear which one to choose. This section provides some guidance.
28978 Benefits of the @file{-gdb.@var{ext}} way:
28982 Can be used with file formats that don't support multiple sections.
28985 Ease of finding scripts for public libraries.
28987 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
28988 in the source search path.
28989 For publicly installed libraries, e.g., @file{libstdc++}, there typically
28990 isn't a source directory in which to find the script.
28993 Doesn't require source code additions.
28997 Benefits of the @code{.debug_gdb_scripts} way:
29001 Works with static linking.
29003 Scripts for libraries done the @file{-gdb.@var{ext}} way require an objfile to
29004 trigger their loading. When an application is statically linked the only
29005 objfile available is the executable, and it is cumbersome to attach all the
29006 scripts from all the input libraries to the executable's
29007 @file{-gdb.@var{ext}} script.
29010 Works with classes that are entirely inlined.
29012 Some classes can be entirely inlined, and thus there may not be an associated
29013 shared library to attach a @file{-gdb.@var{ext}} script to.
29016 Scripts needn't be copied out of the source tree.
29018 In some circumstances, apps can be built out of large collections of internal
29019 libraries, and the build infrastructure necessary to install the
29020 @file{-gdb.@var{ext}} scripts in a place where @value{GDBN} can find them is
29021 cumbersome. It may be easier to specify the scripts in the
29022 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
29023 top of the source tree to the source search path.
29026 @node Multiple Extension Languages
29027 @section Multiple Extension Languages
29029 The Guile and Python extension languages do not share any state,
29030 and generally do not interfere with each other.
29031 There are some things to be aware of, however.
29033 @subsection Python comes first
29035 Python was @value{GDBN}'s first extension language, and to avoid breaking
29036 existing behaviour Python comes first. This is generally solved by the
29037 ``first one wins'' principle. @value{GDBN} maintains a list of enabled
29038 extension languages, and when it makes a call to an extension language,
29039 (say to pretty-print a value), it tries each in turn until an extension
29040 language indicates it has performed the request (e.g., has returned the
29041 pretty-printed form of a value).
29042 This extends to errors while performing such requests: If an error happens
29043 while, for example, trying to pretty-print an object then the error is
29044 reported and any following extension languages are not tried.
29047 @section Creating new spellings of existing commands
29048 @cindex aliases for commands
29050 It is often useful to define alternate spellings of existing commands.
29051 For example, if a new @value{GDBN} command defined in Python has
29052 a long name to type, it is handy to have an abbreviated version of it
29053 that involves less typing.
29055 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
29056 of the @samp{step} command even though it is otherwise an ambiguous
29057 abbreviation of other commands like @samp{set} and @samp{show}.
29059 Aliases are also used to provide shortened or more common versions
29060 of multi-word commands. For example, @value{GDBN} provides the
29061 @samp{tty} alias of the @samp{set inferior-tty} command.
29063 You can define a new alias with the @samp{alias} command.
29068 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
29072 @var{ALIAS} specifies the name of the new alias.
29073 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
29076 @var{COMMAND} specifies the name of an existing command
29077 that is being aliased.
29079 The @samp{-a} option specifies that the new alias is an abbreviation
29080 of the command. Abbreviations are not shown in command
29081 lists displayed by the @samp{help} command.
29083 The @samp{--} option specifies the end of options,
29084 and is useful when @var{ALIAS} begins with a dash.
29086 Here is a simple example showing how to make an abbreviation
29087 of a command so that there is less to type.
29088 Suppose you were tired of typing @samp{disas}, the current
29089 shortest unambiguous abbreviation of the @samp{disassemble} command
29090 and you wanted an even shorter version named @samp{di}.
29091 The following will accomplish this.
29094 (@value{GDBP}) alias -a di = disas
29097 Note that aliases are different from user-defined commands.
29098 With a user-defined command, you also need to write documentation
29099 for it with the @samp{document} command.
29100 An alias automatically picks up the documentation of the existing command.
29102 Here is an example where we make @samp{elms} an abbreviation of
29103 @samp{elements} in the @samp{set print elements} command.
29104 This is to show that you can make an abbreviation of any part
29108 (@value{GDBP}) alias -a set print elms = set print elements
29109 (@value{GDBP}) alias -a show print elms = show print elements
29110 (@value{GDBP}) set p elms 20
29111 (@value{GDBP}) show p elms
29112 Limit on string chars or array elements to print is 200.
29115 Note that if you are defining an alias of a @samp{set} command,
29116 and you want to have an alias for the corresponding @samp{show}
29117 command, then you need to define the latter separately.
29119 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
29120 @var{ALIAS}, just as they are normally.
29123 (@value{GDBP}) alias -a set pr elms = set p ele
29126 Finally, here is an example showing the creation of a one word
29127 alias for a more complex command.
29128 This creates alias @samp{spe} of the command @samp{set print elements}.
29131 (@value{GDBP}) alias spe = set print elements
29132 (@value{GDBP}) spe 20
29136 @chapter Command Interpreters
29137 @cindex command interpreters
29139 @value{GDBN} supports multiple command interpreters, and some command
29140 infrastructure to allow users or user interface writers to switch
29141 between interpreters or run commands in other interpreters.
29143 @value{GDBN} currently supports two command interpreters, the console
29144 interpreter (sometimes called the command-line interpreter or @sc{cli})
29145 and the machine interface interpreter (or @sc{gdb/mi}). This manual
29146 describes both of these interfaces in great detail.
29148 By default, @value{GDBN} will start with the console interpreter.
29149 However, the user may choose to start @value{GDBN} with another
29150 interpreter by specifying the @option{-i} or @option{--interpreter}
29151 startup options. Defined interpreters include:
29155 @cindex console interpreter
29156 The traditional console or command-line interpreter. This is the most often
29157 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
29158 @value{GDBN} will use this interpreter.
29161 @cindex mi interpreter
29162 The newest @sc{gdb/mi} interface (currently @code{mi3}). Used primarily
29163 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
29164 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
29168 @cindex mi3 interpreter
29169 The @sc{gdb/mi} interface introduced in @value{GDBN} 9.1.
29172 @cindex mi2 interpreter
29173 The @sc{gdb/mi} interface introduced in @value{GDBN} 6.0.
29176 @cindex mi1 interpreter
29177 The @sc{gdb/mi} interface introduced in @value{GDBN} 5.1.
29181 @cindex invoke another interpreter
29183 @kindex interpreter-exec
29184 You may execute commands in any interpreter from the current
29185 interpreter using the appropriate command. If you are running the
29186 console interpreter, simply use the @code{interpreter-exec} command:
29189 interpreter-exec mi "-data-list-register-names"
29192 @sc{gdb/mi} has a similar command, although it is only available in versions of
29193 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
29195 Note that @code{interpreter-exec} only changes the interpreter for the
29196 duration of the specified command. It does not change the interpreter
29199 @cindex start a new independent interpreter
29201 Although you may only choose a single interpreter at startup, it is
29202 possible to run an independent interpreter on a specified input/output
29203 device (usually a tty).
29205 For example, consider a debugger GUI or IDE that wants to provide a
29206 @value{GDBN} console view. It may do so by embedding a terminal
29207 emulator widget in its GUI, starting @value{GDBN} in the traditional
29208 command-line mode with stdin/stdout/stderr redirected to that
29209 terminal, and then creating an MI interpreter running on a specified
29210 input/output device. The console interpreter created by @value{GDBN}
29211 at startup handles commands the user types in the terminal widget,
29212 while the GUI controls and synchronizes state with @value{GDBN} using
29213 the separate MI interpreter.
29215 To start a new secondary @dfn{user interface} running MI, use the
29216 @code{new-ui} command:
29219 @cindex new user interface
29221 new-ui @var{interpreter} @var{tty}
29224 The @var{interpreter} parameter specifies the interpreter to run.
29225 This accepts the same values as the @code{interpreter-exec} command.
29226 For example, @samp{console}, @samp{mi}, @samp{mi2}, etc. The
29227 @var{tty} parameter specifies the name of the bidirectional file the
29228 interpreter uses for input/output, usually the name of a
29229 pseudoterminal slave on Unix systems. For example:
29232 (@value{GDBP}) new-ui mi /dev/pts/9
29236 runs an MI interpreter on @file{/dev/pts/9}.
29239 @chapter @value{GDBN} Text User Interface
29241 @cindex Text User Interface
29244 * TUI Overview:: TUI overview
29245 * TUI Keys:: TUI key bindings
29246 * TUI Single Key Mode:: TUI single key mode
29247 * TUI Commands:: TUI-specific commands
29248 * TUI Configuration:: TUI configuration variables
29251 The @value{GDBN} Text User Interface (TUI) is a terminal
29252 interface which uses the @code{curses} library to show the source
29253 file, the assembly output, the program registers and @value{GDBN}
29254 commands in separate text windows. The TUI mode is supported only
29255 on platforms where a suitable version of the @code{curses} library
29258 The TUI mode is enabled by default when you invoke @value{GDBN} as
29259 @samp{@value{GDBP} -tui}.
29260 You can also switch in and out of TUI mode while @value{GDBN} runs by
29261 using various TUI commands and key bindings, such as @command{tui
29262 enable} or @kbd{C-x C-a}. @xref{TUI Commands, ,TUI Commands}, and
29263 @ref{TUI Keys, ,TUI Key Bindings}.
29266 @section TUI Overview
29268 In TUI mode, @value{GDBN} can display several text windows:
29272 This window is the @value{GDBN} command window with the @value{GDBN}
29273 prompt and the @value{GDBN} output. The @value{GDBN} input is still
29274 managed using readline.
29277 The source window shows the source file of the program. The current
29278 line and active breakpoints are displayed in this window.
29281 The assembly window shows the disassembly output of the program.
29284 This window shows the processor registers. Registers are highlighted
29285 when their values change.
29288 The source and assembly windows show the current program position
29289 by highlighting the current line and marking it with a @samp{>} marker.
29290 Breakpoints are indicated with two markers. The first marker
29291 indicates the breakpoint type:
29295 Breakpoint which was hit at least once.
29298 Breakpoint which was never hit.
29301 Hardware breakpoint which was hit at least once.
29304 Hardware breakpoint which was never hit.
29307 The second marker indicates whether the breakpoint is enabled or not:
29311 Breakpoint is enabled.
29314 Breakpoint is disabled.
29317 The source, assembly and register windows are updated when the current
29318 thread changes, when the frame changes, or when the program counter
29321 These windows are not all visible at the same time. The command
29322 window is always visible. The others can be arranged in several
29333 source and assembly,
29336 source and registers, or
29339 assembly and registers.
29342 A status line above the command window shows the following information:
29346 Indicates the current @value{GDBN} target.
29347 (@pxref{Targets, ,Specifying a Debugging Target}).
29350 Gives the current process or thread number.
29351 When no process is being debugged, this field is set to @code{No process}.
29354 Gives the current function name for the selected frame.
29355 The name is demangled if demangling is turned on (@pxref{Print Settings}).
29356 When there is no symbol corresponding to the current program counter,
29357 the string @code{??} is displayed.
29360 Indicates the current line number for the selected frame.
29361 When the current line number is not known, the string @code{??} is displayed.
29364 Indicates the current program counter address.
29368 @section TUI Key Bindings
29369 @cindex TUI key bindings
29371 The TUI installs several key bindings in the readline keymaps
29372 @ifset SYSTEM_READLINE
29373 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
29375 @ifclear SYSTEM_READLINE
29376 (@pxref{Command Line Editing}).
29378 The following key bindings are installed for both TUI mode and the
29379 @value{GDBN} standard mode.
29388 Enter or leave the TUI mode. When leaving the TUI mode,
29389 the curses window management stops and @value{GDBN} operates using
29390 its standard mode, writing on the terminal directly. When reentering
29391 the TUI mode, control is given back to the curses windows.
29392 The screen is then refreshed.
29394 This key binding uses the bindable Readline function
29395 @code{tui-switch-mode}.
29399 Use a TUI layout with only one window. The layout will
29400 either be @samp{source} or @samp{assembly}. When the TUI mode
29401 is not active, it will switch to the TUI mode.
29403 Think of this key binding as the Emacs @kbd{C-x 1} binding.
29405 This key binding uses the bindable Readline function
29406 @code{tui-delete-other-windows}.
29410 Use a TUI layout with at least two windows. When the current
29411 layout already has two windows, the next layout with two windows is used.
29412 When a new layout is chosen, one window will always be common to the
29413 previous layout and the new one.
29415 Think of it as the Emacs @kbd{C-x 2} binding.
29417 This key binding uses the bindable Readline function
29418 @code{tui-change-windows}.
29422 Change the active window. The TUI associates several key bindings
29423 (like scrolling and arrow keys) with the active window. This command
29424 gives the focus to the next TUI window.
29426 Think of it as the Emacs @kbd{C-x o} binding.
29428 This key binding uses the bindable Readline function
29429 @code{tui-other-window}.
29433 Switch in and out of the TUI SingleKey mode that binds single
29434 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
29436 This key binding uses the bindable Readline function
29437 @code{next-keymap}.
29440 The following key bindings only work in the TUI mode:
29445 Scroll the active window one page up.
29449 Scroll the active window one page down.
29453 Scroll the active window one line up.
29457 Scroll the active window one line down.
29461 Scroll the active window one column left.
29465 Scroll the active window one column right.
29469 Refresh the screen.
29472 Because the arrow keys scroll the active window in the TUI mode, they
29473 are not available for their normal use by readline unless the command
29474 window has the focus. When another window is active, you must use
29475 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
29476 and @kbd{C-f} to control the command window.
29478 @node TUI Single Key Mode
29479 @section TUI Single Key Mode
29480 @cindex TUI single key mode
29482 The TUI also provides a @dfn{SingleKey} mode, which binds several
29483 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
29484 switch into this mode, where the following key bindings are used:
29487 @kindex c @r{(SingleKey TUI key)}
29491 @kindex d @r{(SingleKey TUI key)}
29495 @kindex f @r{(SingleKey TUI key)}
29499 @kindex n @r{(SingleKey TUI key)}
29503 @kindex o @r{(SingleKey TUI key)}
29505 nexti. The shortcut letter @samp{o} stands for ``step Over''.
29507 @kindex q @r{(SingleKey TUI key)}
29509 exit the SingleKey mode.
29511 @kindex r @r{(SingleKey TUI key)}
29515 @kindex s @r{(SingleKey TUI key)}
29519 @kindex i @r{(SingleKey TUI key)}
29521 stepi. The shortcut letter @samp{i} stands for ``step Into''.
29523 @kindex u @r{(SingleKey TUI key)}
29527 @kindex v @r{(SingleKey TUI key)}
29531 @kindex w @r{(SingleKey TUI key)}
29536 Other keys temporarily switch to the @value{GDBN} command prompt.
29537 The key that was pressed is inserted in the editing buffer so that
29538 it is possible to type most @value{GDBN} commands without interaction
29539 with the TUI SingleKey mode. Once the command is entered the TUI
29540 SingleKey mode is restored. The only way to permanently leave
29541 this mode is by typing @kbd{q} or @kbd{C-x s}.
29543 @cindex SingleKey keymap name
29544 If @value{GDBN} was built with Readline 8.0 or later, the TUI
29545 SingleKey keymap will be named @samp{SingleKey}. This can be used in
29546 @file{.inputrc} to add additional bindings to this keymap.
29549 @section TUI-specific Commands
29550 @cindex TUI commands
29552 The TUI has specific commands to control the text windows.
29553 These commands are always available, even when @value{GDBN} is not in
29554 the TUI mode. When @value{GDBN} is in the standard mode, most
29555 of these commands will automatically switch to the TUI mode.
29557 Note that if @value{GDBN}'s @code{stdout} is not connected to a
29558 terminal, or @value{GDBN} has been started with the machine interface
29559 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
29560 these commands will fail with an error, because it would not be
29561 possible or desirable to enable curses window management.
29566 Activate TUI mode. The last active TUI window layout will be used if
29567 TUI mode has previously been used in the current debugging session,
29568 otherwise a default layout is used.
29571 @kindex tui disable
29572 Disable TUI mode, returning to the console interpreter.
29576 List and give the size of all displayed windows.
29578 @item layout @var{name}
29580 Changes which TUI windows are displayed. In each layout the command
29581 window is always displayed, the @var{name} parameter controls which
29582 additional windows are displayed, and can be any of the following:
29586 Display the next layout.
29589 Display the previous layout.
29592 Display the source and command windows.
29595 Display the assembly and command windows.
29598 Display the source, assembly, and command windows.
29601 When in @code{src} layout display the register, source, and command
29602 windows. When in @code{asm} or @code{split} layout display the
29603 register, assembler, and command windows.
29606 @item focus @var{name}
29608 Changes which TUI window is currently active for scrolling. The
29609 @var{name} parameter can be any of the following:
29613 Make the next window active for scrolling.
29616 Make the previous window active for scrolling.
29619 Make the source window active for scrolling.
29622 Make the assembly window active for scrolling.
29625 Make the register window active for scrolling.
29628 Make the command window active for scrolling.
29633 Refresh the screen. This is similar to typing @kbd{C-L}.
29635 @item tui reg @var{group}
29637 Changes the register group displayed in the tui register window to
29638 @var{group}. If the register window is not currently displayed this
29639 command will cause the register window to be displayed. The list of
29640 register groups, as well as their order is target specific. The
29641 following groups are available on most targets:
29644 Repeatedly selecting this group will cause the display to cycle
29645 through all of the available register groups.
29648 Repeatedly selecting this group will cause the display to cycle
29649 through all of the available register groups in the reverse order to
29653 Display the general registers.
29655 Display the floating point registers.
29657 Display the system registers.
29659 Display the vector registers.
29661 Display all registers.
29666 Update the source window and the current execution point.
29668 @item winheight @var{name} +@var{count}
29669 @itemx winheight @var{name} -@var{count}
29671 Change the height of the window @var{name} by @var{count}
29672 lines. Positive counts increase the height, while negative counts
29673 decrease it. The @var{name} parameter can be one of @code{src} (the
29674 source window), @code{cmd} (the command window), @code{asm} (the
29675 disassembly window), or @code{regs} (the register display window).
29678 @node TUI Configuration
29679 @section TUI Configuration Variables
29680 @cindex TUI configuration variables
29682 Several configuration variables control the appearance of TUI windows.
29685 @item set tui border-kind @var{kind}
29686 @kindex set tui border-kind
29687 Select the border appearance for the source, assembly and register windows.
29688 The possible values are the following:
29691 Use a space character to draw the border.
29694 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
29697 Use the Alternate Character Set to draw the border. The border is
29698 drawn using character line graphics if the terminal supports them.
29701 @item set tui border-mode @var{mode}
29702 @kindex set tui border-mode
29703 @itemx set tui active-border-mode @var{mode}
29704 @kindex set tui active-border-mode
29705 Select the display attributes for the borders of the inactive windows
29706 or the active window. The @var{mode} can be one of the following:
29709 Use normal attributes to display the border.
29715 Use reverse video mode.
29718 Use half bright mode.
29720 @item half-standout
29721 Use half bright and standout mode.
29724 Use extra bright or bold mode.
29726 @item bold-standout
29727 Use extra bright or bold and standout mode.
29730 @item set tui tab-width @var{nchars}
29731 @kindex set tui tab-width
29733 Set the width of tab stops to be @var{nchars} characters. This
29734 setting affects the display of TAB characters in the source and
29737 @item set tui compact-source @r{[}on@r{|}off@r{]}
29738 @kindex set tui compact-source
29739 Set whether the TUI source window is displayed in ``compact'' form.
29740 The default display uses more space for line numbers and starts the
29741 source text at the next tab stop; the compact display uses only as
29742 much space as is needed for the line numbers in the current file, and
29743 only a single space to separate the line numbers from the source.
29746 Note that the colors of the TUI borders can be controlled using the
29747 appropriate @code{set style} commands. @xref{Output Styling}.
29750 @chapter Using @value{GDBN} under @sc{gnu} Emacs
29753 @cindex @sc{gnu} Emacs
29754 A special interface allows you to use @sc{gnu} Emacs to view (and
29755 edit) the source files for the program you are debugging with
29758 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
29759 executable file you want to debug as an argument. This command starts
29760 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
29761 created Emacs buffer.
29762 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
29764 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
29769 All ``terminal'' input and output goes through an Emacs buffer, called
29772 This applies both to @value{GDBN} commands and their output, and to the input
29773 and output done by the program you are debugging.
29775 This is useful because it means that you can copy the text of previous
29776 commands and input them again; you can even use parts of the output
29779 All the facilities of Emacs' Shell mode are available for interacting
29780 with your program. In particular, you can send signals the usual
29781 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
29785 @value{GDBN} displays source code through Emacs.
29787 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
29788 source file for that frame and puts an arrow (@samp{=>}) at the
29789 left margin of the current line. Emacs uses a separate buffer for
29790 source display, and splits the screen to show both your @value{GDBN} session
29793 Explicit @value{GDBN} @code{list} or search commands still produce output as
29794 usual, but you probably have no reason to use them from Emacs.
29797 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
29798 a graphical mode, enabled by default, which provides further buffers
29799 that can control the execution and describe the state of your program.
29800 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
29802 If you specify an absolute file name when prompted for the @kbd{M-x
29803 gdb} argument, then Emacs sets your current working directory to where
29804 your program resides. If you only specify the file name, then Emacs
29805 sets your current working directory to the directory associated
29806 with the previous buffer. In this case, @value{GDBN} may find your
29807 program by searching your environment's @code{PATH} variable, but on
29808 some operating systems it might not find the source. So, although the
29809 @value{GDBN} input and output session proceeds normally, the auxiliary
29810 buffer does not display the current source and line of execution.
29812 The initial working directory of @value{GDBN} is printed on the top
29813 line of the GUD buffer and this serves as a default for the commands
29814 that specify files for @value{GDBN} to operate on. @xref{Files,
29815 ,Commands to Specify Files}.
29817 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
29818 need to call @value{GDBN} by a different name (for example, if you
29819 keep several configurations around, with different names) you can
29820 customize the Emacs variable @code{gud-gdb-command-name} to run the
29823 In the GUD buffer, you can use these special Emacs commands in
29824 addition to the standard Shell mode commands:
29828 Describe the features of Emacs' GUD Mode.
29831 Execute to another source line, like the @value{GDBN} @code{step} command; also
29832 update the display window to show the current file and location.
29835 Execute to next source line in this function, skipping all function
29836 calls, like the @value{GDBN} @code{next} command. Then update the display window
29837 to show the current file and location.
29840 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
29841 display window accordingly.
29844 Execute until exit from the selected stack frame, like the @value{GDBN}
29845 @code{finish} command.
29848 Continue execution of your program, like the @value{GDBN} @code{continue}
29852 Go up the number of frames indicated by the numeric argument
29853 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
29854 like the @value{GDBN} @code{up} command.
29857 Go down the number of frames indicated by the numeric argument, like the
29858 @value{GDBN} @code{down} command.
29861 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
29862 tells @value{GDBN} to set a breakpoint on the source line point is on.
29864 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
29865 separate frame which shows a backtrace when the GUD buffer is current.
29866 Move point to any frame in the stack and type @key{RET} to make it
29867 become the current frame and display the associated source in the
29868 source buffer. Alternatively, click @kbd{Mouse-2} to make the
29869 selected frame become the current one. In graphical mode, the
29870 speedbar displays watch expressions.
29872 If you accidentally delete the source-display buffer, an easy way to get
29873 it back is to type the command @code{f} in the @value{GDBN} buffer, to
29874 request a frame display; when you run under Emacs, this recreates
29875 the source buffer if necessary to show you the context of the current
29878 The source files displayed in Emacs are in ordinary Emacs buffers
29879 which are visiting the source files in the usual way. You can edit
29880 the files with these buffers if you wish; but keep in mind that @value{GDBN}
29881 communicates with Emacs in terms of line numbers. If you add or
29882 delete lines from the text, the line numbers that @value{GDBN} knows cease
29883 to correspond properly with the code.
29885 A more detailed description of Emacs' interaction with @value{GDBN} is
29886 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
29890 @chapter The @sc{gdb/mi} Interface
29892 @unnumberedsec Function and Purpose
29894 @cindex @sc{gdb/mi}, its purpose
29895 @sc{gdb/mi} is a line based machine oriented text interface to
29896 @value{GDBN} and is activated by specifying using the
29897 @option{--interpreter} command line option (@pxref{Mode Options}). It
29898 is specifically intended to support the development of systems which
29899 use the debugger as just one small component of a larger system.
29901 This chapter is a specification of the @sc{gdb/mi} interface. It is written
29902 in the form of a reference manual.
29904 Note that @sc{gdb/mi} is still under construction, so some of the
29905 features described below are incomplete and subject to change
29906 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
29908 @unnumberedsec Notation and Terminology
29910 @cindex notational conventions, for @sc{gdb/mi}
29911 This chapter uses the following notation:
29915 @code{|} separates two alternatives.
29918 @code{[ @var{something} ]} indicates that @var{something} is optional:
29919 it may or may not be given.
29922 @code{( @var{group} )*} means that @var{group} inside the parentheses
29923 may repeat zero or more times.
29926 @code{( @var{group} )+} means that @var{group} inside the parentheses
29927 may repeat one or more times.
29930 @code{"@var{string}"} means a literal @var{string}.
29934 @heading Dependencies
29938 * GDB/MI General Design::
29939 * GDB/MI Command Syntax::
29940 * GDB/MI Compatibility with CLI::
29941 * GDB/MI Development and Front Ends::
29942 * GDB/MI Output Records::
29943 * GDB/MI Simple Examples::
29944 * GDB/MI Command Description Format::
29945 * GDB/MI Breakpoint Commands::
29946 * GDB/MI Catchpoint Commands::
29947 * GDB/MI Program Context::
29948 * GDB/MI Thread Commands::
29949 * GDB/MI Ada Tasking Commands::
29950 * GDB/MI Program Execution::
29951 * GDB/MI Stack Manipulation::
29952 * GDB/MI Variable Objects::
29953 * GDB/MI Data Manipulation::
29954 * GDB/MI Tracepoint Commands::
29955 * GDB/MI Symbol Query::
29956 * GDB/MI File Commands::
29958 * GDB/MI Kod Commands::
29959 * GDB/MI Memory Overlay Commands::
29960 * GDB/MI Signal Handling Commands::
29962 * GDB/MI Target Manipulation::
29963 * GDB/MI File Transfer Commands::
29964 * GDB/MI Ada Exceptions Commands::
29965 * GDB/MI Support Commands::
29966 * GDB/MI Miscellaneous Commands::
29969 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29970 @node GDB/MI General Design
29971 @section @sc{gdb/mi} General Design
29972 @cindex GDB/MI General Design
29974 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
29975 parts---commands sent to @value{GDBN}, responses to those commands
29976 and notifications. Each command results in exactly one response,
29977 indicating either successful completion of the command, or an error.
29978 For the commands that do not resume the target, the response contains the
29979 requested information. For the commands that resume the target, the
29980 response only indicates whether the target was successfully resumed.
29981 Notifications is the mechanism for reporting changes in the state of the
29982 target, or in @value{GDBN} state, that cannot conveniently be associated with
29983 a command and reported as part of that command response.
29985 The important examples of notifications are:
29989 Exec notifications. These are used to report changes in
29990 target state---when a target is resumed, or stopped. It would not
29991 be feasible to include this information in response of resuming
29992 commands, because one resume commands can result in multiple events in
29993 different threads. Also, quite some time may pass before any event
29994 happens in the target, while a frontend needs to know whether the resuming
29995 command itself was successfully executed.
29998 Console output, and status notifications. Console output
29999 notifications are used to report output of CLI commands, as well as
30000 diagnostics for other commands. Status notifications are used to
30001 report the progress of a long-running operation. Naturally, including
30002 this information in command response would mean no output is produced
30003 until the command is finished, which is undesirable.
30006 General notifications. Commands may have various side effects on
30007 the @value{GDBN} or target state beyond their official purpose. For example,
30008 a command may change the selected thread. Although such changes can
30009 be included in command response, using notification allows for more
30010 orthogonal frontend design.
30014 There's no guarantee that whenever an MI command reports an error,
30015 @value{GDBN} or the target are in any specific state, and especially,
30016 the state is not reverted to the state before the MI command was
30017 processed. Therefore, whenever an MI command results in an error,
30018 we recommend that the frontend refreshes all the information shown in
30019 the user interface.
30023 * Context management::
30024 * Asynchronous and non-stop modes::
30028 @node Context management
30029 @subsection Context management
30031 @subsubsection Threads and Frames
30033 In most cases when @value{GDBN} accesses the target, this access is
30034 done in context of a specific thread and frame (@pxref{Frames}).
30035 Often, even when accessing global data, the target requires that a thread
30036 be specified. The CLI interface maintains the selected thread and frame,
30037 and supplies them to target on each command. This is convenient,
30038 because a command line user would not want to specify that information
30039 explicitly on each command, and because user interacts with
30040 @value{GDBN} via a single terminal, so no confusion is possible as
30041 to what thread and frame are the current ones.
30043 In the case of MI, the concept of selected thread and frame is less
30044 useful. First, a frontend can easily remember this information
30045 itself. Second, a graphical frontend can have more than one window,
30046 each one used for debugging a different thread, and the frontend might
30047 want to access additional threads for internal purposes. This
30048 increases the risk that by relying on implicitly selected thread, the
30049 frontend may be operating on a wrong one. Therefore, each MI command
30050 should explicitly specify which thread and frame to operate on. To
30051 make it possible, each MI command accepts the @samp{--thread} and
30052 @samp{--frame} options, the value to each is @value{GDBN} global
30053 identifier for thread and frame to operate on.
30055 Usually, each top-level window in a frontend allows the user to select
30056 a thread and a frame, and remembers the user selection for further
30057 operations. However, in some cases @value{GDBN} may suggest that the
30058 current thread or frame be changed. For example, when stopping on a
30059 breakpoint it is reasonable to switch to the thread where breakpoint is
30060 hit. For another example, if the user issues the CLI @samp{thread} or
30061 @samp{frame} commands via the frontend, it is desirable to change the
30062 frontend's selection to the one specified by user. @value{GDBN}
30063 communicates the suggestion to change current thread and frame using the
30064 @samp{=thread-selected} notification.
30066 Note that historically, MI shares the selected thread with CLI, so
30067 frontends used the @code{-thread-select} to execute commands in the
30068 right context. However, getting this to work right is cumbersome. The
30069 simplest way is for frontend to emit @code{-thread-select} command
30070 before every command. This doubles the number of commands that need
30071 to be sent. The alternative approach is to suppress @code{-thread-select}
30072 if the selected thread in @value{GDBN} is supposed to be identical to the
30073 thread the frontend wants to operate on. However, getting this
30074 optimization right can be tricky. In particular, if the frontend
30075 sends several commands to @value{GDBN}, and one of the commands changes the
30076 selected thread, then the behaviour of subsequent commands will
30077 change. So, a frontend should either wait for response from such
30078 problematic commands, or explicitly add @code{-thread-select} for
30079 all subsequent commands. No frontend is known to do this exactly
30080 right, so it is suggested to just always pass the @samp{--thread} and
30081 @samp{--frame} options.
30083 @subsubsection Language
30085 The execution of several commands depends on which language is selected.
30086 By default, the current language (@pxref{show language}) is used.
30087 But for commands known to be language-sensitive, it is recommended
30088 to use the @samp{--language} option. This option takes one argument,
30089 which is the name of the language to use while executing the command.
30093 -data-evaluate-expression --language c "sizeof (void*)"
30098 The valid language names are the same names accepted by the
30099 @samp{set language} command (@pxref{Manually}), excluding @samp{auto},
30100 @samp{local} or @samp{unknown}.
30102 @node Asynchronous and non-stop modes
30103 @subsection Asynchronous command execution and non-stop mode
30105 On some targets, @value{GDBN} is capable of processing MI commands
30106 even while the target is running. This is called @dfn{asynchronous
30107 command execution} (@pxref{Background Execution}). The frontend may
30108 specify a preference for asynchronous execution using the
30109 @code{-gdb-set mi-async 1} command, which should be emitted before
30110 either running the executable or attaching to the target. After the
30111 frontend has started the executable or attached to the target, it can
30112 find if asynchronous execution is enabled using the
30113 @code{-list-target-features} command.
30116 @item -gdb-set mi-async on
30117 @item -gdb-set mi-async off
30118 Set whether MI is in asynchronous mode.
30120 When @code{off}, which is the default, MI execution commands (e.g.,
30121 @code{-exec-continue}) are foreground commands, and @value{GDBN} waits
30122 for the program to stop before processing further commands.
30124 When @code{on}, MI execution commands are background execution
30125 commands (e.g., @code{-exec-continue} becomes the equivalent of the
30126 @code{c&} CLI command), and so @value{GDBN} is capable of processing
30127 MI commands even while the target is running.
30129 @item -gdb-show mi-async
30130 Show whether MI asynchronous mode is enabled.
30133 Note: In @value{GDBN} version 7.7 and earlier, this option was called
30134 @code{target-async} instead of @code{mi-async}, and it had the effect
30135 of both putting MI in asynchronous mode and making CLI background
30136 commands possible. CLI background commands are now always possible
30137 ``out of the box'' if the target supports them. The old spelling is
30138 kept as a deprecated alias for backwards compatibility.
30140 Even if @value{GDBN} can accept a command while target is running,
30141 many commands that access the target do not work when the target is
30142 running. Therefore, asynchronous command execution is most useful
30143 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
30144 it is possible to examine the state of one thread, while other threads
30147 When a given thread is running, MI commands that try to access the
30148 target in the context of that thread may not work, or may work only on
30149 some targets. In particular, commands that try to operate on thread's
30150 stack will not work, on any target. Commands that read memory, or
30151 modify breakpoints, may work or not work, depending on the target. Note
30152 that even commands that operate on global state, such as @code{print},
30153 @code{set}, and breakpoint commands, still access the target in the
30154 context of a specific thread, so frontend should try to find a
30155 stopped thread and perform the operation on that thread (using the
30156 @samp{--thread} option).
30158 Which commands will work in the context of a running thread is
30159 highly target dependent. However, the two commands
30160 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
30161 to find the state of a thread, will always work.
30163 @node Thread groups
30164 @subsection Thread groups
30165 @value{GDBN} may be used to debug several processes at the same time.
30166 On some platforms, @value{GDBN} may support debugging of several
30167 hardware systems, each one having several cores with several different
30168 processes running on each core. This section describes the MI
30169 mechanism to support such debugging scenarios.
30171 The key observation is that regardless of the structure of the
30172 target, MI can have a global list of threads, because most commands that
30173 accept the @samp{--thread} option do not need to know what process that
30174 thread belongs to. Therefore, it is not necessary to introduce
30175 neither additional @samp{--process} option, nor an notion of the
30176 current process in the MI interface. The only strictly new feature
30177 that is required is the ability to find how the threads are grouped
30180 To allow the user to discover such grouping, and to support arbitrary
30181 hierarchy of machines/cores/processes, MI introduces the concept of a
30182 @dfn{thread group}. Thread group is a collection of threads and other
30183 thread groups. A thread group always has a string identifier, a type,
30184 and may have additional attributes specific to the type. A new
30185 command, @code{-list-thread-groups}, returns the list of top-level
30186 thread groups, which correspond to processes that @value{GDBN} is
30187 debugging at the moment. By passing an identifier of a thread group
30188 to the @code{-list-thread-groups} command, it is possible to obtain
30189 the members of specific thread group.
30191 To allow the user to easily discover processes, and other objects, he
30192 wishes to debug, a concept of @dfn{available thread group} is
30193 introduced. Available thread group is an thread group that
30194 @value{GDBN} is not debugging, but that can be attached to, using the
30195 @code{-target-attach} command. The list of available top-level thread
30196 groups can be obtained using @samp{-list-thread-groups --available}.
30197 In general, the content of a thread group may be only retrieved only
30198 after attaching to that thread group.
30200 Thread groups are related to inferiors (@pxref{Inferiors and
30201 Programs}). Each inferior corresponds to a thread group of a special
30202 type @samp{process}, and some additional operations are permitted on
30203 such thread groups.
30205 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30206 @node GDB/MI Command Syntax
30207 @section @sc{gdb/mi} Command Syntax
30210 * GDB/MI Input Syntax::
30211 * GDB/MI Output Syntax::
30214 @node GDB/MI Input Syntax
30215 @subsection @sc{gdb/mi} Input Syntax
30217 @cindex input syntax for @sc{gdb/mi}
30218 @cindex @sc{gdb/mi}, input syntax
30220 @item @var{command} @expansion{}
30221 @code{@var{cli-command} | @var{mi-command}}
30223 @item @var{cli-command} @expansion{}
30224 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
30225 @var{cli-command} is any existing @value{GDBN} CLI command.
30227 @item @var{mi-command} @expansion{}
30228 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
30229 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
30231 @item @var{token} @expansion{}
30232 "any sequence of digits"
30234 @item @var{option} @expansion{}
30235 @code{"-" @var{parameter} [ " " @var{parameter} ]}
30237 @item @var{parameter} @expansion{}
30238 @code{@var{non-blank-sequence} | @var{c-string}}
30240 @item @var{operation} @expansion{}
30241 @emph{any of the operations described in this chapter}
30243 @item @var{non-blank-sequence} @expansion{}
30244 @emph{anything, provided it doesn't contain special characters such as
30245 "-", @var{nl}, """ and of course " "}
30247 @item @var{c-string} @expansion{}
30248 @code{""" @var{seven-bit-iso-c-string-content} """}
30250 @item @var{nl} @expansion{}
30259 The CLI commands are still handled by the @sc{mi} interpreter; their
30260 output is described below.
30263 The @code{@var{token}}, when present, is passed back when the command
30267 Some @sc{mi} commands accept optional arguments as part of the parameter
30268 list. Each option is identified by a leading @samp{-} (dash) and may be
30269 followed by an optional argument parameter. Options occur first in the
30270 parameter list and can be delimited from normal parameters using
30271 @samp{--} (this is useful when some parameters begin with a dash).
30278 We want easy access to the existing CLI syntax (for debugging).
30281 We want it to be easy to spot a @sc{mi} operation.
30284 @node GDB/MI Output Syntax
30285 @subsection @sc{gdb/mi} Output Syntax
30287 @cindex output syntax of @sc{gdb/mi}
30288 @cindex @sc{gdb/mi}, output syntax
30289 The output from @sc{gdb/mi} consists of zero or more out-of-band records
30290 followed, optionally, by a single result record. This result record
30291 is for the most recent command. The sequence of output records is
30292 terminated by @samp{(gdb)}.
30294 If an input command was prefixed with a @code{@var{token}} then the
30295 corresponding output for that command will also be prefixed by that same
30299 @item @var{output} @expansion{}
30300 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
30302 @item @var{result-record} @expansion{}
30303 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
30305 @item @var{out-of-band-record} @expansion{}
30306 @code{@var{async-record} | @var{stream-record}}
30308 @item @var{async-record} @expansion{}
30309 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
30311 @item @var{exec-async-output} @expansion{}
30312 @code{[ @var{token} ] "*" @var{async-output nl}}
30314 @item @var{status-async-output} @expansion{}
30315 @code{[ @var{token} ] "+" @var{async-output nl}}
30317 @item @var{notify-async-output} @expansion{}
30318 @code{[ @var{token} ] "=" @var{async-output nl}}
30320 @item @var{async-output} @expansion{}
30321 @code{@var{async-class} ( "," @var{result} )*}
30323 @item @var{result-class} @expansion{}
30324 @code{"done" | "running" | "connected" | "error" | "exit"}
30326 @item @var{async-class} @expansion{}
30327 @code{"stopped" | @var{others}} (where @var{others} will be added
30328 depending on the needs---this is still in development).
30330 @item @var{result} @expansion{}
30331 @code{ @var{variable} "=" @var{value}}
30333 @item @var{variable} @expansion{}
30334 @code{ @var{string} }
30336 @item @var{value} @expansion{}
30337 @code{ @var{const} | @var{tuple} | @var{list} }
30339 @item @var{const} @expansion{}
30340 @code{@var{c-string}}
30342 @item @var{tuple} @expansion{}
30343 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
30345 @item @var{list} @expansion{}
30346 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
30347 @var{result} ( "," @var{result} )* "]" }
30349 @item @var{stream-record} @expansion{}
30350 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
30352 @item @var{console-stream-output} @expansion{}
30353 @code{"~" @var{c-string nl}}
30355 @item @var{target-stream-output} @expansion{}
30356 @code{"@@" @var{c-string nl}}
30358 @item @var{log-stream-output} @expansion{}
30359 @code{"&" @var{c-string nl}}
30361 @item @var{nl} @expansion{}
30364 @item @var{token} @expansion{}
30365 @emph{any sequence of digits}.
30373 All output sequences end in a single line containing a period.
30376 The @code{@var{token}} is from the corresponding request. Note that
30377 for all async output, while the token is allowed by the grammar and
30378 may be output by future versions of @value{GDBN} for select async
30379 output messages, it is generally omitted. Frontends should treat
30380 all async output as reporting general changes in the state of the
30381 target and there should be no need to associate async output to any
30385 @cindex status output in @sc{gdb/mi}
30386 @var{status-async-output} contains on-going status information about the
30387 progress of a slow operation. It can be discarded. All status output is
30388 prefixed by @samp{+}.
30391 @cindex async output in @sc{gdb/mi}
30392 @var{exec-async-output} contains asynchronous state change on the target
30393 (stopped, started, disappeared). All async output is prefixed by
30397 @cindex notify output in @sc{gdb/mi}
30398 @var{notify-async-output} contains supplementary information that the
30399 client should handle (e.g., a new breakpoint information). All notify
30400 output is prefixed by @samp{=}.
30403 @cindex console output in @sc{gdb/mi}
30404 @var{console-stream-output} is output that should be displayed as is in the
30405 console. It is the textual response to a CLI command. All the console
30406 output is prefixed by @samp{~}.
30409 @cindex target output in @sc{gdb/mi}
30410 @var{target-stream-output} is the output produced by the target program.
30411 All the target output is prefixed by @samp{@@}.
30414 @cindex log output in @sc{gdb/mi}
30415 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
30416 instance messages that should be displayed as part of an error log. All
30417 the log output is prefixed by @samp{&}.
30420 @cindex list output in @sc{gdb/mi}
30421 New @sc{gdb/mi} commands should only output @var{lists} containing
30427 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
30428 details about the various output records.
30430 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30431 @node GDB/MI Compatibility with CLI
30432 @section @sc{gdb/mi} Compatibility with CLI
30434 @cindex compatibility, @sc{gdb/mi} and CLI
30435 @cindex @sc{gdb/mi}, compatibility with CLI
30437 For the developers convenience CLI commands can be entered directly,
30438 but there may be some unexpected behaviour. For example, commands
30439 that query the user will behave as if the user replied yes, breakpoint
30440 command lists are not executed and some CLI commands, such as
30441 @code{if}, @code{when} and @code{define}, prompt for further input with
30442 @samp{>}, which is not valid MI output.
30444 This feature may be removed at some stage in the future and it is
30445 recommended that front ends use the @code{-interpreter-exec} command
30446 (@pxref{-interpreter-exec}).
30448 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30449 @node GDB/MI Development and Front Ends
30450 @section @sc{gdb/mi} Development and Front Ends
30451 @cindex @sc{gdb/mi} development
30453 The application which takes the MI output and presents the state of the
30454 program being debugged to the user is called a @dfn{front end}.
30456 Since @sc{gdb/mi} is used by a variety of front ends to @value{GDBN}, changes
30457 to the MI interface may break existing usage. This section describes how the
30458 protocol changes and how to request previous version of the protocol when it
30461 Some changes in MI need not break a carefully designed front end, and
30462 for these the MI version will remain unchanged. The following is a
30463 list of changes that may occur within one level, so front ends should
30464 parse MI output in a way that can handle them:
30468 New MI commands may be added.
30471 New fields may be added to the output of any MI command.
30474 The range of values for fields with specified values, e.g.,
30475 @code{in_scope} (@pxref{-var-update}) may be extended.
30477 @c The format of field's content e.g type prefix, may change so parse it
30478 @c at your own risk. Yes, in general?
30480 @c The order of fields may change? Shouldn't really matter but it might
30481 @c resolve inconsistencies.
30484 If the changes are likely to break front ends, the MI version level
30485 will be increased by one. The new versions of the MI protocol are not compatible
30486 with the old versions. Old versions of MI remain available, allowing front ends
30487 to keep using them until they are modified to use the latest MI version.
30489 Since @code{--interpreter=mi} always points to the latest MI version, it is
30490 recommended that front ends request a specific version of MI when launching
30491 @value{GDBN} (e.g. @code{--interpreter=mi2}) to make sure they get an
30492 interpreter with the MI version they expect.
30494 The following table gives a summary of the the released versions of the MI
30495 interface: the version number, the version of GDB in which it first appeared
30496 and the breaking changes compared to the previous version.
30498 @multitable @columnfractions .05 .05 .9
30499 @headitem MI version @tab GDB version @tab Breaking changes
30516 The @code{-environment-pwd}, @code{-environment-directory} and
30517 @code{-environment-path} commands now returns values using the MI output
30518 syntax, rather than CLI output syntax.
30521 @code{-var-list-children}'s @code{children} result field is now a list, rather
30525 @code{-var-update}'s @code{changelist} result field is now a list, rather than
30537 The output of information about multi-location breakpoints has changed in the
30538 responses to the @code{-break-insert} and @code{-break-info} commands, as well
30539 as in the @code{=breakpoint-created} and @code{=breakpoint-modified} events.
30540 The multiple locations are now placed in a @code{locations} field, whose value
30546 If your front end cannot yet migrate to a more recent version of the
30547 MI protocol, you can nevertheless selectively enable specific features
30548 available in those recent MI versions, using the following commands:
30552 @item -fix-multi-location-breakpoint-output
30553 Use the output for multi-location breakpoints which was introduced by
30554 MI 3, even when using MI versions 2 or 1. This command has no
30555 effect when using MI version 3 or later.
30559 The best way to avoid unexpected changes in MI that might break your front
30560 end is to make your project known to @value{GDBN} developers and
30561 follow development on @email{gdb@@sourceware.org} and
30562 @email{gdb-patches@@sourceware.org}.
30563 @cindex mailing lists
30565 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30566 @node GDB/MI Output Records
30567 @section @sc{gdb/mi} Output Records
30570 * GDB/MI Result Records::
30571 * GDB/MI Stream Records::
30572 * GDB/MI Async Records::
30573 * GDB/MI Breakpoint Information::
30574 * GDB/MI Frame Information::
30575 * GDB/MI Thread Information::
30576 * GDB/MI Ada Exception Information::
30579 @node GDB/MI Result Records
30580 @subsection @sc{gdb/mi} Result Records
30582 @cindex result records in @sc{gdb/mi}
30583 @cindex @sc{gdb/mi}, result records
30584 In addition to a number of out-of-band notifications, the response to a
30585 @sc{gdb/mi} command includes one of the following result indications:
30589 @item "^done" [ "," @var{results} ]
30590 The synchronous operation was successful, @code{@var{results}} are the return
30595 This result record is equivalent to @samp{^done}. Historically, it
30596 was output instead of @samp{^done} if the command has resumed the
30597 target. This behaviour is maintained for backward compatibility, but
30598 all frontends should treat @samp{^done} and @samp{^running}
30599 identically and rely on the @samp{*running} output record to determine
30600 which threads are resumed.
30604 @value{GDBN} has connected to a remote target.
30606 @item "^error" "," "msg=" @var{c-string} [ "," "code=" @var{c-string} ]
30608 The operation failed. The @code{msg=@var{c-string}} variable contains
30609 the corresponding error message.
30611 If present, the @code{code=@var{c-string}} variable provides an error
30612 code on which consumers can rely on to detect the corresponding
30613 error condition. At present, only one error code is defined:
30616 @item "undefined-command"
30617 Indicates that the command causing the error does not exist.
30622 @value{GDBN} has terminated.
30626 @node GDB/MI Stream Records
30627 @subsection @sc{gdb/mi} Stream Records
30629 @cindex @sc{gdb/mi}, stream records
30630 @cindex stream records in @sc{gdb/mi}
30631 @value{GDBN} internally maintains a number of output streams: the console, the
30632 target, and the log. The output intended for each of these streams is
30633 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
30635 Each stream record begins with a unique @dfn{prefix character} which
30636 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
30637 Syntax}). In addition to the prefix, each stream record contains a
30638 @code{@var{string-output}}. This is either raw text (with an implicit new
30639 line) or a quoted C string (which does not contain an implicit newline).
30642 @item "~" @var{string-output}
30643 The console output stream contains text that should be displayed in the
30644 CLI console window. It contains the textual responses to CLI commands.
30646 @item "@@" @var{string-output}
30647 The target output stream contains any textual output from the running
30648 target. This is only present when GDB's event loop is truly
30649 asynchronous, which is currently only the case for remote targets.
30651 @item "&" @var{string-output}
30652 The log stream contains debugging messages being produced by @value{GDBN}'s
30656 @node GDB/MI Async Records
30657 @subsection @sc{gdb/mi} Async Records
30659 @cindex async records in @sc{gdb/mi}
30660 @cindex @sc{gdb/mi}, async records
30661 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
30662 additional changes that have occurred. Those changes can either be a
30663 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
30664 target activity (e.g., target stopped).
30666 The following is the list of possible async records:
30670 @item *running,thread-id="@var{thread}"
30671 The target is now running. The @var{thread} field can be the global
30672 thread ID of the the thread that is now running, and it can be
30673 @samp{all} if all threads are running. The frontend should assume
30674 that no interaction with a running thread is possible after this
30675 notification is produced. The frontend should not assume that this
30676 notification is output only once for any command. @value{GDBN} may
30677 emit this notification several times, either for different threads,
30678 because it cannot resume all threads together, or even for a single
30679 thread, if the thread must be stepped though some code before letting
30682 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
30683 The target has stopped. The @var{reason} field can have one of the
30687 @item breakpoint-hit
30688 A breakpoint was reached.
30689 @item watchpoint-trigger
30690 A watchpoint was triggered.
30691 @item read-watchpoint-trigger
30692 A read watchpoint was triggered.
30693 @item access-watchpoint-trigger
30694 An access watchpoint was triggered.
30695 @item function-finished
30696 An -exec-finish or similar CLI command was accomplished.
30697 @item location-reached
30698 An -exec-until or similar CLI command was accomplished.
30699 @item watchpoint-scope
30700 A watchpoint has gone out of scope.
30701 @item end-stepping-range
30702 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
30703 similar CLI command was accomplished.
30704 @item exited-signalled
30705 The inferior exited because of a signal.
30707 The inferior exited.
30708 @item exited-normally
30709 The inferior exited normally.
30710 @item signal-received
30711 A signal was received by the inferior.
30713 The inferior has stopped due to a library being loaded or unloaded.
30714 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
30715 set or when a @code{catch load} or @code{catch unload} catchpoint is
30716 in use (@pxref{Set Catchpoints}).
30718 The inferior has forked. This is reported when @code{catch fork}
30719 (@pxref{Set Catchpoints}) has been used.
30721 The inferior has vforked. This is reported in when @code{catch vfork}
30722 (@pxref{Set Catchpoints}) has been used.
30723 @item syscall-entry
30724 The inferior entered a system call. This is reported when @code{catch
30725 syscall} (@pxref{Set Catchpoints}) has been used.
30726 @item syscall-return
30727 The inferior returned from a system call. This is reported when
30728 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
30730 The inferior called @code{exec}. This is reported when @code{catch exec}
30731 (@pxref{Set Catchpoints}) has been used.
30734 The @var{id} field identifies the global thread ID of the thread
30735 that directly caused the stop -- for example by hitting a breakpoint.
30736 Depending on whether all-stop
30737 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
30738 stop all threads, or only the thread that directly triggered the stop.
30739 If all threads are stopped, the @var{stopped} field will have the
30740 value of @code{"all"}. Otherwise, the value of the @var{stopped}
30741 field will be a list of thread identifiers. Presently, this list will
30742 always include a single thread, but frontend should be prepared to see
30743 several threads in the list. The @var{core} field reports the
30744 processor core on which the stop event has happened. This field may be absent
30745 if such information is not available.
30747 @item =thread-group-added,id="@var{id}"
30748 @itemx =thread-group-removed,id="@var{id}"
30749 A thread group was either added or removed. The @var{id} field
30750 contains the @value{GDBN} identifier of the thread group. When a thread
30751 group is added, it generally might not be associated with a running
30752 process. When a thread group is removed, its id becomes invalid and
30753 cannot be used in any way.
30755 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
30756 A thread group became associated with a running program,
30757 either because the program was just started or the thread group
30758 was attached to a program. The @var{id} field contains the
30759 @value{GDBN} identifier of the thread group. The @var{pid} field
30760 contains process identifier, specific to the operating system.
30762 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
30763 A thread group is no longer associated with a running program,
30764 either because the program has exited, or because it was detached
30765 from. The @var{id} field contains the @value{GDBN} identifier of the
30766 thread group. The @var{code} field is the exit code of the inferior; it exists
30767 only when the inferior exited with some code.
30769 @item =thread-created,id="@var{id}",group-id="@var{gid}"
30770 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
30771 A thread either was created, or has exited. The @var{id} field
30772 contains the global @value{GDBN} identifier of the thread. The @var{gid}
30773 field identifies the thread group this thread belongs to.
30775 @item =thread-selected,id="@var{id}"[,frame="@var{frame}"]
30776 Informs that the selected thread or frame were changed. This notification
30777 is not emitted as result of the @code{-thread-select} or
30778 @code{-stack-select-frame} commands, but is emitted whenever an MI command
30779 that is not documented to change the selected thread and frame actually
30780 changes them. In particular, invoking, directly or indirectly
30781 (via user-defined command), the CLI @code{thread} or @code{frame} commands,
30782 will generate this notification. Changing the thread or frame from another
30783 user interface (see @ref{Interpreters}) will also generate this notification.
30785 The @var{frame} field is only present if the newly selected thread is
30786 stopped. See @ref{GDB/MI Frame Information} for the format of its value.
30788 We suggest that in response to this notification, front ends
30789 highlight the selected thread and cause subsequent commands to apply to
30792 @item =library-loaded,...
30793 Reports that a new library file was loaded by the program. This
30794 notification has 5 fields---@var{id}, @var{target-name},
30795 @var{host-name}, @var{symbols-loaded} and @var{ranges}. The @var{id} field is an
30796 opaque identifier of the library. For remote debugging case,
30797 @var{target-name} and @var{host-name} fields give the name of the
30798 library file on the target, and on the host respectively. For native
30799 debugging, both those fields have the same value. The
30800 @var{symbols-loaded} field is emitted only for backward compatibility
30801 and should not be relied on to convey any useful information. The
30802 @var{thread-group} field, if present, specifies the id of the thread
30803 group in whose context the library was loaded. If the field is
30804 absent, it means the library was loaded in the context of all present
30805 thread groups. The @var{ranges} field specifies the ranges of addresses belonging
30808 @item =library-unloaded,...
30809 Reports that a library was unloaded by the program. This notification
30810 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
30811 the same meaning as for the @code{=library-loaded} notification.
30812 The @var{thread-group} field, if present, specifies the id of the
30813 thread group in whose context the library was unloaded. If the field is
30814 absent, it means the library was unloaded in the context of all present
30817 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
30818 @itemx =traceframe-changed,end
30819 Reports that the trace frame was changed and its new number is
30820 @var{tfnum}. The number of the tracepoint associated with this trace
30821 frame is @var{tpnum}.
30823 @item =tsv-created,name=@var{name},initial=@var{initial}
30824 Reports that the new trace state variable @var{name} is created with
30825 initial value @var{initial}.
30827 @item =tsv-deleted,name=@var{name}
30828 @itemx =tsv-deleted
30829 Reports that the trace state variable @var{name} is deleted or all
30830 trace state variables are deleted.
30832 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
30833 Reports that the trace state variable @var{name} is modified with
30834 the initial value @var{initial}. The current value @var{current} of
30835 trace state variable is optional and is reported if the current
30836 value of trace state variable is known.
30838 @item =breakpoint-created,bkpt=@{...@}
30839 @itemx =breakpoint-modified,bkpt=@{...@}
30840 @itemx =breakpoint-deleted,id=@var{number}
30841 Reports that a breakpoint was created, modified, or deleted,
30842 respectively. Only user-visible breakpoints are reported to the MI
30845 The @var{bkpt} argument is of the same form as returned by the various
30846 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
30847 @var{number} is the ordinal number of the breakpoint.
30849 Note that if a breakpoint is emitted in the result record of a
30850 command, then it will not also be emitted in an async record.
30852 @item =record-started,thread-group="@var{id}",method="@var{method}"[,format="@var{format}"]
30853 @itemx =record-stopped,thread-group="@var{id}"
30854 Execution log recording was either started or stopped on an
30855 inferior. The @var{id} is the @value{GDBN} identifier of the thread
30856 group corresponding to the affected inferior.
30858 The @var{method} field indicates the method used to record execution. If the
30859 method in use supports multiple recording formats, @var{format} will be present
30860 and contain the currently used format. @xref{Process Record and Replay},
30861 for existing method and format values.
30863 @item =cmd-param-changed,param=@var{param},value=@var{value}
30864 Reports that a parameter of the command @code{set @var{param}} is
30865 changed to @var{value}. In the multi-word @code{set} command,
30866 the @var{param} is the whole parameter list to @code{set} command.
30867 For example, In command @code{set check type on}, @var{param}
30868 is @code{check type} and @var{value} is @code{on}.
30870 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
30871 Reports that bytes from @var{addr} to @var{data} + @var{len} were
30872 written in an inferior. The @var{id} is the identifier of the
30873 thread group corresponding to the affected inferior. The optional
30874 @code{type="code"} part is reported if the memory written to holds
30878 @node GDB/MI Breakpoint Information
30879 @subsection @sc{gdb/mi} Breakpoint Information
30881 When @value{GDBN} reports information about a breakpoint, a
30882 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
30887 The breakpoint number.
30890 The type of the breakpoint. For ordinary breakpoints this will be
30891 @samp{breakpoint}, but many values are possible.
30894 If the type of the breakpoint is @samp{catchpoint}, then this
30895 indicates the exact type of catchpoint.
30898 This is the breakpoint disposition---either @samp{del}, meaning that
30899 the breakpoint will be deleted at the next stop, or @samp{keep},
30900 meaning that the breakpoint will not be deleted.
30903 This indicates whether the breakpoint is enabled, in which case the
30904 value is @samp{y}, or disabled, in which case the value is @samp{n}.
30905 Note that this is not the same as the field @code{enable}.
30908 The address of the breakpoint. This may be a hexidecimal number,
30909 giving the address; or the string @samp{<PENDING>}, for a pending
30910 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
30911 multiple locations. This field will not be present if no address can
30912 be determined. For example, a watchpoint does not have an address.
30915 Optional field containing any flags related to the address. These flags are
30916 architecture-dependent; see @ref{Architectures} for their meaning for a
30920 If known, the function in which the breakpoint appears.
30921 If not known, this field is not present.
30924 The name of the source file which contains this function, if known.
30925 If not known, this field is not present.
30928 The full file name of the source file which contains this function, if
30929 known. If not known, this field is not present.
30932 The line number at which this breakpoint appears, if known.
30933 If not known, this field is not present.
30936 If the source file is not known, this field may be provided. If
30937 provided, this holds the address of the breakpoint, possibly followed
30941 If this breakpoint is pending, this field is present and holds the
30942 text used to set the breakpoint, as entered by the user.
30945 Where this breakpoint's condition is evaluated, either @samp{host} or
30949 If this is a thread-specific breakpoint, then this identifies the
30950 thread in which the breakpoint can trigger.
30953 If this breakpoint is restricted to a particular Ada task, then this
30954 field will hold the task identifier.
30957 If the breakpoint is conditional, this is the condition expression.
30960 The ignore count of the breakpoint.
30963 The enable count of the breakpoint.
30965 @item traceframe-usage
30968 @item static-tracepoint-marker-string-id
30969 For a static tracepoint, the name of the static tracepoint marker.
30972 For a masked watchpoint, this is the mask.
30975 A tracepoint's pass count.
30977 @item original-location
30978 The location of the breakpoint as originally specified by the user.
30979 This field is optional.
30982 The number of times the breakpoint has been hit.
30985 This field is only given for tracepoints. This is either @samp{y},
30986 meaning that the tracepoint is installed, or @samp{n}, meaning that it
30990 Some extra data, the exact contents of which are type-dependent.
30993 This field is present if the breakpoint has multiple locations. It is also
30994 exceptionally present if the breakpoint is enabled and has a single, disabled
30997 The value is a list of locations. The format of a location is described below.
31001 A location in a multi-location breakpoint is represented as a tuple with the
31007 The location number as a dotted pair, like @samp{1.2}. The first digit is the
31008 number of the parent breakpoint. The second digit is the number of the
31009 location within that breakpoint.
31012 This indicates whether the location is enabled, in which case the
31013 value is @samp{y}, or disabled, in which case the value is @samp{n}.
31014 Note that this is not the same as the field @code{enable}.
31017 The address of this location as an hexidecimal number.
31020 Optional field containing any flags related to the address. These flags are
31021 architecture-dependent; see @ref{Architectures} for their meaning for a
31025 If known, the function in which the location appears.
31026 If not known, this field is not present.
31029 The name of the source file which contains this location, if known.
31030 If not known, this field is not present.
31033 The full file name of the source file which contains this location, if
31034 known. If not known, this field is not present.
31037 The line number at which this location appears, if known.
31038 If not known, this field is not present.
31040 @item thread-groups
31041 The thread groups this location is in.
31045 For example, here is what the output of @code{-break-insert}
31046 (@pxref{GDB/MI Breakpoint Commands}) might be:
31049 -> -break-insert main
31050 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
31051 enabled="y",addr="0x08048564",func="main",file="myprog.c",
31052 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
31057 @node GDB/MI Frame Information
31058 @subsection @sc{gdb/mi} Frame Information
31060 Response from many MI commands includes an information about stack
31061 frame. This information is a tuple that may have the following
31066 The level of the stack frame. The innermost frame has the level of
31067 zero. This field is always present.
31070 The name of the function corresponding to the frame. This field may
31071 be absent if @value{GDBN} is unable to determine the function name.
31074 The code address for the frame. This field is always present.
31077 Optional field containing any flags related to the address. These flags are
31078 architecture-dependent; see @ref{Architectures} for their meaning for a
31082 The name of the source files that correspond to the frame's code
31083 address. This field may be absent.
31086 The source line corresponding to the frames' code address. This field
31090 The name of the binary file (either executable or shared library) the
31091 corresponds to the frame's code address. This field may be absent.
31095 @node GDB/MI Thread Information
31096 @subsection @sc{gdb/mi} Thread Information
31098 Whenever @value{GDBN} has to report an information about a thread, it
31099 uses a tuple with the following fields. The fields are always present unless
31104 The global numeric id assigned to the thread by @value{GDBN}.
31107 The target-specific string identifying the thread.
31110 Additional information about the thread provided by the target.
31111 It is supposed to be human-readable and not interpreted by the
31112 frontend. This field is optional.
31115 The name of the thread. If the user specified a name using the
31116 @code{thread name} command, then this name is given. Otherwise, if
31117 @value{GDBN} can extract the thread name from the target, then that
31118 name is given. If @value{GDBN} cannot find the thread name, then this
31122 The execution state of the thread, either @samp{stopped} or @samp{running},
31123 depending on whether the thread is presently running.
31126 The stack frame currently executing in the thread. This field is only present
31127 if the thread is stopped. Its format is documented in
31128 @ref{GDB/MI Frame Information}.
31131 The value of this field is an integer number of the processor core the
31132 thread was last seen on. This field is optional.
31135 @node GDB/MI Ada Exception Information
31136 @subsection @sc{gdb/mi} Ada Exception Information
31138 Whenever a @code{*stopped} record is emitted because the program
31139 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
31140 @value{GDBN} provides the name of the exception that was raised via
31141 the @code{exception-name} field. Also, for exceptions that were raised
31142 with an exception message, @value{GDBN} provides that message via
31143 the @code{exception-message} field.
31145 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31146 @node GDB/MI Simple Examples
31147 @section Simple Examples of @sc{gdb/mi} Interaction
31148 @cindex @sc{gdb/mi}, simple examples
31150 This subsection presents several simple examples of interaction using
31151 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
31152 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
31153 the output received from @sc{gdb/mi}.
31155 Note the line breaks shown in the examples are here only for
31156 readability, they don't appear in the real output.
31158 @subheading Setting a Breakpoint
31160 Setting a breakpoint generates synchronous output which contains detailed
31161 information of the breakpoint.
31164 -> -break-insert main
31165 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
31166 enabled="y",addr="0x08048564",func="main",file="myprog.c",
31167 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
31172 @subheading Program Execution
31174 Program execution generates asynchronous records and MI gives the
31175 reason that execution stopped.
31181 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
31182 frame=@{addr="0x08048564",func="main",
31183 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
31184 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68",
31185 arch="i386:x86_64"@}
31190 <- *stopped,reason="exited-normally"
31194 @subheading Quitting @value{GDBN}
31196 Quitting @value{GDBN} just prints the result class @samp{^exit}.
31204 Please note that @samp{^exit} is printed immediately, but it might
31205 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
31206 performs necessary cleanups, including killing programs being debugged
31207 or disconnecting from debug hardware, so the frontend should wait till
31208 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
31209 fails to exit in reasonable time.
31211 @subheading A Bad Command
31213 Here's what happens if you pass a non-existent command:
31217 <- ^error,msg="Undefined MI command: rubbish"
31222 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31223 @node GDB/MI Command Description Format
31224 @section @sc{gdb/mi} Command Description Format
31226 The remaining sections describe blocks of commands. Each block of
31227 commands is laid out in a fashion similar to this section.
31229 @subheading Motivation
31231 The motivation for this collection of commands.
31233 @subheading Introduction
31235 A brief introduction to this collection of commands as a whole.
31237 @subheading Commands
31239 For each command in the block, the following is described:
31241 @subsubheading Synopsis
31244 -command @var{args}@dots{}
31247 @subsubheading Result
31249 @subsubheading @value{GDBN} Command
31251 The corresponding @value{GDBN} CLI command(s), if any.
31253 @subsubheading Example
31255 Example(s) formatted for readability. Some of the described commands have
31256 not been implemented yet and these are labeled N.A.@: (not available).
31259 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31260 @node GDB/MI Breakpoint Commands
31261 @section @sc{gdb/mi} Breakpoint Commands
31263 @cindex breakpoint commands for @sc{gdb/mi}
31264 @cindex @sc{gdb/mi}, breakpoint commands
31265 This section documents @sc{gdb/mi} commands for manipulating
31268 @subheading The @code{-break-after} Command
31269 @findex -break-after
31271 @subsubheading Synopsis
31274 -break-after @var{number} @var{count}
31277 The breakpoint number @var{number} is not in effect until it has been
31278 hit @var{count} times. To see how this is reflected in the output of
31279 the @samp{-break-list} command, see the description of the
31280 @samp{-break-list} command below.
31282 @subsubheading @value{GDBN} Command
31284 The corresponding @value{GDBN} command is @samp{ignore}.
31286 @subsubheading Example
31291 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
31292 enabled="y",addr="0x000100d0",func="main",file="hello.c",
31293 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
31301 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
31302 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
31303 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
31304 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
31305 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
31306 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
31307 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
31308 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
31309 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
31310 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
31315 @subheading The @code{-break-catch} Command
31316 @findex -break-catch
31319 @subheading The @code{-break-commands} Command
31320 @findex -break-commands
31322 @subsubheading Synopsis
31325 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
31328 Specifies the CLI commands that should be executed when breakpoint
31329 @var{number} is hit. The parameters @var{command1} to @var{commandN}
31330 are the commands. If no command is specified, any previously-set
31331 commands are cleared. @xref{Break Commands}. Typical use of this
31332 functionality is tracing a program, that is, printing of values of
31333 some variables whenever breakpoint is hit and then continuing.
31335 @subsubheading @value{GDBN} Command
31337 The corresponding @value{GDBN} command is @samp{commands}.
31339 @subsubheading Example
31344 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
31345 enabled="y",addr="0x000100d0",func="main",file="hello.c",
31346 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
31349 -break-commands 1 "print v" "continue"
31354 @subheading The @code{-break-condition} Command
31355 @findex -break-condition
31357 @subsubheading Synopsis
31360 -break-condition @var{number} @var{expr}
31363 Breakpoint @var{number} will stop the program only if the condition in
31364 @var{expr} is true. The condition becomes part of the
31365 @samp{-break-list} output (see the description of the @samp{-break-list}
31368 @subsubheading @value{GDBN} Command
31370 The corresponding @value{GDBN} command is @samp{condition}.
31372 @subsubheading Example
31376 -break-condition 1 1
31380 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
31381 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
31382 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
31383 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
31384 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
31385 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
31386 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
31387 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
31388 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
31389 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
31393 @subheading The @code{-break-delete} Command
31394 @findex -break-delete
31396 @subsubheading Synopsis
31399 -break-delete ( @var{breakpoint} )+
31402 Delete the breakpoint(s) whose number(s) are specified in the argument
31403 list. This is obviously reflected in the breakpoint list.
31405 @subsubheading @value{GDBN} Command
31407 The corresponding @value{GDBN} command is @samp{delete}.
31409 @subsubheading Example
31417 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
31418 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
31419 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
31420 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
31421 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
31422 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
31423 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
31428 @subheading The @code{-break-disable} Command
31429 @findex -break-disable
31431 @subsubheading Synopsis
31434 -break-disable ( @var{breakpoint} )+
31437 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
31438 break list is now set to @samp{n} for the named @var{breakpoint}(s).
31440 @subsubheading @value{GDBN} Command
31442 The corresponding @value{GDBN} command is @samp{disable}.
31444 @subsubheading Example
31452 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
31453 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
31454 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
31455 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
31456 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
31457 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
31458 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
31459 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
31460 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
31461 line="5",thread-groups=["i1"],times="0"@}]@}
31465 @subheading The @code{-break-enable} Command
31466 @findex -break-enable
31468 @subsubheading Synopsis
31471 -break-enable ( @var{breakpoint} )+
31474 Enable (previously disabled) @var{breakpoint}(s).
31476 @subsubheading @value{GDBN} Command
31478 The corresponding @value{GDBN} command is @samp{enable}.
31480 @subsubheading Example
31488 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
31489 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
31490 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
31491 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
31492 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
31493 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
31494 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
31495 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
31496 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
31497 line="5",thread-groups=["i1"],times="0"@}]@}
31501 @subheading The @code{-break-info} Command
31502 @findex -break-info
31504 @subsubheading Synopsis
31507 -break-info @var{breakpoint}
31511 Get information about a single breakpoint.
31513 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
31514 Information}, for details on the format of each breakpoint in the
31517 @subsubheading @value{GDBN} Command
31519 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
31521 @subsubheading Example
31524 @subheading The @code{-break-insert} Command
31525 @findex -break-insert
31526 @anchor{-break-insert}
31528 @subsubheading Synopsis
31531 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
31532 [ -c @var{condition} ] [ -i @var{ignore-count} ]
31533 [ -p @var{thread-id} ] [ @var{location} ]
31537 If specified, @var{location}, can be one of:
31540 @item linespec location
31541 A linespec location. @xref{Linespec Locations}.
31543 @item explicit location
31544 An explicit location. @sc{gdb/mi} explicit locations are
31545 analogous to the CLI's explicit locations using the option names
31546 listed below. @xref{Explicit Locations}.
31549 @item --source @var{filename}
31550 The source file name of the location. This option requires the use
31551 of either @samp{--function} or @samp{--line}.
31553 @item --function @var{function}
31554 The name of a function or method.
31556 @item --label @var{label}
31557 The name of a label.
31559 @item --line @var{lineoffset}
31560 An absolute or relative line offset from the start of the location.
31563 @item address location
31564 An address location, *@var{address}. @xref{Address Locations}.
31568 The possible optional parameters of this command are:
31572 Insert a temporary breakpoint.
31574 Insert a hardware breakpoint.
31576 If @var{location} cannot be parsed (for example if it
31577 refers to unknown files or functions), create a pending
31578 breakpoint. Without this flag, @value{GDBN} will report
31579 an error, and won't create a breakpoint, if @var{location}
31582 Create a disabled breakpoint.
31584 Create a tracepoint. @xref{Tracepoints}. When this parameter
31585 is used together with @samp{-h}, a fast tracepoint is created.
31586 @item -c @var{condition}
31587 Make the breakpoint conditional on @var{condition}.
31588 @item -i @var{ignore-count}
31589 Initialize the @var{ignore-count}.
31590 @item -p @var{thread-id}
31591 Restrict the breakpoint to the thread with the specified global
31595 @subsubheading Result
31597 @xref{GDB/MI Breakpoint Information}, for details on the format of the
31598 resulting breakpoint.
31600 Note: this format is open to change.
31601 @c An out-of-band breakpoint instead of part of the result?
31603 @subsubheading @value{GDBN} Command
31605 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
31606 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
31608 @subsubheading Example
31613 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
31614 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
31617 -break-insert -t foo
31618 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
31619 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
31623 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
31624 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
31625 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
31626 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
31627 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
31628 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
31629 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
31630 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
31631 addr="0x0001072c", func="main",file="recursive2.c",
31632 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
31634 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
31635 addr="0x00010774",func="foo",file="recursive2.c",
31636 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
31639 @c -break-insert -r foo.*
31640 @c ~int foo(int, int);
31641 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
31642 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
31647 @subheading The @code{-dprintf-insert} Command
31648 @findex -dprintf-insert
31650 @subsubheading Synopsis
31653 -dprintf-insert [ -t ] [ -f ] [ -d ]
31654 [ -c @var{condition} ] [ -i @var{ignore-count} ]
31655 [ -p @var{thread-id} ] [ @var{location} ] [ @var{format} ]
31660 If supplied, @var{location} may be specified the same way as for
31661 the @code{-break-insert} command. @xref{-break-insert}.
31663 The possible optional parameters of this command are:
31667 Insert a temporary breakpoint.
31669 If @var{location} cannot be parsed (for example, if it
31670 refers to unknown files or functions), create a pending
31671 breakpoint. Without this flag, @value{GDBN} will report
31672 an error, and won't create a breakpoint, if @var{location}
31675 Create a disabled breakpoint.
31676 @item -c @var{condition}
31677 Make the breakpoint conditional on @var{condition}.
31678 @item -i @var{ignore-count}
31679 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
31680 to @var{ignore-count}.
31681 @item -p @var{thread-id}
31682 Restrict the breakpoint to the thread with the specified global
31686 @subsubheading Result
31688 @xref{GDB/MI Breakpoint Information}, for details on the format of the
31689 resulting breakpoint.
31691 @c An out-of-band breakpoint instead of part of the result?
31693 @subsubheading @value{GDBN} Command
31695 The corresponding @value{GDBN} command is @samp{dprintf}.
31697 @subsubheading Example
31701 4-dprintf-insert foo "At foo entry\n"
31702 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
31703 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
31704 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
31705 times="0",script=@{"printf \"At foo entry\\n\"","continue"@},
31706 original-location="foo"@}
31708 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
31709 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
31710 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
31711 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
31712 times="0",script=@{"printf \"arg=%d, g=%d\\n\", arg, g","continue"@},
31713 original-location="mi-dprintf.c:26"@}
31717 @subheading The @code{-break-list} Command
31718 @findex -break-list
31720 @subsubheading Synopsis
31726 Displays the list of inserted breakpoints, showing the following fields:
31730 number of the breakpoint
31732 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
31734 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
31737 is the breakpoint enabled or no: @samp{y} or @samp{n}
31739 memory location at which the breakpoint is set
31741 logical location of the breakpoint, expressed by function name, file
31743 @item Thread-groups
31744 list of thread groups to which this breakpoint applies
31746 number of times the breakpoint has been hit
31749 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
31750 @code{body} field is an empty list.
31752 @subsubheading @value{GDBN} Command
31754 The corresponding @value{GDBN} command is @samp{info break}.
31756 @subsubheading Example
31761 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
31762 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
31763 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
31764 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
31765 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
31766 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
31767 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
31768 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
31769 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
31771 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
31772 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
31773 line="13",thread-groups=["i1"],times="0"@}]@}
31777 Here's an example of the result when there are no breakpoints:
31782 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
31783 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
31784 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
31785 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
31786 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
31787 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
31788 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
31793 @subheading The @code{-break-passcount} Command
31794 @findex -break-passcount
31796 @subsubheading Synopsis
31799 -break-passcount @var{tracepoint-number} @var{passcount}
31802 Set the passcount for tracepoint @var{tracepoint-number} to
31803 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
31804 is not a tracepoint, error is emitted. This corresponds to CLI
31805 command @samp{passcount}.
31807 @subheading The @code{-break-watch} Command
31808 @findex -break-watch
31810 @subsubheading Synopsis
31813 -break-watch [ -a | -r ]
31816 Create a watchpoint. With the @samp{-a} option it will create an
31817 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
31818 read from or on a write to the memory location. With the @samp{-r}
31819 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
31820 trigger only when the memory location is accessed for reading. Without
31821 either of the options, the watchpoint created is a regular watchpoint,
31822 i.e., it will trigger when the memory location is accessed for writing.
31823 @xref{Set Watchpoints, , Setting Watchpoints}.
31825 Note that @samp{-break-list} will report a single list of watchpoints and
31826 breakpoints inserted.
31828 @subsubheading @value{GDBN} Command
31830 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
31833 @subsubheading Example
31835 Setting a watchpoint on a variable in the @code{main} function:
31840 ^done,wpt=@{number="2",exp="x"@}
31845 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
31846 value=@{old="-268439212",new="55"@},
31847 frame=@{func="main",args=[],file="recursive2.c",
31848 fullname="/home/foo/bar/recursive2.c",line="5",arch="i386:x86_64"@}
31852 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
31853 the program execution twice: first for the variable changing value, then
31854 for the watchpoint going out of scope.
31859 ^done,wpt=@{number="5",exp="C"@}
31864 *stopped,reason="watchpoint-trigger",
31865 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
31866 frame=@{func="callee4",args=[],
31867 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31868 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
31869 arch="i386:x86_64"@}
31874 *stopped,reason="watchpoint-scope",wpnum="5",
31875 frame=@{func="callee3",args=[@{name="strarg",
31876 value="0x11940 \"A string argument.\""@}],
31877 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31878 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
31879 arch="i386:x86_64"@}
31883 Listing breakpoints and watchpoints, at different points in the program
31884 execution. Note that once the watchpoint goes out of scope, it is
31890 ^done,wpt=@{number="2",exp="C"@}
31893 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
31894 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
31895 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
31896 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
31897 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
31898 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
31899 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
31900 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
31901 addr="0x00010734",func="callee4",
31902 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31903 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
31905 bkpt=@{number="2",type="watchpoint",disp="keep",
31906 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
31911 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
31912 value=@{old="-276895068",new="3"@},
31913 frame=@{func="callee4",args=[],
31914 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31915 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
31916 arch="i386:x86_64"@}
31919 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
31920 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
31921 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
31922 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
31923 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
31924 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
31925 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
31926 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
31927 addr="0x00010734",func="callee4",
31928 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31929 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
31931 bkpt=@{number="2",type="watchpoint",disp="keep",
31932 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
31936 ^done,reason="watchpoint-scope",wpnum="2",
31937 frame=@{func="callee3",args=[@{name="strarg",
31938 value="0x11940 \"A string argument.\""@}],
31939 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31940 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
31941 arch="i386:x86_64"@}
31944 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
31945 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
31946 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
31947 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
31948 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
31949 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
31950 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
31951 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
31952 addr="0x00010734",func="callee4",
31953 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31954 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
31955 thread-groups=["i1"],times="1"@}]@}
31960 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31961 @node GDB/MI Catchpoint Commands
31962 @section @sc{gdb/mi} Catchpoint Commands
31964 This section documents @sc{gdb/mi} commands for manipulating
31968 * Shared Library GDB/MI Catchpoint Commands::
31969 * Ada Exception GDB/MI Catchpoint Commands::
31970 * C++ Exception GDB/MI Catchpoint Commands::
31973 @node Shared Library GDB/MI Catchpoint Commands
31974 @subsection Shared Library @sc{gdb/mi} Catchpoints
31976 @subheading The @code{-catch-load} Command
31977 @findex -catch-load
31979 @subsubheading Synopsis
31982 -catch-load [ -t ] [ -d ] @var{regexp}
31985 Add a catchpoint for library load events. If the @samp{-t} option is used,
31986 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
31987 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
31988 in a disabled state. The @samp{regexp} argument is a regular
31989 expression used to match the name of the loaded library.
31992 @subsubheading @value{GDBN} Command
31994 The corresponding @value{GDBN} command is @samp{catch load}.
31996 @subsubheading Example
31999 -catch-load -t foo.so
32000 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
32001 what="load of library matching foo.so",catch-type="load",times="0"@}
32006 @subheading The @code{-catch-unload} Command
32007 @findex -catch-unload
32009 @subsubheading Synopsis
32012 -catch-unload [ -t ] [ -d ] @var{regexp}
32015 Add a catchpoint for library unload events. If the @samp{-t} option is
32016 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
32017 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
32018 created in a disabled state. The @samp{regexp} argument is a regular
32019 expression used to match the name of the unloaded library.
32021 @subsubheading @value{GDBN} Command
32023 The corresponding @value{GDBN} command is @samp{catch unload}.
32025 @subsubheading Example
32028 -catch-unload -d bar.so
32029 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
32030 what="load of library matching bar.so",catch-type="unload",times="0"@}
32034 @node Ada Exception GDB/MI Catchpoint Commands
32035 @subsection Ada Exception @sc{gdb/mi} Catchpoints
32037 The following @sc{gdb/mi} commands can be used to create catchpoints
32038 that stop the execution when Ada exceptions are being raised.
32040 @subheading The @code{-catch-assert} Command
32041 @findex -catch-assert
32043 @subsubheading Synopsis
32046 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
32049 Add a catchpoint for failed Ada assertions.
32051 The possible optional parameters for this command are:
32054 @item -c @var{condition}
32055 Make the catchpoint conditional on @var{condition}.
32057 Create a disabled catchpoint.
32059 Create a temporary catchpoint.
32062 @subsubheading @value{GDBN} Command
32064 The corresponding @value{GDBN} command is @samp{catch assert}.
32066 @subsubheading Example
32070 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
32071 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
32072 thread-groups=["i1"],times="0",
32073 original-location="__gnat_debug_raise_assert_failure"@}
32077 @subheading The @code{-catch-exception} Command
32078 @findex -catch-exception
32080 @subsubheading Synopsis
32083 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
32087 Add a catchpoint stopping when Ada exceptions are raised.
32088 By default, the command stops the program when any Ada exception
32089 gets raised. But it is also possible, by using some of the
32090 optional parameters described below, to create more selective
32093 The possible optional parameters for this command are:
32096 @item -c @var{condition}
32097 Make the catchpoint conditional on @var{condition}.
32099 Create a disabled catchpoint.
32100 @item -e @var{exception-name}
32101 Only stop when @var{exception-name} is raised. This option cannot
32102 be used combined with @samp{-u}.
32104 Create a temporary catchpoint.
32106 Stop only when an unhandled exception gets raised. This option
32107 cannot be used combined with @samp{-e}.
32110 @subsubheading @value{GDBN} Command
32112 The corresponding @value{GDBN} commands are @samp{catch exception}
32113 and @samp{catch exception unhandled}.
32115 @subsubheading Example
32118 -catch-exception -e Program_Error
32119 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
32120 enabled="y",addr="0x0000000000404874",
32121 what="`Program_Error' Ada exception", thread-groups=["i1"],
32122 times="0",original-location="__gnat_debug_raise_exception"@}
32126 @subheading The @code{-catch-handlers} Command
32127 @findex -catch-handlers
32129 @subsubheading Synopsis
32132 -catch-handlers [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
32136 Add a catchpoint stopping when Ada exceptions are handled.
32137 By default, the command stops the program when any Ada exception
32138 gets handled. But it is also possible, by using some of the
32139 optional parameters described below, to create more selective
32142 The possible optional parameters for this command are:
32145 @item -c @var{condition}
32146 Make the catchpoint conditional on @var{condition}.
32148 Create a disabled catchpoint.
32149 @item -e @var{exception-name}
32150 Only stop when @var{exception-name} is handled.
32152 Create a temporary catchpoint.
32155 @subsubheading @value{GDBN} Command
32157 The corresponding @value{GDBN} command is @samp{catch handlers}.
32159 @subsubheading Example
32162 -catch-handlers -e Constraint_Error
32163 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
32164 enabled="y",addr="0x0000000000402f68",
32165 what="`Constraint_Error' Ada exception handlers",thread-groups=["i1"],
32166 times="0",original-location="__gnat_begin_handler"@}
32170 @node C++ Exception GDB/MI Catchpoint Commands
32171 @subsection C@t{++} Exception @sc{gdb/mi} Catchpoints
32173 The following @sc{gdb/mi} commands can be used to create catchpoints
32174 that stop the execution when C@t{++} exceptions are being throw, rethrown,
32177 @subheading The @code{-catch-throw} Command
32178 @findex -catch-throw
32180 @subsubheading Synopsis
32183 -catch-throw [ -t ] [ -r @var{regexp}]
32186 Stop when the debuggee throws a C@t{++} exception. If @var{regexp} is
32187 given, then only exceptions whose type matches the regular expression
32190 If @samp{-t} is given, then the catchpoint is enabled only for one
32191 stop, the catchpoint is automatically deleted after stopping once for
32194 @subsubheading @value{GDBN} Command
32196 The corresponding @value{GDBN} commands are @samp{catch throw}
32197 and @samp{tcatch throw} (@pxref{Set Catchpoints}).
32199 @subsubheading Example
32202 -catch-throw -r exception_type
32203 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
32204 what="exception throw",catch-type="throw",
32205 thread-groups=["i1"],
32206 regexp="exception_type",times="0"@}
32212 ~"Catchpoint 1 (exception thrown), 0x00007ffff7ae00ed
32213 in __cxa_throw () from /lib64/libstdc++.so.6\n"
32214 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
32215 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_throw",
32216 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
32217 thread-id="1",stopped-threads="all",core="6"
32221 @subheading The @code{-catch-rethrow} Command
32222 @findex -catch-rethrow
32224 @subsubheading Synopsis
32227 -catch-rethrow [ -t ] [ -r @var{regexp}]
32230 Stop when a C@t{++} exception is re-thrown. If @var{regexp} is given,
32231 then only exceptions whose type matches the regular expression will be
32234 If @samp{-t} is given, then the catchpoint is enabled only for one
32235 stop, the catchpoint is automatically deleted after the first event is
32238 @subsubheading @value{GDBN} Command
32240 The corresponding @value{GDBN} commands are @samp{catch rethrow}
32241 and @samp{tcatch rethrow} (@pxref{Set Catchpoints}).
32243 @subsubheading Example
32246 -catch-rethrow -r exception_type
32247 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
32248 what="exception rethrow",catch-type="rethrow",
32249 thread-groups=["i1"],
32250 regexp="exception_type",times="0"@}
32256 ~"Catchpoint 1 (exception rethrown), 0x00007ffff7ae00ed
32257 in __cxa_rethrow () from /lib64/libstdc++.so.6\n"
32258 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
32259 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_rethrow",
32260 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
32261 thread-id="1",stopped-threads="all",core="6"
32265 @subheading The @code{-catch-catch} Command
32266 @findex -catch-catch
32268 @subsubheading Synopsis
32271 -catch-catch [ -t ] [ -r @var{regexp}]
32274 Stop when the debuggee catches a C@t{++} exception. If @var{regexp}
32275 is given, then only exceptions whose type matches the regular
32276 expression will be caught.
32278 If @samp{-t} is given, then the catchpoint is enabled only for one
32279 stop, the catchpoint is automatically deleted after the first event is
32282 @subsubheading @value{GDBN} Command
32284 The corresponding @value{GDBN} commands are @samp{catch catch}
32285 and @samp{tcatch catch} (@pxref{Set Catchpoints}).
32287 @subsubheading Example
32290 -catch-catch -r exception_type
32291 ^done,bkpt=@{number="1",type="catchpoint",disp="keep",enabled="y",
32292 what="exception catch",catch-type="catch",
32293 thread-groups=["i1"],
32294 regexp="exception_type",times="0"@}
32300 ~"Catchpoint 1 (exception caught), 0x00007ffff7ae00ed
32301 in __cxa_begin_catch () from /lib64/libstdc++.so.6\n"
32302 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
32303 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_begin_catch",
32304 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
32305 thread-id="1",stopped-threads="all",core="6"
32309 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32310 @node GDB/MI Program Context
32311 @section @sc{gdb/mi} Program Context
32313 @subheading The @code{-exec-arguments} Command
32314 @findex -exec-arguments
32317 @subsubheading Synopsis
32320 -exec-arguments @var{args}
32323 Set the inferior program arguments, to be used in the next
32326 @subsubheading @value{GDBN} Command
32328 The corresponding @value{GDBN} command is @samp{set args}.
32330 @subsubheading Example
32334 -exec-arguments -v word
32341 @subheading The @code{-exec-show-arguments} Command
32342 @findex -exec-show-arguments
32344 @subsubheading Synopsis
32347 -exec-show-arguments
32350 Print the arguments of the program.
32352 @subsubheading @value{GDBN} Command
32354 The corresponding @value{GDBN} command is @samp{show args}.
32356 @subsubheading Example
32361 @subheading The @code{-environment-cd} Command
32362 @findex -environment-cd
32364 @subsubheading Synopsis
32367 -environment-cd @var{pathdir}
32370 Set @value{GDBN}'s working directory.
32372 @subsubheading @value{GDBN} Command
32374 The corresponding @value{GDBN} command is @samp{cd}.
32376 @subsubheading Example
32380 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
32386 @subheading The @code{-environment-directory} Command
32387 @findex -environment-directory
32389 @subsubheading Synopsis
32392 -environment-directory [ -r ] [ @var{pathdir} ]+
32395 Add directories @var{pathdir} to beginning of search path for source files.
32396 If the @samp{-r} option is used, the search path is reset to the default
32397 search path. If directories @var{pathdir} are supplied in addition to the
32398 @samp{-r} option, the search path is first reset and then addition
32400 Multiple directories may be specified, separated by blanks. Specifying
32401 multiple directories in a single command
32402 results in the directories added to the beginning of the
32403 search path in the same order they were presented in the command.
32404 If blanks are needed as
32405 part of a directory name, double-quotes should be used around
32406 the name. In the command output, the path will show up separated
32407 by the system directory-separator character. The directory-separator
32408 character must not be used
32409 in any directory name.
32410 If no directories are specified, the current search path is displayed.
32412 @subsubheading @value{GDBN} Command
32414 The corresponding @value{GDBN} command is @samp{dir}.
32416 @subsubheading Example
32420 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
32421 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
32423 -environment-directory ""
32424 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
32426 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
32427 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
32429 -environment-directory -r
32430 ^done,source-path="$cdir:$cwd"
32435 @subheading The @code{-environment-path} Command
32436 @findex -environment-path
32438 @subsubheading Synopsis
32441 -environment-path [ -r ] [ @var{pathdir} ]+
32444 Add directories @var{pathdir} to beginning of search path for object files.
32445 If the @samp{-r} option is used, the search path is reset to the original
32446 search path that existed at gdb start-up. If directories @var{pathdir} are
32447 supplied in addition to the
32448 @samp{-r} option, the search path is first reset and then addition
32450 Multiple directories may be specified, separated by blanks. Specifying
32451 multiple directories in a single command
32452 results in the directories added to the beginning of the
32453 search path in the same order they were presented in the command.
32454 If blanks are needed as
32455 part of a directory name, double-quotes should be used around
32456 the name. In the command output, the path will show up separated
32457 by the system directory-separator character. The directory-separator
32458 character must not be used
32459 in any directory name.
32460 If no directories are specified, the current path is displayed.
32463 @subsubheading @value{GDBN} Command
32465 The corresponding @value{GDBN} command is @samp{path}.
32467 @subsubheading Example
32472 ^done,path="/usr/bin"
32474 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
32475 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
32477 -environment-path -r /usr/local/bin
32478 ^done,path="/usr/local/bin:/usr/bin"
32483 @subheading The @code{-environment-pwd} Command
32484 @findex -environment-pwd
32486 @subsubheading Synopsis
32492 Show the current working directory.
32494 @subsubheading @value{GDBN} Command
32496 The corresponding @value{GDBN} command is @samp{pwd}.
32498 @subsubheading Example
32503 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
32507 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32508 @node GDB/MI Thread Commands
32509 @section @sc{gdb/mi} Thread Commands
32512 @subheading The @code{-thread-info} Command
32513 @findex -thread-info
32515 @subsubheading Synopsis
32518 -thread-info [ @var{thread-id} ]
32521 Reports information about either a specific thread, if the
32522 @var{thread-id} parameter is present, or about all threads.
32523 @var{thread-id} is the thread's global thread ID. When printing
32524 information about all threads, also reports the global ID of the
32527 @subsubheading @value{GDBN} Command
32529 The @samp{info thread} command prints the same information
32532 @subsubheading Result
32534 The result contains the following attributes:
32538 A list of threads. The format of the elements of the list is described in
32539 @ref{GDB/MI Thread Information}.
32541 @item current-thread-id
32542 The global id of the currently selected thread. This field is omitted if there
32543 is no selected thread (for example, when the selected inferior is not running,
32544 and therefore has no threads) or if a @var{thread-id} argument was passed to
32549 @subsubheading Example
32554 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
32555 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
32556 args=[]@},state="running"@},
32557 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
32558 frame=@{level="0",addr="0x0804891f",func="foo",
32559 args=[@{name="i",value="10"@}],
32560 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},
32561 state="running"@}],
32562 current-thread-id="1"
32566 @subheading The @code{-thread-list-ids} Command
32567 @findex -thread-list-ids
32569 @subsubheading Synopsis
32575 Produces a list of the currently known global @value{GDBN} thread ids.
32576 At the end of the list it also prints the total number of such
32579 This command is retained for historical reasons, the
32580 @code{-thread-info} command should be used instead.
32582 @subsubheading @value{GDBN} Command
32584 Part of @samp{info threads} supplies the same information.
32586 @subsubheading Example
32591 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
32592 current-thread-id="1",number-of-threads="3"
32597 @subheading The @code{-thread-select} Command
32598 @findex -thread-select
32600 @subsubheading Synopsis
32603 -thread-select @var{thread-id}
32606 Make thread with global thread number @var{thread-id} the current
32607 thread. It prints the number of the new current thread, and the
32608 topmost frame for that thread.
32610 This command is deprecated in favor of explicitly using the
32611 @samp{--thread} option to each command.
32613 @subsubheading @value{GDBN} Command
32615 The corresponding @value{GDBN} command is @samp{thread}.
32617 @subsubheading Example
32624 *stopped,reason="end-stepping-range",thread-id="2",line="187",
32625 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
32629 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
32630 number-of-threads="3"
32633 ^done,new-thread-id="3",
32634 frame=@{level="0",func="vprintf",
32635 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
32636 @{name="arg",value="0x2"@}],file="vprintf.c",line="31",arch="i386:x86_64"@}
32640 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32641 @node GDB/MI Ada Tasking Commands
32642 @section @sc{gdb/mi} Ada Tasking Commands
32644 @subheading The @code{-ada-task-info} Command
32645 @findex -ada-task-info
32647 @subsubheading Synopsis
32650 -ada-task-info [ @var{task-id} ]
32653 Reports information about either a specific Ada task, if the
32654 @var{task-id} parameter is present, or about all Ada tasks.
32656 @subsubheading @value{GDBN} Command
32658 The @samp{info tasks} command prints the same information
32659 about all Ada tasks (@pxref{Ada Tasks}).
32661 @subsubheading Result
32663 The result is a table of Ada tasks. The following columns are
32664 defined for each Ada task:
32668 This field exists only for the current thread. It has the value @samp{*}.
32671 The identifier that @value{GDBN} uses to refer to the Ada task.
32674 The identifier that the target uses to refer to the Ada task.
32677 The global thread identifier of the thread corresponding to the Ada
32680 This field should always exist, as Ada tasks are always implemented
32681 on top of a thread. But if @value{GDBN} cannot find this corresponding
32682 thread for any reason, the field is omitted.
32685 This field exists only when the task was created by another task.
32686 In this case, it provides the ID of the parent task.
32689 The base priority of the task.
32692 The current state of the task. For a detailed description of the
32693 possible states, see @ref{Ada Tasks}.
32696 The name of the task.
32700 @subsubheading Example
32704 ^done,tasks=@{nr_rows="3",nr_cols="8",
32705 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
32706 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
32707 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
32708 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
32709 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
32710 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
32711 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
32712 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
32713 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
32714 state="Child Termination Wait",name="main_task"@}]@}
32718 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32719 @node GDB/MI Program Execution
32720 @section @sc{gdb/mi} Program Execution
32722 These are the asynchronous commands which generate the out-of-band
32723 record @samp{*stopped}. Currently @value{GDBN} only really executes
32724 asynchronously with remote targets and this interaction is mimicked in
32727 @subheading The @code{-exec-continue} Command
32728 @findex -exec-continue
32730 @subsubheading Synopsis
32733 -exec-continue [--reverse] [--all|--thread-group N]
32736 Resumes the execution of the inferior program, which will continue
32737 to execute until it reaches a debugger stop event. If the
32738 @samp{--reverse} option is specified, execution resumes in reverse until
32739 it reaches a stop event. Stop events may include
32742 breakpoints or watchpoints
32744 signals or exceptions
32746 the end of the process (or its beginning under @samp{--reverse})
32748 the end or beginning of a replay log if one is being used.
32750 In all-stop mode (@pxref{All-Stop
32751 Mode}), may resume only one thread, or all threads, depending on the
32752 value of the @samp{scheduler-locking} variable. If @samp{--all} is
32753 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
32754 ignored in all-stop mode. If the @samp{--thread-group} options is
32755 specified, then all threads in that thread group are resumed.
32757 @subsubheading @value{GDBN} Command
32759 The corresponding @value{GDBN} corresponding is @samp{continue}.
32761 @subsubheading Example
32768 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
32769 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
32770 line="13",arch="i386:x86_64"@}
32775 @subheading The @code{-exec-finish} Command
32776 @findex -exec-finish
32778 @subsubheading Synopsis
32781 -exec-finish [--reverse]
32784 Resumes the execution of the inferior program until the current
32785 function is exited. Displays the results returned by the function.
32786 If the @samp{--reverse} option is specified, resumes the reverse
32787 execution of the inferior program until the point where current
32788 function was called.
32790 @subsubheading @value{GDBN} Command
32792 The corresponding @value{GDBN} command is @samp{finish}.
32794 @subsubheading Example
32796 Function returning @code{void}.
32803 *stopped,reason="function-finished",frame=@{func="main",args=[],
32804 file="hello.c",fullname="/home/foo/bar/hello.c",line="7",arch="i386:x86_64"@}
32808 Function returning other than @code{void}. The name of the internal
32809 @value{GDBN} variable storing the result is printed, together with the
32816 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
32817 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
32818 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
32819 arch="i386:x86_64"@},
32820 gdb-result-var="$1",return-value="0"
32825 @subheading The @code{-exec-interrupt} Command
32826 @findex -exec-interrupt
32828 @subsubheading Synopsis
32831 -exec-interrupt [--all|--thread-group N]
32834 Interrupts the background execution of the target. Note how the token
32835 associated with the stop message is the one for the execution command
32836 that has been interrupted. The token for the interrupt itself only
32837 appears in the @samp{^done} output. If the user is trying to
32838 interrupt a non-running program, an error message will be printed.
32840 Note that when asynchronous execution is enabled, this command is
32841 asynchronous just like other execution commands. That is, first the
32842 @samp{^done} response will be printed, and the target stop will be
32843 reported after that using the @samp{*stopped} notification.
32845 In non-stop mode, only the context thread is interrupted by default.
32846 All threads (in all inferiors) will be interrupted if the
32847 @samp{--all} option is specified. If the @samp{--thread-group}
32848 option is specified, all threads in that group will be interrupted.
32850 @subsubheading @value{GDBN} Command
32852 The corresponding @value{GDBN} command is @samp{interrupt}.
32854 @subsubheading Example
32865 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
32866 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
32867 fullname="/home/foo/bar/try.c",line="13",arch="i386:x86_64"@}
32872 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
32876 @subheading The @code{-exec-jump} Command
32879 @subsubheading Synopsis
32882 -exec-jump @var{location}
32885 Resumes execution of the inferior program at the location specified by
32886 parameter. @xref{Specify Location}, for a description of the
32887 different forms of @var{location}.
32889 @subsubheading @value{GDBN} Command
32891 The corresponding @value{GDBN} command is @samp{jump}.
32893 @subsubheading Example
32896 -exec-jump foo.c:10
32897 *running,thread-id="all"
32902 @subheading The @code{-exec-next} Command
32905 @subsubheading Synopsis
32908 -exec-next [--reverse]
32911 Resumes execution of the inferior program, stopping when the beginning
32912 of the next source line is reached.
32914 If the @samp{--reverse} option is specified, resumes reverse execution
32915 of the inferior program, stopping at the beginning of the previous
32916 source line. If you issue this command on the first line of a
32917 function, it will take you back to the caller of that function, to the
32918 source line where the function was called.
32921 @subsubheading @value{GDBN} Command
32923 The corresponding @value{GDBN} command is @samp{next}.
32925 @subsubheading Example
32931 *stopped,reason="end-stepping-range",line="8",file="hello.c"
32936 @subheading The @code{-exec-next-instruction} Command
32937 @findex -exec-next-instruction
32939 @subsubheading Synopsis
32942 -exec-next-instruction [--reverse]
32945 Executes one machine instruction. If the instruction is a function
32946 call, continues until the function returns. If the program stops at an
32947 instruction in the middle of a source line, the address will be
32950 If the @samp{--reverse} option is specified, resumes reverse execution
32951 of the inferior program, stopping at the previous instruction. If the
32952 previously executed instruction was a return from another function,
32953 it will continue to execute in reverse until the call to that function
32954 (from the current stack frame) is reached.
32956 @subsubheading @value{GDBN} Command
32958 The corresponding @value{GDBN} command is @samp{nexti}.
32960 @subsubheading Example
32964 -exec-next-instruction
32968 *stopped,reason="end-stepping-range",
32969 addr="0x000100d4",line="5",file="hello.c"
32974 @subheading The @code{-exec-return} Command
32975 @findex -exec-return
32977 @subsubheading Synopsis
32983 Makes current function return immediately. Doesn't execute the inferior.
32984 Displays the new current frame.
32986 @subsubheading @value{GDBN} Command
32988 The corresponding @value{GDBN} command is @samp{return}.
32990 @subsubheading Example
32994 200-break-insert callee4
32995 200^done,bkpt=@{number="1",addr="0x00010734",
32996 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
33001 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
33002 frame=@{func="callee4",args=[],
33003 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33004 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
33005 arch="i386:x86_64"@}
33011 111^done,frame=@{level="0",func="callee3",
33012 args=[@{name="strarg",
33013 value="0x11940 \"A string argument.\""@}],
33014 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33015 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
33016 arch="i386:x86_64"@}
33021 @subheading The @code{-exec-run} Command
33024 @subsubheading Synopsis
33027 -exec-run [ --all | --thread-group N ] [ --start ]
33030 Starts execution of the inferior from the beginning. The inferior
33031 executes until either a breakpoint is encountered or the program
33032 exits. In the latter case the output will include an exit code, if
33033 the program has exited exceptionally.
33035 When neither the @samp{--all} nor the @samp{--thread-group} option
33036 is specified, the current inferior is started. If the
33037 @samp{--thread-group} option is specified, it should refer to a thread
33038 group of type @samp{process}, and that thread group will be started.
33039 If the @samp{--all} option is specified, then all inferiors will be started.
33041 Using the @samp{--start} option instructs the debugger to stop
33042 the execution at the start of the inferior's main subprogram,
33043 following the same behavior as the @code{start} command
33044 (@pxref{Starting}).
33046 @subsubheading @value{GDBN} Command
33048 The corresponding @value{GDBN} command is @samp{run}.
33050 @subsubheading Examples
33055 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
33060 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
33061 frame=@{func="main",args=[],file="recursive2.c",
33062 fullname="/home/foo/bar/recursive2.c",line="4",arch="i386:x86_64"@}
33067 Program exited normally:
33075 *stopped,reason="exited-normally"
33080 Program exited exceptionally:
33088 *stopped,reason="exited",exit-code="01"
33092 Another way the program can terminate is if it receives a signal such as
33093 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
33097 *stopped,reason="exited-signalled",signal-name="SIGINT",
33098 signal-meaning="Interrupt"
33102 @c @subheading -exec-signal
33105 @subheading The @code{-exec-step} Command
33108 @subsubheading Synopsis
33111 -exec-step [--reverse]
33114 Resumes execution of the inferior program, stopping when the beginning
33115 of the next source line is reached, if the next source line is not a
33116 function call. If it is, stop at the first instruction of the called
33117 function. If the @samp{--reverse} option is specified, resumes reverse
33118 execution of the inferior program, stopping at the beginning of the
33119 previously executed source line.
33121 @subsubheading @value{GDBN} Command
33123 The corresponding @value{GDBN} command is @samp{step}.
33125 @subsubheading Example
33127 Stepping into a function:
33133 *stopped,reason="end-stepping-range",
33134 frame=@{func="foo",args=[@{name="a",value="10"@},
33135 @{name="b",value="0"@}],file="recursive2.c",
33136 fullname="/home/foo/bar/recursive2.c",line="11",arch="i386:x86_64"@}
33146 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
33151 @subheading The @code{-exec-step-instruction} Command
33152 @findex -exec-step-instruction
33154 @subsubheading Synopsis
33157 -exec-step-instruction [--reverse]
33160 Resumes the inferior which executes one machine instruction. If the
33161 @samp{--reverse} option is specified, resumes reverse execution of the
33162 inferior program, stopping at the previously executed instruction.
33163 The output, once @value{GDBN} has stopped, will vary depending on
33164 whether we have stopped in the middle of a source line or not. In the
33165 former case, the address at which the program stopped will be printed
33168 @subsubheading @value{GDBN} Command
33170 The corresponding @value{GDBN} command is @samp{stepi}.
33172 @subsubheading Example
33176 -exec-step-instruction
33180 *stopped,reason="end-stepping-range",
33181 frame=@{func="foo",args=[],file="try.c",
33182 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
33184 -exec-step-instruction
33188 *stopped,reason="end-stepping-range",
33189 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
33190 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
33195 @subheading The @code{-exec-until} Command
33196 @findex -exec-until
33198 @subsubheading Synopsis
33201 -exec-until [ @var{location} ]
33204 Executes the inferior until the @var{location} specified in the
33205 argument is reached. If there is no argument, the inferior executes
33206 until a source line greater than the current one is reached. The
33207 reason for stopping in this case will be @samp{location-reached}.
33209 @subsubheading @value{GDBN} Command
33211 The corresponding @value{GDBN} command is @samp{until}.
33213 @subsubheading Example
33217 -exec-until recursive2.c:6
33221 *stopped,reason="location-reached",frame=@{func="main",args=[],
33222 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6",
33223 arch="i386:x86_64"@}
33228 @subheading -file-clear
33229 Is this going away????
33232 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33233 @node GDB/MI Stack Manipulation
33234 @section @sc{gdb/mi} Stack Manipulation Commands
33236 @subheading The @code{-enable-frame-filters} Command
33237 @findex -enable-frame-filters
33240 -enable-frame-filters
33243 @value{GDBN} allows Python-based frame filters to affect the output of
33244 the MI commands relating to stack traces. As there is no way to
33245 implement this in a fully backward-compatible way, a front end must
33246 request that this functionality be enabled.
33248 Once enabled, this feature cannot be disabled.
33250 Note that if Python support has not been compiled into @value{GDBN},
33251 this command will still succeed (and do nothing).
33253 @subheading The @code{-stack-info-frame} Command
33254 @findex -stack-info-frame
33256 @subsubheading Synopsis
33262 Get info on the selected frame.
33264 @subsubheading @value{GDBN} Command
33266 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
33267 (without arguments).
33269 @subsubheading Example
33274 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
33275 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33276 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
33277 arch="i386:x86_64"@}
33281 @subheading The @code{-stack-info-depth} Command
33282 @findex -stack-info-depth
33284 @subsubheading Synopsis
33287 -stack-info-depth [ @var{max-depth} ]
33290 Return the depth of the stack. If the integer argument @var{max-depth}
33291 is specified, do not count beyond @var{max-depth} frames.
33293 @subsubheading @value{GDBN} Command
33295 There's no equivalent @value{GDBN} command.
33297 @subsubheading Example
33299 For a stack with frame levels 0 through 11:
33306 -stack-info-depth 4
33309 -stack-info-depth 12
33312 -stack-info-depth 11
33315 -stack-info-depth 13
33320 @anchor{-stack-list-arguments}
33321 @subheading The @code{-stack-list-arguments} Command
33322 @findex -stack-list-arguments
33324 @subsubheading Synopsis
33327 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
33328 [ @var{low-frame} @var{high-frame} ]
33331 Display a list of the arguments for the frames between @var{low-frame}
33332 and @var{high-frame} (inclusive). If @var{low-frame} and
33333 @var{high-frame} are not provided, list the arguments for the whole
33334 call stack. If the two arguments are equal, show the single frame
33335 at the corresponding level. It is an error if @var{low-frame} is
33336 larger than the actual number of frames. On the other hand,
33337 @var{high-frame} may be larger than the actual number of frames, in
33338 which case only existing frames will be returned.
33340 If @var{print-values} is 0 or @code{--no-values}, print only the names of
33341 the variables; if it is 1 or @code{--all-values}, print also their
33342 values; and if it is 2 or @code{--simple-values}, print the name,
33343 type and value for simple data types, and the name and type for arrays,
33344 structures and unions. If the option @code{--no-frame-filters} is
33345 supplied, then Python frame filters will not be executed.
33347 If the @code{--skip-unavailable} option is specified, arguments that
33348 are not available are not listed. Partially available arguments
33349 are still displayed, however.
33351 Use of this command to obtain arguments in a single frame is
33352 deprecated in favor of the @samp{-stack-list-variables} command.
33354 @subsubheading @value{GDBN} Command
33356 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
33357 @samp{gdb_get_args} command which partially overlaps with the
33358 functionality of @samp{-stack-list-arguments}.
33360 @subsubheading Example
33367 frame=@{level="0",addr="0x00010734",func="callee4",
33368 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33369 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
33370 arch="i386:x86_64"@},
33371 frame=@{level="1",addr="0x0001076c",func="callee3",
33372 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33373 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
33374 arch="i386:x86_64"@},
33375 frame=@{level="2",addr="0x0001078c",func="callee2",
33376 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33377 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22",
33378 arch="i386:x86_64"@},
33379 frame=@{level="3",addr="0x000107b4",func="callee1",
33380 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33381 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27",
33382 arch="i386:x86_64"@},
33383 frame=@{level="4",addr="0x000107e0",func="main",
33384 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
33385 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32",
33386 arch="i386:x86_64"@}]
33388 -stack-list-arguments 0
33391 frame=@{level="0",args=[]@},
33392 frame=@{level="1",args=[name="strarg"]@},
33393 frame=@{level="2",args=[name="intarg",name="strarg"]@},
33394 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
33395 frame=@{level="4",args=[]@}]
33397 -stack-list-arguments 1
33400 frame=@{level="0",args=[]@},
33402 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
33403 frame=@{level="2",args=[
33404 @{name="intarg",value="2"@},
33405 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
33406 @{frame=@{level="3",args=[
33407 @{name="intarg",value="2"@},
33408 @{name="strarg",value="0x11940 \"A string argument.\""@},
33409 @{name="fltarg",value="3.5"@}]@},
33410 frame=@{level="4",args=[]@}]
33412 -stack-list-arguments 0 2 2
33413 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
33415 -stack-list-arguments 1 2 2
33416 ^done,stack-args=[frame=@{level="2",
33417 args=[@{name="intarg",value="2"@},
33418 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
33422 @c @subheading -stack-list-exception-handlers
33425 @anchor{-stack-list-frames}
33426 @subheading The @code{-stack-list-frames} Command
33427 @findex -stack-list-frames
33429 @subsubheading Synopsis
33432 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
33435 List the frames currently on the stack. For each frame it displays the
33440 The frame number, 0 being the topmost frame, i.e., the innermost function.
33442 The @code{$pc} value for that frame.
33446 File name of the source file where the function lives.
33447 @item @var{fullname}
33448 The full file name of the source file where the function lives.
33450 Line number corresponding to the @code{$pc}.
33452 The shared library where this function is defined. This is only given
33453 if the frame's function is not known.
33455 Frame's architecture.
33458 If invoked without arguments, this command prints a backtrace for the
33459 whole stack. If given two integer arguments, it shows the frames whose
33460 levels are between the two arguments (inclusive). If the two arguments
33461 are equal, it shows the single frame at the corresponding level. It is
33462 an error if @var{low-frame} is larger than the actual number of
33463 frames. On the other hand, @var{high-frame} may be larger than the
33464 actual number of frames, in which case only existing frames will be
33465 returned. If the option @code{--no-frame-filters} is supplied, then
33466 Python frame filters will not be executed.
33468 @subsubheading @value{GDBN} Command
33470 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
33472 @subsubheading Example
33474 Full stack backtrace:
33480 [frame=@{level="0",addr="0x0001076c",func="foo",
33481 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11",
33482 arch="i386:x86_64"@},
33483 frame=@{level="1",addr="0x000107a4",func="foo",
33484 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33485 arch="i386:x86_64"@},
33486 frame=@{level="2",addr="0x000107a4",func="foo",
33487 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33488 arch="i386:x86_64"@},
33489 frame=@{level="3",addr="0x000107a4",func="foo",
33490 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33491 arch="i386:x86_64"@},
33492 frame=@{level="4",addr="0x000107a4",func="foo",
33493 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33494 arch="i386:x86_64"@},
33495 frame=@{level="5",addr="0x000107a4",func="foo",
33496 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33497 arch="i386:x86_64"@},
33498 frame=@{level="6",addr="0x000107a4",func="foo",
33499 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33500 arch="i386:x86_64"@},
33501 frame=@{level="7",addr="0x000107a4",func="foo",
33502 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33503 arch="i386:x86_64"@},
33504 frame=@{level="8",addr="0x000107a4",func="foo",
33505 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33506 arch="i386:x86_64"@},
33507 frame=@{level="9",addr="0x000107a4",func="foo",
33508 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33509 arch="i386:x86_64"@},
33510 frame=@{level="10",addr="0x000107a4",func="foo",
33511 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33512 arch="i386:x86_64"@},
33513 frame=@{level="11",addr="0x00010738",func="main",
33514 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4",
33515 arch="i386:x86_64"@}]
33519 Show frames between @var{low_frame} and @var{high_frame}:
33523 -stack-list-frames 3 5
33525 [frame=@{level="3",addr="0x000107a4",func="foo",
33526 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33527 arch="i386:x86_64"@},
33528 frame=@{level="4",addr="0x000107a4",func="foo",
33529 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33530 arch="i386:x86_64"@},
33531 frame=@{level="5",addr="0x000107a4",func="foo",
33532 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33533 arch="i386:x86_64"@}]
33537 Show a single frame:
33541 -stack-list-frames 3 3
33543 [frame=@{level="3",addr="0x000107a4",func="foo",
33544 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
33545 arch="i386:x86_64"@}]
33550 @subheading The @code{-stack-list-locals} Command
33551 @findex -stack-list-locals
33552 @anchor{-stack-list-locals}
33554 @subsubheading Synopsis
33557 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
33560 Display the local variable names for the selected frame. If
33561 @var{print-values} is 0 or @code{--no-values}, print only the names of
33562 the variables; if it is 1 or @code{--all-values}, print also their
33563 values; and if it is 2 or @code{--simple-values}, print the name,
33564 type and value for simple data types, and the name and type for arrays,
33565 structures and unions. In this last case, a frontend can immediately
33566 display the value of simple data types and create variable objects for
33567 other data types when the user wishes to explore their values in
33568 more detail. If the option @code{--no-frame-filters} is supplied, then
33569 Python frame filters will not be executed.
33571 If the @code{--skip-unavailable} option is specified, local variables
33572 that are not available are not listed. Partially available local
33573 variables are still displayed, however.
33575 This command is deprecated in favor of the
33576 @samp{-stack-list-variables} command.
33578 @subsubheading @value{GDBN} Command
33580 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
33582 @subsubheading Example
33586 -stack-list-locals 0
33587 ^done,locals=[name="A",name="B",name="C"]
33589 -stack-list-locals --all-values
33590 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
33591 @{name="C",value="@{1, 2, 3@}"@}]
33592 -stack-list-locals --simple-values
33593 ^done,locals=[@{name="A",type="int",value="1"@},
33594 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
33598 @anchor{-stack-list-variables}
33599 @subheading The @code{-stack-list-variables} Command
33600 @findex -stack-list-variables
33602 @subsubheading Synopsis
33605 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
33608 Display the names of local variables and function arguments for the selected frame. If
33609 @var{print-values} is 0 or @code{--no-values}, print only the names of
33610 the variables; if it is 1 or @code{--all-values}, print also their
33611 values; and if it is 2 or @code{--simple-values}, print the name,
33612 type and value for simple data types, and the name and type for arrays,
33613 structures and unions. If the option @code{--no-frame-filters} is
33614 supplied, then Python frame filters will not be executed.
33616 If the @code{--skip-unavailable} option is specified, local variables
33617 and arguments that are not available are not listed. Partially
33618 available arguments and local variables are still displayed, however.
33620 @subsubheading Example
33624 -stack-list-variables --thread 1 --frame 0 --all-values
33625 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
33630 @subheading The @code{-stack-select-frame} Command
33631 @findex -stack-select-frame
33633 @subsubheading Synopsis
33636 -stack-select-frame @var{framenum}
33639 Change the selected frame. Select a different frame @var{framenum} on
33642 This command in deprecated in favor of passing the @samp{--frame}
33643 option to every command.
33645 @subsubheading @value{GDBN} Command
33647 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
33648 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
33650 @subsubheading Example
33654 -stack-select-frame 2
33659 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33660 @node GDB/MI Variable Objects
33661 @section @sc{gdb/mi} Variable Objects
33665 @subheading Motivation for Variable Objects in @sc{gdb/mi}
33667 For the implementation of a variable debugger window (locals, watched
33668 expressions, etc.), we are proposing the adaptation of the existing code
33669 used by @code{Insight}.
33671 The two main reasons for that are:
33675 It has been proven in practice (it is already on its second generation).
33678 It will shorten development time (needless to say how important it is
33682 The original interface was designed to be used by Tcl code, so it was
33683 slightly changed so it could be used through @sc{gdb/mi}. This section
33684 describes the @sc{gdb/mi} operations that will be available and gives some
33685 hints about their use.
33687 @emph{Note}: In addition to the set of operations described here, we
33688 expect the @sc{gui} implementation of a variable window to require, at
33689 least, the following operations:
33692 @item @code{-gdb-show} @code{output-radix}
33693 @item @code{-stack-list-arguments}
33694 @item @code{-stack-list-locals}
33695 @item @code{-stack-select-frame}
33700 @subheading Introduction to Variable Objects
33702 @cindex variable objects in @sc{gdb/mi}
33704 Variable objects are "object-oriented" MI interface for examining and
33705 changing values of expressions. Unlike some other MI interfaces that
33706 work with expressions, variable objects are specifically designed for
33707 simple and efficient presentation in the frontend. A variable object
33708 is identified by string name. When a variable object is created, the
33709 frontend specifies the expression for that variable object. The
33710 expression can be a simple variable, or it can be an arbitrary complex
33711 expression, and can even involve CPU registers. After creating a
33712 variable object, the frontend can invoke other variable object
33713 operations---for example to obtain or change the value of a variable
33714 object, or to change display format.
33716 Variable objects have hierarchical tree structure. Any variable object
33717 that corresponds to a composite type, such as structure in C, has
33718 a number of child variable objects, for example corresponding to each
33719 element of a structure. A child variable object can itself have
33720 children, recursively. Recursion ends when we reach
33721 leaf variable objects, which always have built-in types. Child variable
33722 objects are created only by explicit request, so if a frontend
33723 is not interested in the children of a particular variable object, no
33724 child will be created.
33726 For a leaf variable object it is possible to obtain its value as a
33727 string, or set the value from a string. String value can be also
33728 obtained for a non-leaf variable object, but it's generally a string
33729 that only indicates the type of the object, and does not list its
33730 contents. Assignment to a non-leaf variable object is not allowed.
33732 A frontend does not need to read the values of all variable objects each time
33733 the program stops. Instead, MI provides an update command that lists all
33734 variable objects whose values has changed since the last update
33735 operation. This considerably reduces the amount of data that must
33736 be transferred to the frontend. As noted above, children variable
33737 objects are created on demand, and only leaf variable objects have a
33738 real value. As result, gdb will read target memory only for leaf
33739 variables that frontend has created.
33741 The automatic update is not always desirable. For example, a frontend
33742 might want to keep a value of some expression for future reference,
33743 and never update it. For another example, fetching memory is
33744 relatively slow for embedded targets, so a frontend might want
33745 to disable automatic update for the variables that are either not
33746 visible on the screen, or ``closed''. This is possible using so
33747 called ``frozen variable objects''. Such variable objects are never
33748 implicitly updated.
33750 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
33751 fixed variable object, the expression is parsed when the variable
33752 object is created, including associating identifiers to specific
33753 variables. The meaning of expression never changes. For a floating
33754 variable object the values of variables whose names appear in the
33755 expressions are re-evaluated every time in the context of the current
33756 frame. Consider this example:
33761 struct work_state state;
33768 If a fixed variable object for the @code{state} variable is created in
33769 this function, and we enter the recursive call, the variable
33770 object will report the value of @code{state} in the top-level
33771 @code{do_work} invocation. On the other hand, a floating variable
33772 object will report the value of @code{state} in the current frame.
33774 If an expression specified when creating a fixed variable object
33775 refers to a local variable, the variable object becomes bound to the
33776 thread and frame in which the variable object is created. When such
33777 variable object is updated, @value{GDBN} makes sure that the
33778 thread/frame combination the variable object is bound to still exists,
33779 and re-evaluates the variable object in context of that thread/frame.
33781 The following is the complete set of @sc{gdb/mi} operations defined to
33782 access this functionality:
33784 @multitable @columnfractions .4 .6
33785 @item @strong{Operation}
33786 @tab @strong{Description}
33788 @item @code{-enable-pretty-printing}
33789 @tab enable Python-based pretty-printing
33790 @item @code{-var-create}
33791 @tab create a variable object
33792 @item @code{-var-delete}
33793 @tab delete the variable object and/or its children
33794 @item @code{-var-set-format}
33795 @tab set the display format of this variable
33796 @item @code{-var-show-format}
33797 @tab show the display format of this variable
33798 @item @code{-var-info-num-children}
33799 @tab tells how many children this object has
33800 @item @code{-var-list-children}
33801 @tab return a list of the object's children
33802 @item @code{-var-info-type}
33803 @tab show the type of this variable object
33804 @item @code{-var-info-expression}
33805 @tab print parent-relative expression that this variable object represents
33806 @item @code{-var-info-path-expression}
33807 @tab print full expression that this variable object represents
33808 @item @code{-var-show-attributes}
33809 @tab is this variable editable? does it exist here?
33810 @item @code{-var-evaluate-expression}
33811 @tab get the value of this variable
33812 @item @code{-var-assign}
33813 @tab set the value of this variable
33814 @item @code{-var-update}
33815 @tab update the variable and its children
33816 @item @code{-var-set-frozen}
33817 @tab set frozenness attribute
33818 @item @code{-var-set-update-range}
33819 @tab set range of children to display on update
33822 In the next subsection we describe each operation in detail and suggest
33823 how it can be used.
33825 @subheading Description And Use of Operations on Variable Objects
33827 @subheading The @code{-enable-pretty-printing} Command
33828 @findex -enable-pretty-printing
33831 -enable-pretty-printing
33834 @value{GDBN} allows Python-based visualizers to affect the output of the
33835 MI variable object commands. However, because there was no way to
33836 implement this in a fully backward-compatible way, a front end must
33837 request that this functionality be enabled.
33839 Once enabled, this feature cannot be disabled.
33841 Note that if Python support has not been compiled into @value{GDBN},
33842 this command will still succeed (and do nothing).
33844 This feature is currently (as of @value{GDBN} 7.0) experimental, and
33845 may work differently in future versions of @value{GDBN}.
33847 @subheading The @code{-var-create} Command
33848 @findex -var-create
33850 @subsubheading Synopsis
33853 -var-create @{@var{name} | "-"@}
33854 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
33857 This operation creates a variable object, which allows the monitoring of
33858 a variable, the result of an expression, a memory cell or a CPU
33861 The @var{name} parameter is the string by which the object can be
33862 referenced. It must be unique. If @samp{-} is specified, the varobj
33863 system will generate a string ``varNNNNNN'' automatically. It will be
33864 unique provided that one does not specify @var{name} of that format.
33865 The command fails if a duplicate name is found.
33867 The frame under which the expression should be evaluated can be
33868 specified by @var{frame-addr}. A @samp{*} indicates that the current
33869 frame should be used. A @samp{@@} indicates that a floating variable
33870 object must be created.
33872 @var{expression} is any expression valid on the current language set (must not
33873 begin with a @samp{*}), or one of the following:
33877 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
33880 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
33883 @samp{$@var{regname}} --- a CPU register name
33886 @cindex dynamic varobj
33887 A varobj's contents may be provided by a Python-based pretty-printer. In this
33888 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
33889 have slightly different semantics in some cases. If the
33890 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
33891 will never create a dynamic varobj. This ensures backward
33892 compatibility for existing clients.
33894 @subsubheading Result
33896 This operation returns attributes of the newly-created varobj. These
33901 The name of the varobj.
33904 The number of children of the varobj. This number is not necessarily
33905 reliable for a dynamic varobj. Instead, you must examine the
33906 @samp{has_more} attribute.
33909 The varobj's scalar value. For a varobj whose type is some sort of
33910 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
33911 will not be interesting.
33914 The varobj's type. This is a string representation of the type, as
33915 would be printed by the @value{GDBN} CLI. If @samp{print object}
33916 (@pxref{Print Settings, set print object}) is set to @code{on}, the
33917 @emph{actual} (derived) type of the object is shown rather than the
33918 @emph{declared} one.
33921 If a variable object is bound to a specific thread, then this is the
33922 thread's global identifier.
33925 For a dynamic varobj, this indicates whether there appear to be any
33926 children available. For a non-dynamic varobj, this will be 0.
33929 This attribute will be present and have the value @samp{1} if the
33930 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
33931 then this attribute will not be present.
33934 A dynamic varobj can supply a display hint to the front end. The
33935 value comes directly from the Python pretty-printer object's
33936 @code{display_hint} method. @xref{Pretty Printing API}.
33939 Typical output will look like this:
33942 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
33943 has_more="@var{has_more}"
33947 @subheading The @code{-var-delete} Command
33948 @findex -var-delete
33950 @subsubheading Synopsis
33953 -var-delete [ -c ] @var{name}
33956 Deletes a previously created variable object and all of its children.
33957 With the @samp{-c} option, just deletes the children.
33959 Returns an error if the object @var{name} is not found.
33962 @subheading The @code{-var-set-format} Command
33963 @findex -var-set-format
33965 @subsubheading Synopsis
33968 -var-set-format @var{name} @var{format-spec}
33971 Sets the output format for the value of the object @var{name} to be
33974 @anchor{-var-set-format}
33975 The syntax for the @var{format-spec} is as follows:
33978 @var{format-spec} @expansion{}
33979 @{binary | decimal | hexadecimal | octal | natural | zero-hexadecimal@}
33982 The natural format is the default format choosen automatically
33983 based on the variable type (like decimal for an @code{int}, hex
33984 for pointers, etc.).
33986 The zero-hexadecimal format has a representation similar to hexadecimal
33987 but with padding zeroes to the left of the value. For example, a 32-bit
33988 hexadecimal value of 0x1234 would be represented as 0x00001234 in the
33989 zero-hexadecimal format.
33991 For a variable with children, the format is set only on the
33992 variable itself, and the children are not affected.
33994 @subheading The @code{-var-show-format} Command
33995 @findex -var-show-format
33997 @subsubheading Synopsis
34000 -var-show-format @var{name}
34003 Returns the format used to display the value of the object @var{name}.
34006 @var{format} @expansion{}
34011 @subheading The @code{-var-info-num-children} Command
34012 @findex -var-info-num-children
34014 @subsubheading Synopsis
34017 -var-info-num-children @var{name}
34020 Returns the number of children of a variable object @var{name}:
34026 Note that this number is not completely reliable for a dynamic varobj.
34027 It will return the current number of children, but more children may
34031 @subheading The @code{-var-list-children} Command
34032 @findex -var-list-children
34034 @subsubheading Synopsis
34037 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
34039 @anchor{-var-list-children}
34041 Return a list of the children of the specified variable object and
34042 create variable objects for them, if they do not already exist. With
34043 a single argument or if @var{print-values} has a value of 0 or
34044 @code{--no-values}, print only the names of the variables; if
34045 @var{print-values} is 1 or @code{--all-values}, also print their
34046 values; and if it is 2 or @code{--simple-values} print the name and
34047 value for simple data types and just the name for arrays, structures
34050 @var{from} and @var{to}, if specified, indicate the range of children
34051 to report. If @var{from} or @var{to} is less than zero, the range is
34052 reset and all children will be reported. Otherwise, children starting
34053 at @var{from} (zero-based) and up to and excluding @var{to} will be
34056 If a child range is requested, it will only affect the current call to
34057 @code{-var-list-children}, but not future calls to @code{-var-update}.
34058 For this, you must instead use @code{-var-set-update-range}. The
34059 intent of this approach is to enable a front end to implement any
34060 update approach it likes; for example, scrolling a view may cause the
34061 front end to request more children with @code{-var-list-children}, and
34062 then the front end could call @code{-var-set-update-range} with a
34063 different range to ensure that future updates are restricted to just
34066 For each child the following results are returned:
34071 Name of the variable object created for this child.
34074 The expression to be shown to the user by the front end to designate this child.
34075 For example this may be the name of a structure member.
34077 For a dynamic varobj, this value cannot be used to form an
34078 expression. There is no way to do this at all with a dynamic varobj.
34080 For C/C@t{++} structures there are several pseudo children returned to
34081 designate access qualifiers. For these pseudo children @var{exp} is
34082 @samp{public}, @samp{private}, or @samp{protected}. In this case the
34083 type and value are not present.
34085 A dynamic varobj will not report the access qualifying
34086 pseudo-children, regardless of the language. This information is not
34087 available at all with a dynamic varobj.
34090 Number of children this child has. For a dynamic varobj, this will be
34094 The type of the child. If @samp{print object}
34095 (@pxref{Print Settings, set print object}) is set to @code{on}, the
34096 @emph{actual} (derived) type of the object is shown rather than the
34097 @emph{declared} one.
34100 If values were requested, this is the value.
34103 If this variable object is associated with a thread, this is the
34104 thread's global thread id. Otherwise this result is not present.
34107 If the variable object is frozen, this variable will be present with a value of 1.
34110 A dynamic varobj can supply a display hint to the front end. The
34111 value comes directly from the Python pretty-printer object's
34112 @code{display_hint} method. @xref{Pretty Printing API}.
34115 This attribute will be present and have the value @samp{1} if the
34116 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
34117 then this attribute will not be present.
34121 The result may have its own attributes:
34125 A dynamic varobj can supply a display hint to the front end. The
34126 value comes directly from the Python pretty-printer object's
34127 @code{display_hint} method. @xref{Pretty Printing API}.
34130 This is an integer attribute which is nonzero if there are children
34131 remaining after the end of the selected range.
34134 @subsubheading Example
34138 -var-list-children n
34139 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
34140 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
34142 -var-list-children --all-values n
34143 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
34144 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
34148 @subheading The @code{-var-info-type} Command
34149 @findex -var-info-type
34151 @subsubheading Synopsis
34154 -var-info-type @var{name}
34157 Returns the type of the specified variable @var{name}. The type is
34158 returned as a string in the same format as it is output by the
34162 type=@var{typename}
34166 @subheading The @code{-var-info-expression} Command
34167 @findex -var-info-expression
34169 @subsubheading Synopsis
34172 -var-info-expression @var{name}
34175 Returns a string that is suitable for presenting this
34176 variable object in user interface. The string is generally
34177 not valid expression in the current language, and cannot be evaluated.
34179 For example, if @code{a} is an array, and variable object
34180 @code{A} was created for @code{a}, then we'll get this output:
34183 (gdb) -var-info-expression A.1
34184 ^done,lang="C",exp="1"
34188 Here, the value of @code{lang} is the language name, which can be
34189 found in @ref{Supported Languages}.
34191 Note that the output of the @code{-var-list-children} command also
34192 includes those expressions, so the @code{-var-info-expression} command
34195 @subheading The @code{-var-info-path-expression} Command
34196 @findex -var-info-path-expression
34198 @subsubheading Synopsis
34201 -var-info-path-expression @var{name}
34204 Returns an expression that can be evaluated in the current
34205 context and will yield the same value that a variable object has.
34206 Compare this with the @code{-var-info-expression} command, which
34207 result can be used only for UI presentation. Typical use of
34208 the @code{-var-info-path-expression} command is creating a
34209 watchpoint from a variable object.
34211 This command is currently not valid for children of a dynamic varobj,
34212 and will give an error when invoked on one.
34214 For example, suppose @code{C} is a C@t{++} class, derived from class
34215 @code{Base}, and that the @code{Base} class has a member called
34216 @code{m_size}. Assume a variable @code{c} is has the type of
34217 @code{C} and a variable object @code{C} was created for variable
34218 @code{c}. Then, we'll get this output:
34220 (gdb) -var-info-path-expression C.Base.public.m_size
34221 ^done,path_expr=((Base)c).m_size)
34224 @subheading The @code{-var-show-attributes} Command
34225 @findex -var-show-attributes
34227 @subsubheading Synopsis
34230 -var-show-attributes @var{name}
34233 List attributes of the specified variable object @var{name}:
34236 status=@var{attr} [ ( ,@var{attr} )* ]
34240 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
34242 @subheading The @code{-var-evaluate-expression} Command
34243 @findex -var-evaluate-expression
34245 @subsubheading Synopsis
34248 -var-evaluate-expression [-f @var{format-spec}] @var{name}
34251 Evaluates the expression that is represented by the specified variable
34252 object and returns its value as a string. The format of the string
34253 can be specified with the @samp{-f} option. The possible values of
34254 this option are the same as for @code{-var-set-format}
34255 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
34256 the current display format will be used. The current display format
34257 can be changed using the @code{-var-set-format} command.
34263 Note that one must invoke @code{-var-list-children} for a variable
34264 before the value of a child variable can be evaluated.
34266 @subheading The @code{-var-assign} Command
34267 @findex -var-assign
34269 @subsubheading Synopsis
34272 -var-assign @var{name} @var{expression}
34275 Assigns the value of @var{expression} to the variable object specified
34276 by @var{name}. The object must be @samp{editable}. If the variable's
34277 value is altered by the assign, the variable will show up in any
34278 subsequent @code{-var-update} list.
34280 @subsubheading Example
34288 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
34292 @subheading The @code{-var-update} Command
34293 @findex -var-update
34295 @subsubheading Synopsis
34298 -var-update [@var{print-values}] @{@var{name} | "*"@}
34301 Reevaluate the expressions corresponding to the variable object
34302 @var{name} and all its direct and indirect children, and return the
34303 list of variable objects whose values have changed; @var{name} must
34304 be a root variable object. Here, ``changed'' means that the result of
34305 @code{-var-evaluate-expression} before and after the
34306 @code{-var-update} is different. If @samp{*} is used as the variable
34307 object names, all existing variable objects are updated, except
34308 for frozen ones (@pxref{-var-set-frozen}). The option
34309 @var{print-values} determines whether both names and values, or just
34310 names are printed. The possible values of this option are the same
34311 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
34312 recommended to use the @samp{--all-values} option, to reduce the
34313 number of MI commands needed on each program stop.
34315 With the @samp{*} parameter, if a variable object is bound to a
34316 currently running thread, it will not be updated, without any
34319 If @code{-var-set-update-range} was previously used on a varobj, then
34320 only the selected range of children will be reported.
34322 @code{-var-update} reports all the changed varobjs in a tuple named
34325 Each item in the change list is itself a tuple holding:
34329 The name of the varobj.
34332 If values were requested for this update, then this field will be
34333 present and will hold the value of the varobj.
34336 @anchor{-var-update}
34337 This field is a string which may take one of three values:
34341 The variable object's current value is valid.
34344 The variable object does not currently hold a valid value but it may
34345 hold one in the future if its associated expression comes back into
34349 The variable object no longer holds a valid value.
34350 This can occur when the executable file being debugged has changed,
34351 either through recompilation or by using the @value{GDBN} @code{file}
34352 command. The front end should normally choose to delete these variable
34356 In the future new values may be added to this list so the front should
34357 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
34360 This is only present if the varobj is still valid. If the type
34361 changed, then this will be the string @samp{true}; otherwise it will
34364 When a varobj's type changes, its children are also likely to have
34365 become incorrect. Therefore, the varobj's children are automatically
34366 deleted when this attribute is @samp{true}. Also, the varobj's update
34367 range, when set using the @code{-var-set-update-range} command, is
34371 If the varobj's type changed, then this field will be present and will
34374 @item new_num_children
34375 For a dynamic varobj, if the number of children changed, or if the
34376 type changed, this will be the new number of children.
34378 The @samp{numchild} field in other varobj responses is generally not
34379 valid for a dynamic varobj -- it will show the number of children that
34380 @value{GDBN} knows about, but because dynamic varobjs lazily
34381 instantiate their children, this will not reflect the number of
34382 children which may be available.
34384 The @samp{new_num_children} attribute only reports changes to the
34385 number of children known by @value{GDBN}. This is the only way to
34386 detect whether an update has removed children (which necessarily can
34387 only happen at the end of the update range).
34390 The display hint, if any.
34393 This is an integer value, which will be 1 if there are more children
34394 available outside the varobj's update range.
34397 This attribute will be present and have the value @samp{1} if the
34398 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
34399 then this attribute will not be present.
34402 If new children were added to a dynamic varobj within the selected
34403 update range (as set by @code{-var-set-update-range}), then they will
34404 be listed in this attribute.
34407 @subsubheading Example
34414 -var-update --all-values var1
34415 ^done,changelist=[@{name="var1",value="3",in_scope="true",
34416 type_changed="false"@}]
34420 @subheading The @code{-var-set-frozen} Command
34421 @findex -var-set-frozen
34422 @anchor{-var-set-frozen}
34424 @subsubheading Synopsis
34427 -var-set-frozen @var{name} @var{flag}
34430 Set the frozenness flag on the variable object @var{name}. The
34431 @var{flag} parameter should be either @samp{1} to make the variable
34432 frozen or @samp{0} to make it unfrozen. If a variable object is
34433 frozen, then neither itself, nor any of its children, are
34434 implicitly updated by @code{-var-update} of
34435 a parent variable or by @code{-var-update *}. Only
34436 @code{-var-update} of the variable itself will update its value and
34437 values of its children. After a variable object is unfrozen, it is
34438 implicitly updated by all subsequent @code{-var-update} operations.
34439 Unfreezing a variable does not update it, only subsequent
34440 @code{-var-update} does.
34442 @subsubheading Example
34446 -var-set-frozen V 1
34451 @subheading The @code{-var-set-update-range} command
34452 @findex -var-set-update-range
34453 @anchor{-var-set-update-range}
34455 @subsubheading Synopsis
34458 -var-set-update-range @var{name} @var{from} @var{to}
34461 Set the range of children to be returned by future invocations of
34462 @code{-var-update}.
34464 @var{from} and @var{to} indicate the range of children to report. If
34465 @var{from} or @var{to} is less than zero, the range is reset and all
34466 children will be reported. Otherwise, children starting at @var{from}
34467 (zero-based) and up to and excluding @var{to} will be reported.
34469 @subsubheading Example
34473 -var-set-update-range V 1 2
34477 @subheading The @code{-var-set-visualizer} command
34478 @findex -var-set-visualizer
34479 @anchor{-var-set-visualizer}
34481 @subsubheading Synopsis
34484 -var-set-visualizer @var{name} @var{visualizer}
34487 Set a visualizer for the variable object @var{name}.
34489 @var{visualizer} is the visualizer to use. The special value
34490 @samp{None} means to disable any visualizer in use.
34492 If not @samp{None}, @var{visualizer} must be a Python expression.
34493 This expression must evaluate to a callable object which accepts a
34494 single argument. @value{GDBN} will call this object with the value of
34495 the varobj @var{name} as an argument (this is done so that the same
34496 Python pretty-printing code can be used for both the CLI and MI).
34497 When called, this object must return an object which conforms to the
34498 pretty-printing interface (@pxref{Pretty Printing API}).
34500 The pre-defined function @code{gdb.default_visualizer} may be used to
34501 select a visualizer by following the built-in process
34502 (@pxref{Selecting Pretty-Printers}). This is done automatically when
34503 a varobj is created, and so ordinarily is not needed.
34505 This feature is only available if Python support is enabled. The MI
34506 command @code{-list-features} (@pxref{GDB/MI Support Commands})
34507 can be used to check this.
34509 @subsubheading Example
34511 Resetting the visualizer:
34515 -var-set-visualizer V None
34519 Reselecting the default (type-based) visualizer:
34523 -var-set-visualizer V gdb.default_visualizer
34527 Suppose @code{SomeClass} is a visualizer class. A lambda expression
34528 can be used to instantiate this class for a varobj:
34532 -var-set-visualizer V "lambda val: SomeClass()"
34536 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34537 @node GDB/MI Data Manipulation
34538 @section @sc{gdb/mi} Data Manipulation
34540 @cindex data manipulation, in @sc{gdb/mi}
34541 @cindex @sc{gdb/mi}, data manipulation
34542 This section describes the @sc{gdb/mi} commands that manipulate data:
34543 examine memory and registers, evaluate expressions, etc.
34545 For details about what an addressable memory unit is,
34546 @pxref{addressable memory unit}.
34548 @c REMOVED FROM THE INTERFACE.
34549 @c @subheading -data-assign
34550 @c Change the value of a program variable. Plenty of side effects.
34551 @c @subsubheading GDB Command
34553 @c @subsubheading Example
34556 @subheading The @code{-data-disassemble} Command
34557 @findex -data-disassemble
34559 @subsubheading Synopsis
34563 [ -s @var{start-addr} -e @var{end-addr} ]
34564 | [ -a @var{addr} ]
34565 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
34573 @item @var{start-addr}
34574 is the beginning address (or @code{$pc})
34575 @item @var{end-addr}
34578 is an address anywhere within (or the name of) the function to
34579 disassemble. If an address is specified, the whole function
34580 surrounding that address will be disassembled. If a name is
34581 specified, the whole function with that name will be disassembled.
34582 @item @var{filename}
34583 is the name of the file to disassemble
34584 @item @var{linenum}
34585 is the line number to disassemble around
34587 is the number of disassembly lines to be produced. If it is -1,
34588 the whole function will be disassembled, in case no @var{end-addr} is
34589 specified. If @var{end-addr} is specified as a non-zero value, and
34590 @var{lines} is lower than the number of disassembly lines between
34591 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
34592 displayed; if @var{lines} is higher than the number of lines between
34593 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
34598 @item 0 disassembly only
34599 @item 1 mixed source and disassembly (deprecated)
34600 @item 2 disassembly with raw opcodes
34601 @item 3 mixed source and disassembly with raw opcodes (deprecated)
34602 @item 4 mixed source and disassembly
34603 @item 5 mixed source and disassembly with raw opcodes
34606 Modes 1 and 3 are deprecated. The output is ``source centric''
34607 which hasn't proved useful in practice.
34608 @xref{Machine Code}, for a discussion of the difference between
34609 @code{/m} and @code{/s} output of the @code{disassemble} command.
34612 @subsubheading Result
34614 The result of the @code{-data-disassemble} command will be a list named
34615 @samp{asm_insns}, the contents of this list depend on the @var{mode}
34616 used with the @code{-data-disassemble} command.
34618 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
34623 The address at which this instruction was disassembled.
34626 The name of the function this instruction is within.
34629 The decimal offset in bytes from the start of @samp{func-name}.
34632 The text disassembly for this @samp{address}.
34635 This field is only present for modes 2, 3 and 5. This contains the raw opcode
34636 bytes for the @samp{inst} field.
34640 For modes 1, 3, 4 and 5 the @samp{asm_insns} list contains tuples named
34641 @samp{src_and_asm_line}, each of which has the following fields:
34645 The line number within @samp{file}.
34648 The file name from the compilation unit. This might be an absolute
34649 file name or a relative file name depending on the compile command
34653 Absolute file name of @samp{file}. It is converted to a canonical form
34654 using the source file search path
34655 (@pxref{Source Path, ,Specifying Source Directories})
34656 and after resolving all the symbolic links.
34658 If the source file is not found this field will contain the path as
34659 present in the debug information.
34661 @item line_asm_insn
34662 This is a list of tuples containing the disassembly for @samp{line} in
34663 @samp{file}. The fields of each tuple are the same as for
34664 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
34665 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
34670 Note that whatever included in the @samp{inst} field, is not
34671 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
34674 @subsubheading @value{GDBN} Command
34676 The corresponding @value{GDBN} command is @samp{disassemble}.
34678 @subsubheading Example
34680 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
34684 -data-disassemble -s $pc -e "$pc + 20" -- 0
34687 @{address="0x000107c0",func-name="main",offset="4",
34688 inst="mov 2, %o0"@},
34689 @{address="0x000107c4",func-name="main",offset="8",
34690 inst="sethi %hi(0x11800), %o2"@},
34691 @{address="0x000107c8",func-name="main",offset="12",
34692 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
34693 @{address="0x000107cc",func-name="main",offset="16",
34694 inst="sethi %hi(0x11800), %o2"@},
34695 @{address="0x000107d0",func-name="main",offset="20",
34696 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
34700 Disassemble the whole @code{main} function. Line 32 is part of
34704 -data-disassemble -f basics.c -l 32 -- 0
34706 @{address="0x000107bc",func-name="main",offset="0",
34707 inst="save %sp, -112, %sp"@},
34708 @{address="0x000107c0",func-name="main",offset="4",
34709 inst="mov 2, %o0"@},
34710 @{address="0x000107c4",func-name="main",offset="8",
34711 inst="sethi %hi(0x11800), %o2"@},
34713 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
34714 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
34718 Disassemble 3 instructions from the start of @code{main}:
34722 -data-disassemble -f basics.c -l 32 -n 3 -- 0
34724 @{address="0x000107bc",func-name="main",offset="0",
34725 inst="save %sp, -112, %sp"@},
34726 @{address="0x000107c0",func-name="main",offset="4",
34727 inst="mov 2, %o0"@},
34728 @{address="0x000107c4",func-name="main",offset="8",
34729 inst="sethi %hi(0x11800), %o2"@}]
34733 Disassemble 3 instructions from the start of @code{main} in mixed mode:
34737 -data-disassemble -f basics.c -l 32 -n 3 -- 1
34739 src_and_asm_line=@{line="31",
34740 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
34741 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
34742 line_asm_insn=[@{address="0x000107bc",
34743 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
34744 src_and_asm_line=@{line="32",
34745 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
34746 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
34747 line_asm_insn=[@{address="0x000107c0",
34748 func-name="main",offset="4",inst="mov 2, %o0"@},
34749 @{address="0x000107c4",func-name="main",offset="8",
34750 inst="sethi %hi(0x11800), %o2"@}]@}]
34755 @subheading The @code{-data-evaluate-expression} Command
34756 @findex -data-evaluate-expression
34758 @subsubheading Synopsis
34761 -data-evaluate-expression @var{expr}
34764 Evaluate @var{expr} as an expression. The expression could contain an
34765 inferior function call. The function call will execute synchronously.
34766 If the expression contains spaces, it must be enclosed in double quotes.
34768 @subsubheading @value{GDBN} Command
34770 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
34771 @samp{call}. In @code{gdbtk} only, there's a corresponding
34772 @samp{gdb_eval} command.
34774 @subsubheading Example
34776 In the following example, the numbers that precede the commands are the
34777 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
34778 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
34782 211-data-evaluate-expression A
34785 311-data-evaluate-expression &A
34786 311^done,value="0xefffeb7c"
34788 411-data-evaluate-expression A+3
34791 511-data-evaluate-expression "A + 3"
34797 @subheading The @code{-data-list-changed-registers} Command
34798 @findex -data-list-changed-registers
34800 @subsubheading Synopsis
34803 -data-list-changed-registers
34806 Display a list of the registers that have changed.
34808 @subsubheading @value{GDBN} Command
34810 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
34811 has the corresponding command @samp{gdb_changed_register_list}.
34813 @subsubheading Example
34815 On a PPC MBX board:
34823 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
34824 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
34825 line="5",arch="powerpc"@}
34827 -data-list-changed-registers
34828 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
34829 "10","11","13","14","15","16","17","18","19","20","21","22","23",
34830 "24","25","26","27","28","30","31","64","65","66","67","69"]
34835 @subheading The @code{-data-list-register-names} Command
34836 @findex -data-list-register-names
34838 @subsubheading Synopsis
34841 -data-list-register-names [ ( @var{regno} )+ ]
34844 Show a list of register names for the current target. If no arguments
34845 are given, it shows a list of the names of all the registers. If
34846 integer numbers are given as arguments, it will print a list of the
34847 names of the registers corresponding to the arguments. To ensure
34848 consistency between a register name and its number, the output list may
34849 include empty register names.
34851 @subsubheading @value{GDBN} Command
34853 @value{GDBN} does not have a command which corresponds to
34854 @samp{-data-list-register-names}. In @code{gdbtk} there is a
34855 corresponding command @samp{gdb_regnames}.
34857 @subsubheading Example
34859 For the PPC MBX board:
34862 -data-list-register-names
34863 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
34864 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
34865 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
34866 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
34867 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
34868 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
34869 "", "pc","ps","cr","lr","ctr","xer"]
34871 -data-list-register-names 1 2 3
34872 ^done,register-names=["r1","r2","r3"]
34876 @subheading The @code{-data-list-register-values} Command
34877 @findex -data-list-register-values
34879 @subsubheading Synopsis
34882 -data-list-register-values
34883 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
34886 Display the registers' contents. The format according to which the
34887 registers' contents are to be returned is given by @var{fmt}, followed
34888 by an optional list of numbers specifying the registers to display. A
34889 missing list of numbers indicates that the contents of all the
34890 registers must be returned. The @code{--skip-unavailable} option
34891 indicates that only the available registers are to be returned.
34893 Allowed formats for @var{fmt} are:
34910 @subsubheading @value{GDBN} Command
34912 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
34913 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
34915 @subsubheading Example
34917 For a PPC MBX board (note: line breaks are for readability only, they
34918 don't appear in the actual output):
34922 -data-list-register-values r 64 65
34923 ^done,register-values=[@{number="64",value="0xfe00a300"@},
34924 @{number="65",value="0x00029002"@}]
34926 -data-list-register-values x
34927 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
34928 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
34929 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
34930 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
34931 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
34932 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
34933 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
34934 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
34935 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
34936 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
34937 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
34938 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
34939 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
34940 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
34941 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
34942 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
34943 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
34944 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
34945 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
34946 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
34947 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
34948 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
34949 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
34950 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
34951 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
34952 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
34953 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
34954 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
34955 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
34956 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
34957 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
34958 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
34959 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
34960 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
34961 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
34962 @{number="69",value="0x20002b03"@}]
34967 @subheading The @code{-data-read-memory} Command
34968 @findex -data-read-memory
34970 This command is deprecated, use @code{-data-read-memory-bytes} instead.
34972 @subsubheading Synopsis
34975 -data-read-memory [ -o @var{byte-offset} ]
34976 @var{address} @var{word-format} @var{word-size}
34977 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
34984 @item @var{address}
34985 An expression specifying the address of the first memory word to be
34986 read. Complex expressions containing embedded white space should be
34987 quoted using the C convention.
34989 @item @var{word-format}
34990 The format to be used to print the memory words. The notation is the
34991 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
34994 @item @var{word-size}
34995 The size of each memory word in bytes.
34997 @item @var{nr-rows}
34998 The number of rows in the output table.
35000 @item @var{nr-cols}
35001 The number of columns in the output table.
35004 If present, indicates that each row should include an @sc{ascii} dump. The
35005 value of @var{aschar} is used as a padding character when a byte is not a
35006 member of the printable @sc{ascii} character set (printable @sc{ascii}
35007 characters are those whose code is between 32 and 126, inclusively).
35009 @item @var{byte-offset}
35010 An offset to add to the @var{address} before fetching memory.
35013 This command displays memory contents as a table of @var{nr-rows} by
35014 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
35015 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
35016 (returned as @samp{total-bytes}). Should less than the requested number
35017 of bytes be returned by the target, the missing words are identified
35018 using @samp{N/A}. The number of bytes read from the target is returned
35019 in @samp{nr-bytes} and the starting address used to read memory in
35022 The address of the next/previous row or page is available in
35023 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
35026 @subsubheading @value{GDBN} Command
35028 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
35029 @samp{gdb_get_mem} memory read command.
35031 @subsubheading Example
35033 Read six bytes of memory starting at @code{bytes+6} but then offset by
35034 @code{-6} bytes. Format as three rows of two columns. One byte per
35035 word. Display each word in hex.
35039 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
35040 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
35041 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
35042 prev-page="0x0000138a",memory=[
35043 @{addr="0x00001390",data=["0x00","0x01"]@},
35044 @{addr="0x00001392",data=["0x02","0x03"]@},
35045 @{addr="0x00001394",data=["0x04","0x05"]@}]
35049 Read two bytes of memory starting at address @code{shorts + 64} and
35050 display as a single word formatted in decimal.
35054 5-data-read-memory shorts+64 d 2 1 1
35055 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
35056 next-row="0x00001512",prev-row="0x0000150e",
35057 next-page="0x00001512",prev-page="0x0000150e",memory=[
35058 @{addr="0x00001510",data=["128"]@}]
35062 Read thirty two bytes of memory starting at @code{bytes+16} and format
35063 as eight rows of four columns. Include a string encoding with @samp{x}
35064 used as the non-printable character.
35068 4-data-read-memory bytes+16 x 1 8 4 x
35069 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
35070 next-row="0x000013c0",prev-row="0x0000139c",
35071 next-page="0x000013c0",prev-page="0x00001380",memory=[
35072 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
35073 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
35074 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
35075 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
35076 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
35077 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
35078 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
35079 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
35083 @subheading The @code{-data-read-memory-bytes} Command
35084 @findex -data-read-memory-bytes
35086 @subsubheading Synopsis
35089 -data-read-memory-bytes [ -o @var{offset} ]
35090 @var{address} @var{count}
35097 @item @var{address}
35098 An expression specifying the address of the first addressable memory unit
35099 to be read. Complex expressions containing embedded white space should be
35100 quoted using the C convention.
35103 The number of addressable memory units to read. This should be an integer
35107 The offset relative to @var{address} at which to start reading. This
35108 should be an integer literal. This option is provided so that a frontend
35109 is not required to first evaluate address and then perform address
35110 arithmetics itself.
35114 This command attempts to read all accessible memory regions in the
35115 specified range. First, all regions marked as unreadable in the memory
35116 map (if one is defined) will be skipped. @xref{Memory Region
35117 Attributes}. Second, @value{GDBN} will attempt to read the remaining
35118 regions. For each one, if reading full region results in an errors,
35119 @value{GDBN} will try to read a subset of the region.
35121 In general, every single memory unit in the region may be readable or not,
35122 and the only way to read every readable unit is to try a read at
35123 every address, which is not practical. Therefore, @value{GDBN} will
35124 attempt to read all accessible memory units at either beginning or the end
35125 of the region, using a binary division scheme. This heuristic works
35126 well for reading across a memory map boundary. Note that if a region
35127 has a readable range that is neither at the beginning or the end,
35128 @value{GDBN} will not read it.
35130 The result record (@pxref{GDB/MI Result Records}) that is output of
35131 the command includes a field named @samp{memory} whose content is a
35132 list of tuples. Each tuple represent a successfully read memory block
35133 and has the following fields:
35137 The start address of the memory block, as hexadecimal literal.
35140 The end address of the memory block, as hexadecimal literal.
35143 The offset of the memory block, as hexadecimal literal, relative to
35144 the start address passed to @code{-data-read-memory-bytes}.
35147 The contents of the memory block, in hex.
35153 @subsubheading @value{GDBN} Command
35155 The corresponding @value{GDBN} command is @samp{x}.
35157 @subsubheading Example
35161 -data-read-memory-bytes &a 10
35162 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
35164 contents="01000000020000000300"@}]
35169 @subheading The @code{-data-write-memory-bytes} Command
35170 @findex -data-write-memory-bytes
35172 @subsubheading Synopsis
35175 -data-write-memory-bytes @var{address} @var{contents}
35176 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
35183 @item @var{address}
35184 An expression specifying the address of the first addressable memory unit
35185 to be written. Complex expressions containing embedded white space should
35186 be quoted using the C convention.
35188 @item @var{contents}
35189 The hex-encoded data to write. It is an error if @var{contents} does
35190 not represent an integral number of addressable memory units.
35193 Optional argument indicating the number of addressable memory units to be
35194 written. If @var{count} is greater than @var{contents}' length,
35195 @value{GDBN} will repeatedly write @var{contents} until it fills
35196 @var{count} memory units.
35200 @subsubheading @value{GDBN} Command
35202 There's no corresponding @value{GDBN} command.
35204 @subsubheading Example
35208 -data-write-memory-bytes &a "aabbccdd"
35215 -data-write-memory-bytes &a "aabbccdd" 16e
35220 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
35221 @node GDB/MI Tracepoint Commands
35222 @section @sc{gdb/mi} Tracepoint Commands
35224 The commands defined in this section implement MI support for
35225 tracepoints. For detailed introduction, see @ref{Tracepoints}.
35227 @subheading The @code{-trace-find} Command
35228 @findex -trace-find
35230 @subsubheading Synopsis
35233 -trace-find @var{mode} [@var{parameters}@dots{}]
35236 Find a trace frame using criteria defined by @var{mode} and
35237 @var{parameters}. The following table lists permissible
35238 modes and their parameters. For details of operation, see @ref{tfind}.
35243 No parameters are required. Stops examining trace frames.
35246 An integer is required as parameter. Selects tracepoint frame with
35249 @item tracepoint-number
35250 An integer is required as parameter. Finds next
35251 trace frame that corresponds to tracepoint with the specified number.
35254 An address is required as parameter. Finds
35255 next trace frame that corresponds to any tracepoint at the specified
35258 @item pc-inside-range
35259 Two addresses are required as parameters. Finds next trace
35260 frame that corresponds to a tracepoint at an address inside the
35261 specified range. Both bounds are considered to be inside the range.
35263 @item pc-outside-range
35264 Two addresses are required as parameters. Finds
35265 next trace frame that corresponds to a tracepoint at an address outside
35266 the specified range. Both bounds are considered to be inside the range.
35269 Line specification is required as parameter. @xref{Specify Location}.
35270 Finds next trace frame that corresponds to a tracepoint at
35271 the specified location.
35275 If @samp{none} was passed as @var{mode}, the response does not
35276 have fields. Otherwise, the response may have the following fields:
35280 This field has either @samp{0} or @samp{1} as the value, depending
35281 on whether a matching tracepoint was found.
35284 The index of the found traceframe. This field is present iff
35285 the @samp{found} field has value of @samp{1}.
35288 The index of the found tracepoint. This field is present iff
35289 the @samp{found} field has value of @samp{1}.
35292 The information about the frame corresponding to the found trace
35293 frame. This field is present only if a trace frame was found.
35294 @xref{GDB/MI Frame Information}, for description of this field.
35298 @subsubheading @value{GDBN} Command
35300 The corresponding @value{GDBN} command is @samp{tfind}.
35302 @subheading -trace-define-variable
35303 @findex -trace-define-variable
35305 @subsubheading Synopsis
35308 -trace-define-variable @var{name} [ @var{value} ]
35311 Create trace variable @var{name} if it does not exist. If
35312 @var{value} is specified, sets the initial value of the specified
35313 trace variable to that value. Note that the @var{name} should start
35314 with the @samp{$} character.
35316 @subsubheading @value{GDBN} Command
35318 The corresponding @value{GDBN} command is @samp{tvariable}.
35320 @subheading The @code{-trace-frame-collected} Command
35321 @findex -trace-frame-collected
35323 @subsubheading Synopsis
35326 -trace-frame-collected
35327 [--var-print-values @var{var_pval}]
35328 [--comp-print-values @var{comp_pval}]
35329 [--registers-format @var{regformat}]
35330 [--memory-contents]
35333 This command returns the set of collected objects, register names,
35334 trace state variable names, memory ranges and computed expressions
35335 that have been collected at a particular trace frame. The optional
35336 parameters to the command affect the output format in different ways.
35337 See the output description table below for more details.
35339 The reported names can be used in the normal manner to create
35340 varobjs and inspect the objects themselves. The items returned by
35341 this command are categorized so that it is clear which is a variable,
35342 which is a register, which is a trace state variable, which is a
35343 memory range and which is a computed expression.
35345 For instance, if the actions were
35347 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
35348 collect *(int*)0xaf02bef0@@40
35352 the object collected in its entirety would be @code{myVar}. The
35353 object @code{myArray} would be partially collected, because only the
35354 element at index @code{myIndex} would be collected. The remaining
35355 objects would be computed expressions.
35357 An example output would be:
35361 -trace-frame-collected
35363 explicit-variables=[@{name="myVar",value="1"@}],
35364 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
35365 @{name="myObj.field",value="0"@},
35366 @{name="myPtr->field",value="1"@},
35367 @{name="myCount + 2",value="3"@},
35368 @{name="$tvar1 + 1",value="43970027"@}],
35369 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
35370 @{number="1",value="0x0"@},
35371 @{number="2",value="0x4"@},
35373 @{number="125",value="0x0"@}],
35374 tvars=[@{name="$tvar1",current="43970026"@}],
35375 memory=[@{address="0x0000000000602264",length="4"@},
35376 @{address="0x0000000000615bc0",length="4"@}]
35383 @item explicit-variables
35384 The set of objects that have been collected in their entirety (as
35385 opposed to collecting just a few elements of an array or a few struct
35386 members). For each object, its name and value are printed.
35387 The @code{--var-print-values} option affects how or whether the value
35388 field is output. If @var{var_pval} is 0, then print only the names;
35389 if it is 1, print also their values; and if it is 2, print the name,
35390 type and value for simple data types, and the name and type for
35391 arrays, structures and unions.
35393 @item computed-expressions
35394 The set of computed expressions that have been collected at the
35395 current trace frame. The @code{--comp-print-values} option affects
35396 this set like the @code{--var-print-values} option affects the
35397 @code{explicit-variables} set. See above.
35400 The registers that have been collected at the current trace frame.
35401 For each register collected, the name and current value are returned.
35402 The value is formatted according to the @code{--registers-format}
35403 option. See the @command{-data-list-register-values} command for a
35404 list of the allowed formats. The default is @samp{x}.
35407 The trace state variables that have been collected at the current
35408 trace frame. For each trace state variable collected, the name and
35409 current value are returned.
35412 The set of memory ranges that have been collected at the current trace
35413 frame. Its content is a list of tuples. Each tuple represents a
35414 collected memory range and has the following fields:
35418 The start address of the memory range, as hexadecimal literal.
35421 The length of the memory range, as decimal literal.
35424 The contents of the memory block, in hex. This field is only present
35425 if the @code{--memory-contents} option is specified.
35431 @subsubheading @value{GDBN} Command
35433 There is no corresponding @value{GDBN} command.
35435 @subsubheading Example
35437 @subheading -trace-list-variables
35438 @findex -trace-list-variables
35440 @subsubheading Synopsis
35443 -trace-list-variables
35446 Return a table of all defined trace variables. Each element of the
35447 table has the following fields:
35451 The name of the trace variable. This field is always present.
35454 The initial value. This is a 64-bit signed integer. This
35455 field is always present.
35458 The value the trace variable has at the moment. This is a 64-bit
35459 signed integer. This field is absent iff current value is
35460 not defined, for example if the trace was never run, or is
35465 @subsubheading @value{GDBN} Command
35467 The corresponding @value{GDBN} command is @samp{tvariables}.
35469 @subsubheading Example
35473 -trace-list-variables
35474 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
35475 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
35476 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
35477 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
35478 body=[variable=@{name="$trace_timestamp",initial="0"@}
35479 variable=@{name="$foo",initial="10",current="15"@}]@}
35483 @subheading -trace-save
35484 @findex -trace-save
35486 @subsubheading Synopsis
35489 -trace-save [ -r ] [ -ctf ] @var{filename}
35492 Saves the collected trace data to @var{filename}. Without the
35493 @samp{-r} option, the data is downloaded from the target and saved
35494 in a local file. With the @samp{-r} option the target is asked
35495 to perform the save.
35497 By default, this command will save the trace in the tfile format. You can
35498 supply the optional @samp{-ctf} argument to save it the CTF format. See
35499 @ref{Trace Files} for more information about CTF.
35501 @subsubheading @value{GDBN} Command
35503 The corresponding @value{GDBN} command is @samp{tsave}.
35506 @subheading -trace-start
35507 @findex -trace-start
35509 @subsubheading Synopsis
35515 Starts a tracing experiment. The result of this command does not
35518 @subsubheading @value{GDBN} Command
35520 The corresponding @value{GDBN} command is @samp{tstart}.
35522 @subheading -trace-status
35523 @findex -trace-status
35525 @subsubheading Synopsis
35531 Obtains the status of a tracing experiment. The result may include
35532 the following fields:
35537 May have a value of either @samp{0}, when no tracing operations are
35538 supported, @samp{1}, when all tracing operations are supported, or
35539 @samp{file} when examining trace file. In the latter case, examining
35540 of trace frame is possible but new tracing experiement cannot be
35541 started. This field is always present.
35544 May have a value of either @samp{0} or @samp{1} depending on whether
35545 tracing experiement is in progress on target. This field is present
35546 if @samp{supported} field is not @samp{0}.
35549 Report the reason why the tracing was stopped last time. This field
35550 may be absent iff tracing was never stopped on target yet. The
35551 value of @samp{request} means the tracing was stopped as result of
35552 the @code{-trace-stop} command. The value of @samp{overflow} means
35553 the tracing buffer is full. The value of @samp{disconnection} means
35554 tracing was automatically stopped when @value{GDBN} has disconnected.
35555 The value of @samp{passcount} means tracing was stopped when a
35556 tracepoint was passed a maximal number of times for that tracepoint.
35557 This field is present if @samp{supported} field is not @samp{0}.
35559 @item stopping-tracepoint
35560 The number of tracepoint whose passcount as exceeded. This field is
35561 present iff the @samp{stop-reason} field has the value of
35565 @itemx frames-created
35566 The @samp{frames} field is a count of the total number of trace frames
35567 in the trace buffer, while @samp{frames-created} is the total created
35568 during the run, including ones that were discarded, such as when a
35569 circular trace buffer filled up. Both fields are optional.
35573 These fields tell the current size of the tracing buffer and the
35574 remaining space. These fields are optional.
35577 The value of the circular trace buffer flag. @code{1} means that the
35578 trace buffer is circular and old trace frames will be discarded if
35579 necessary to make room, @code{0} means that the trace buffer is linear
35583 The value of the disconnected tracing flag. @code{1} means that
35584 tracing will continue after @value{GDBN} disconnects, @code{0} means
35585 that the trace run will stop.
35588 The filename of the trace file being examined. This field is
35589 optional, and only present when examining a trace file.
35593 @subsubheading @value{GDBN} Command
35595 The corresponding @value{GDBN} command is @samp{tstatus}.
35597 @subheading -trace-stop
35598 @findex -trace-stop
35600 @subsubheading Synopsis
35606 Stops a tracing experiment. The result of this command has the same
35607 fields as @code{-trace-status}, except that the @samp{supported} and
35608 @samp{running} fields are not output.
35610 @subsubheading @value{GDBN} Command
35612 The corresponding @value{GDBN} command is @samp{tstop}.
35615 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
35616 @node GDB/MI Symbol Query
35617 @section @sc{gdb/mi} Symbol Query Commands
35621 @subheading The @code{-symbol-info-address} Command
35622 @findex -symbol-info-address
35624 @subsubheading Synopsis
35627 -symbol-info-address @var{symbol}
35630 Describe where @var{symbol} is stored.
35632 @subsubheading @value{GDBN} Command
35634 The corresponding @value{GDBN} command is @samp{info address}.
35636 @subsubheading Example
35640 @subheading The @code{-symbol-info-file} Command
35641 @findex -symbol-info-file
35643 @subsubheading Synopsis
35649 Show the file for the symbol.
35651 @subsubheading @value{GDBN} Command
35653 There's no equivalent @value{GDBN} command. @code{gdbtk} has
35654 @samp{gdb_find_file}.
35656 @subsubheading Example
35660 @subheading The @code{-symbol-info-functions} Command
35661 @findex -symbol-info-functions
35662 @anchor{-symbol-info-functions}
35664 @subsubheading Synopsis
35667 -symbol-info-functions [--include-nondebug]
35668 [--type @var{type_regexp}]
35669 [--name @var{name_regexp}]
35670 [--max-results @var{limit}]
35674 Return a list containing the names and types for all global functions
35675 taken from the debug information. The functions are grouped by source
35676 file, and shown with the line number on which each function is
35679 The @code{--include-nondebug} option causes the output to include
35680 code symbols from the symbol table.
35682 The options @code{--type} and @code{--name} allow the symbols returned
35683 to be filtered based on either the name of the function, or the type
35684 signature of the function.
35686 The option @code{--max-results} restricts the command to return no
35687 more than @var{limit} results. If exactly @var{limit} results are
35688 returned then there might be additional results available if a higher
35691 @subsubheading @value{GDBN} Command
35693 The corresponding @value{GDBN} command is @samp{info functions}.
35695 @subsubheading Example
35699 -symbol-info-functions
35702 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
35703 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
35704 symbols=[@{line="36", name="f4", type="void (int *)",
35705 description="void f4(int *);"@},
35706 @{line="42", name="main", type="int ()",
35707 description="int main();"@},
35708 @{line="30", name="f1", type="my_int_t (int, int)",
35709 description="static my_int_t f1(int, int);"@}]@},
35710 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
35711 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
35712 symbols=[@{line="33", name="f2", type="float (another_float_t)",
35713 description="float f2(another_float_t);"@},
35714 @{line="39", name="f3", type="int (another_int_t)",
35715 description="int f3(another_int_t);"@},
35716 @{line="27", name="f1", type="another_float_t (int)",
35717 description="static another_float_t f1(int);"@}]@}]@}
35721 -symbol-info-functions --name f1
35724 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
35725 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
35726 symbols=[@{line="30", name="f1", type="my_int_t (int, int)",
35727 description="static my_int_t f1(int, int);"@}]@},
35728 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
35729 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
35730 symbols=[@{line="27", name="f1", type="another_float_t (int)",
35731 description="static another_float_t f1(int);"@}]@}]@}
35735 -symbol-info-functions --type void
35738 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
35739 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
35740 symbols=[@{line="36", name="f4", type="void (int *)",
35741 description="void f4(int *);"@}]@}]@}
35745 -symbol-info-functions --include-nondebug
35748 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
35749 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
35750 symbols=[@{line="36", name="f4", type="void (int *)",
35751 description="void f4(int *);"@},
35752 @{line="42", name="main", type="int ()",
35753 description="int main();"@},
35754 @{line="30", name="f1", type="my_int_t (int, int)",
35755 description="static my_int_t f1(int, int);"@}]@},
35756 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
35757 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
35758 symbols=[@{line="33", name="f2", type="float (another_float_t)",
35759 description="float f2(another_float_t);"@},
35760 @{line="39", name="f3", type="int (another_int_t)",
35761 description="int f3(another_int_t);"@},
35762 @{line="27", name="f1", type="another_float_t (int)",
35763 description="static another_float_t f1(int);"@}]@}],
35765 [@{address="0x0000000000400398",name="_init"@},
35766 @{address="0x00000000004003b0",name="_start"@},
35772 @subheading The @code{-symbol-info-module-functions} Command
35773 @findex -symbol-info-module-functions
35774 @anchor{-symbol-info-module-functions}
35776 @subsubheading Synopsis
35779 -symbol-info-module-functions [--module @var{module_regexp}]
35780 [--name @var{name_regexp}]
35781 [--type @var{type_regexp}]
35785 Return a list containing the names of all known functions within all
35786 know Fortran modules. The functions are grouped by source file and
35787 containing module, and shown with the line number on which each
35788 function is defined.
35790 The option @code{--module} only returns results for modules matching
35791 @var{module_regexp}. The option @code{--name} only returns functions
35792 whose name matches @var{name_regexp}, and @code{--type} only returns
35793 functions whose type matches @var{type_regexp}.
35795 @subsubheading @value{GDBN} Command
35797 The corresponding @value{GDBN} command is @samp{info module functions}.
35799 @subsubheading Example
35804 -symbol-info-module-functions
35807 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
35808 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
35809 symbols=[@{line="21",name="mod1::check_all",type="void (void)",
35810 description="void mod1::check_all(void);"@}]@}]@},
35812 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
35813 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
35814 symbols=[@{line="30",name="mod2::check_var_i",type="void (void)",
35815 description="void mod2::check_var_i(void);"@}]@}]@},
35817 files=[@{filename="/projec/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35818 fullname="/projec/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35819 symbols=[@{line="21",name="mod3::check_all",type="void (void)",
35820 description="void mod3::check_all(void);"@},
35821 @{line="27",name="mod3::check_mod2",type="void (void)",
35822 description="void mod3::check_mod2(void);"@}]@}]@},
35823 @{module="modmany",
35824 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35825 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35826 symbols=[@{line="35",name="modmany::check_some",type="void (void)",
35827 description="void modmany::check_some(void);"@}]@}]@},
35829 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35830 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35831 symbols=[@{line="44",name="moduse::check_all",type="void (void)",
35832 description="void moduse::check_all(void);"@},
35833 @{line="49",name="moduse::check_var_x",type="void (void)",
35834 description="void moduse::check_var_x(void);"@}]@}]@}]
35838 @subheading The @code{-symbol-info-module-variables} Command
35839 @findex -symbol-info-module-variables
35840 @anchor{-symbol-info-module-variables}
35842 @subsubheading Synopsis
35845 -symbol-info-module-variables [--module @var{module_regexp}]
35846 [--name @var{name_regexp}]
35847 [--type @var{type_regexp}]
35851 Return a list containing the names of all known variables within all
35852 know Fortran modules. The variables are grouped by source file and
35853 containing module, and shown with the line number on which each
35854 variable is defined.
35856 The option @code{--module} only returns results for modules matching
35857 @var{module_regexp}. The option @code{--name} only returns variables
35858 whose name matches @var{name_regexp}, and @code{--type} only returns
35859 variables whose type matches @var{type_regexp}.
35861 @subsubheading @value{GDBN} Command
35863 The corresponding @value{GDBN} command is @samp{info module variables}.
35865 @subsubheading Example
35870 -symbol-info-module-variables
35873 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
35874 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
35875 symbols=[@{line="18",name="mod1::var_const",type="integer(kind=4)",
35876 description="integer(kind=4) mod1::var_const;"@},
35877 @{line="17",name="mod1::var_i",type="integer(kind=4)",
35878 description="integer(kind=4) mod1::var_i;"@}]@}]@},
35880 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
35881 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
35882 symbols=[@{line="28",name="mod2::var_i",type="integer(kind=4)",
35883 description="integer(kind=4) mod2::var_i;"@}]@}]@},
35885 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35886 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35887 symbols=[@{line="18",name="mod3::mod1",type="integer(kind=4)",
35888 description="integer(kind=4) mod3::mod1;"@},
35889 @{line="17",name="mod3::mod2",type="integer(kind=4)",
35890 description="integer(kind=4) mod3::mod2;"@},
35891 @{line="19",name="mod3::var_i",type="integer(kind=4)",
35892 description="integer(kind=4) mod3::var_i;"@}]@}]@},
35893 @{module="modmany",
35894 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35895 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35896 symbols=[@{line="33",name="modmany::var_a",type="integer(kind=4)",
35897 description="integer(kind=4) modmany::var_a;"@},
35898 @{line="33",name="modmany::var_b",type="integer(kind=4)",
35899 description="integer(kind=4) modmany::var_b;"@},
35900 @{line="33",name="modmany::var_c",type="integer(kind=4)",
35901 description="integer(kind=4) modmany::var_c;"@},
35902 @{line="33",name="modmany::var_i",type="integer(kind=4)",
35903 description="integer(kind=4) modmany::var_i;"@}]@}]@},
35905 files=[@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35906 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35907 symbols=[@{line="42",name="moduse::var_x",type="integer(kind=4)",
35908 description="integer(kind=4) moduse::var_x;"@},
35909 @{line="42",name="moduse::var_y",type="integer(kind=4)",
35910 description="integer(kind=4) moduse::var_y;"@}]@}]@}]
35914 @subheading The @code{-symbol-info-modules} Command
35915 @findex -symbol-info-modules
35916 @anchor{-symbol-info-modules}
35918 @subsubheading Synopsis
35921 -symbol-info-modules [--name @var{name_regexp}]
35922 [--max-results @var{limit}]
35927 Return a list containing the names of all known Fortran modules. The
35928 modules are grouped by source file, and shown with the line number on
35929 which each modules is defined.
35931 The option @code{--name} allows the modules returned to be filtered
35932 based the name of the module.
35934 The option @code{--max-results} restricts the command to return no
35935 more than @var{limit} results. If exactly @var{limit} results are
35936 returned then there might be additional results available if a higher
35939 @subsubheading @value{GDBN} Command
35941 The corresponding @value{GDBN} command is @samp{info modules}.
35943 @subsubheading Example
35947 -symbol-info-modules
35950 [@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
35951 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
35952 symbols=[@{line="16",name="mod1"@},
35953 @{line="22",name="mod2"@}]@},
35954 @{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35955 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35956 symbols=[@{line="16",name="mod3"@},
35957 @{line="22",name="modmany"@},
35958 @{line="26",name="moduse"@}]@}]@}
35962 -symbol-info-modules --name mod[123]
35965 [@{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
35966 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules-2.f90",
35967 symbols=[@{line="16",name="mod1"@},
35968 @{line="22",name="mod2"@}]@},
35969 @{filename="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35970 fullname="/project/gdb/testsuite/gdb.mi/mi-fortran-modules.f90",
35971 symbols=[@{line="16",name="mod3"@}]@}]@}
35975 @subheading The @code{-symbol-info-types} Command
35976 @findex -symbol-info-types
35977 @anchor{-symbol-info-types}
35979 @subsubheading Synopsis
35982 -symbol-info-types [--name @var{name_regexp}]
35983 [--max-results @var{limit}]
35988 Return a list of all defined types. The types are grouped by source
35989 file, and shown with the line number on which each user defined type
35990 is defined. Some base types are not defined in the source code but
35991 are added to the debug information by the compiler, for example
35992 @code{int}, @code{float}, etc.; these types do not have an associated
35995 The option @code{--name} allows the list of types returned to be
35998 The option @code{--max-results} restricts the command to return no
35999 more than @var{limit} results. If exactly @var{limit} results are
36000 returned then there might be additional results available if a higher
36003 @subsubheading @value{GDBN} Command
36005 The corresponding @value{GDBN} command is @samp{info types}.
36007 @subsubheading Example
36014 [@{filename="gdb.mi/mi-sym-info-1.c",
36015 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36016 symbols=[@{name="float"@},
36018 @{line="27",name="typedef int my_int_t;"@}]@},
36019 @{filename="gdb.mi/mi-sym-info-2.c",
36020 fullname="/project/gdb.mi/mi-sym-info-2.c",
36021 symbols=[@{line="24",name="typedef float another_float_t;"@},
36022 @{line="23",name="typedef int another_int_t;"@},
36024 @{name="int"@}]@}]@}
36028 -symbol-info-types --name _int_
36031 [@{filename="gdb.mi/mi-sym-info-1.c",
36032 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36033 symbols=[@{line="27",name="typedef int my_int_t;"@}]@},
36034 @{filename="gdb.mi/mi-sym-info-2.c",
36035 fullname="/project/gdb.mi/mi-sym-info-2.c",
36036 symbols=[@{line="23",name="typedef int another_int_t;"@}]@}]@}
36040 @subheading The @code{-symbol-info-variables} Command
36041 @findex -symbol-info-variables
36042 @anchor{-symbol-info-variables}
36044 @subsubheading Synopsis
36047 -symbol-info-variables [--include-nondebug]
36048 [--type @var{type_regexp}]
36049 [--name @var{name_regexp}]
36050 [--max-results @var{limit}]
36055 Return a list containing the names and types for all global variables
36056 taken from the debug information. The variables are grouped by source
36057 file, and shown with the line number on which each variable is
36060 The @code{--include-nondebug} option causes the output to include
36061 data symbols from the symbol table.
36063 The options @code{--type} and @code{--name} allow the symbols returned
36064 to be filtered based on either the name of the variable, or the type
36067 The option @code{--max-results} restricts the command to return no
36068 more than @var{limit} results. If exactly @var{limit} results are
36069 returned then there might be additional results available if a higher
36072 @subsubheading @value{GDBN} Command
36074 The corresponding @value{GDBN} command is @samp{info variables}.
36076 @subsubheading Example
36080 -symbol-info-variables
36083 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36084 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36085 symbols=[@{line="25",name="global_f1",type="float",
36086 description="static float global_f1;"@},
36087 @{line="24",name="global_i1",type="int",
36088 description="static int global_i1;"@}]@},
36089 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36090 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36091 symbols=[@{line="21",name="global_f2",type="int",
36092 description="int global_f2;"@},
36093 @{line="20",name="global_i2",type="int",
36094 description="int global_i2;"@},
36095 @{line="19",name="global_f1",type="float",
36096 description="static float global_f1;"@},
36097 @{line="18",name="global_i1",type="int",
36098 description="static int global_i1;"@}]@}]@}
36102 -symbol-info-variables --name f1
36105 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36106 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36107 symbols=[@{line="25",name="global_f1",type="float",
36108 description="static float global_f1;"@}]@},
36109 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36110 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36111 symbols=[@{line="19",name="global_f1",type="float",
36112 description="static float global_f1;"@}]@}]@}
36116 -symbol-info-variables --type float
36119 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36120 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36121 symbols=[@{line="25",name="global_f1",type="float",
36122 description="static float global_f1;"@}]@},
36123 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36124 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36125 symbols=[@{line="19",name="global_f1",type="float",
36126 description="static float global_f1;"@}]@}]@}
36130 -symbol-info-variables --include-nondebug
36133 [@{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36134 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-1.c",
36135 symbols=[@{line="25",name="global_f1",type="float",
36136 description="static float global_f1;"@},
36137 @{line="24",name="global_i1",type="int",
36138 description="static int global_i1;"@}]@},
36139 @{filename="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36140 fullname="/project/gdb/testsuite/gdb.mi/mi-sym-info-2.c",
36141 symbols=[@{line="21",name="global_f2",type="int",
36142 description="int global_f2;"@},
36143 @{line="20",name="global_i2",type="int",
36144 description="int global_i2;"@},
36145 @{line="19",name="global_f1",type="float",
36146 description="static float global_f1;"@},
36147 @{line="18",name="global_i1",type="int",
36148 description="static int global_i1;"@}]@}],
36150 [@{address="0x00000000004005d0",name="_IO_stdin_used"@},
36151 @{address="0x00000000004005d8",name="__dso_handle"@}
36158 @subheading The @code{-symbol-info-line} Command
36159 @findex -symbol-info-line
36161 @subsubheading Synopsis
36167 Show the core addresses of the code for a source line.
36169 @subsubheading @value{GDBN} Command
36171 The corresponding @value{GDBN} command is @samp{info line}.
36172 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
36174 @subsubheading Example
36178 @subheading The @code{-symbol-info-symbol} Command
36179 @findex -symbol-info-symbol
36181 @subsubheading Synopsis
36184 -symbol-info-symbol @var{addr}
36187 Describe what symbol is at location @var{addr}.
36189 @subsubheading @value{GDBN} Command
36191 The corresponding @value{GDBN} command is @samp{info symbol}.
36193 @subsubheading Example
36197 @subheading The @code{-symbol-list-functions} Command
36198 @findex -symbol-list-functions
36200 @subsubheading Synopsis
36203 -symbol-list-functions
36206 List the functions in the executable.
36208 @subsubheading @value{GDBN} Command
36210 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
36211 @samp{gdb_search} in @code{gdbtk}.
36213 @subsubheading Example
36218 @subheading The @code{-symbol-list-lines} Command
36219 @findex -symbol-list-lines
36221 @subsubheading Synopsis
36224 -symbol-list-lines @var{filename}
36227 Print the list of lines that contain code and their associated program
36228 addresses for the given source filename. The entries are sorted in
36229 ascending PC order.
36231 @subsubheading @value{GDBN} Command
36233 There is no corresponding @value{GDBN} command.
36235 @subsubheading Example
36238 -symbol-list-lines basics.c
36239 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
36245 @subheading The @code{-symbol-list-types} Command
36246 @findex -symbol-list-types
36248 @subsubheading Synopsis
36254 List all the type names.
36256 @subsubheading @value{GDBN} Command
36258 The corresponding commands are @samp{info types} in @value{GDBN},
36259 @samp{gdb_search} in @code{gdbtk}.
36261 @subsubheading Example
36265 @subheading The @code{-symbol-list-variables} Command
36266 @findex -symbol-list-variables
36268 @subsubheading Synopsis
36271 -symbol-list-variables
36274 List all the global and static variable names.
36276 @subsubheading @value{GDBN} Command
36278 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
36280 @subsubheading Example
36284 @subheading The @code{-symbol-locate} Command
36285 @findex -symbol-locate
36287 @subsubheading Synopsis
36293 @subsubheading @value{GDBN} Command
36295 @samp{gdb_loc} in @code{gdbtk}.
36297 @subsubheading Example
36301 @subheading The @code{-symbol-type} Command
36302 @findex -symbol-type
36304 @subsubheading Synopsis
36307 -symbol-type @var{variable}
36310 Show type of @var{variable}.
36312 @subsubheading @value{GDBN} Command
36314 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
36315 @samp{gdb_obj_variable}.
36317 @subsubheading Example
36322 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
36323 @node GDB/MI File Commands
36324 @section @sc{gdb/mi} File Commands
36326 This section describes the GDB/MI commands to specify executable file names
36327 and to read in and obtain symbol table information.
36329 @subheading The @code{-file-exec-and-symbols} Command
36330 @findex -file-exec-and-symbols
36332 @subsubheading Synopsis
36335 -file-exec-and-symbols @var{file}
36338 Specify the executable file to be debugged. This file is the one from
36339 which the symbol table is also read. If no file is specified, the
36340 command clears the executable and symbol information. If breakpoints
36341 are set when using this command with no arguments, @value{GDBN} will produce
36342 error messages. Otherwise, no output is produced, except a completion
36345 @subsubheading @value{GDBN} Command
36347 The corresponding @value{GDBN} command is @samp{file}.
36349 @subsubheading Example
36353 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
36359 @subheading The @code{-file-exec-file} Command
36360 @findex -file-exec-file
36362 @subsubheading Synopsis
36365 -file-exec-file @var{file}
36368 Specify the executable file to be debugged. Unlike
36369 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
36370 from this file. If used without argument, @value{GDBN} clears the information
36371 about the executable file. No output is produced, except a completion
36374 @subsubheading @value{GDBN} Command
36376 The corresponding @value{GDBN} command is @samp{exec-file}.
36378 @subsubheading Example
36382 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
36389 @subheading The @code{-file-list-exec-sections} Command
36390 @findex -file-list-exec-sections
36392 @subsubheading Synopsis
36395 -file-list-exec-sections
36398 List the sections of the current executable file.
36400 @subsubheading @value{GDBN} Command
36402 The @value{GDBN} command @samp{info file} shows, among the rest, the same
36403 information as this command. @code{gdbtk} has a corresponding command
36404 @samp{gdb_load_info}.
36406 @subsubheading Example
36411 @subheading The @code{-file-list-exec-source-file} Command
36412 @findex -file-list-exec-source-file
36414 @subsubheading Synopsis
36417 -file-list-exec-source-file
36420 List the line number, the current source file, and the absolute path
36421 to the current source file for the current executable. The macro
36422 information field has a value of @samp{1} or @samp{0} depending on
36423 whether or not the file includes preprocessor macro information.
36425 @subsubheading @value{GDBN} Command
36427 The @value{GDBN} equivalent is @samp{info source}
36429 @subsubheading Example
36433 123-file-list-exec-source-file
36434 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
36439 @subheading The @code{-file-list-exec-source-files} Command
36440 @findex -file-list-exec-source-files
36442 @subsubheading Synopsis
36445 -file-list-exec-source-files
36448 List the source files for the current executable.
36450 It will always output both the filename and fullname (absolute file
36451 name) of a source file.
36453 @subsubheading @value{GDBN} Command
36455 The @value{GDBN} equivalent is @samp{info sources}.
36456 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
36458 @subsubheading Example
36461 -file-list-exec-source-files
36463 @{file=foo.c,fullname=/home/foo.c@},
36464 @{file=/home/bar.c,fullname=/home/bar.c@},
36465 @{file=gdb_could_not_find_fullpath.c@}]
36469 @subheading The @code{-file-list-shared-libraries} Command
36470 @findex -file-list-shared-libraries
36472 @subsubheading Synopsis
36475 -file-list-shared-libraries [ @var{regexp} ]
36478 List the shared libraries in the program.
36479 With a regular expression @var{regexp}, only those libraries whose
36480 names match @var{regexp} are listed.
36482 @subsubheading @value{GDBN} Command
36484 The corresponding @value{GDBN} command is @samp{info shared}. The fields
36485 have a similar meaning to the @code{=library-loaded} notification.
36486 The @code{ranges} field specifies the multiple segments belonging to this
36487 library. Each range has the following fields:
36491 The address defining the inclusive lower bound of the segment.
36493 The address defining the exclusive upper bound of the segment.
36496 @subsubheading Example
36499 -file-list-exec-source-files
36500 ^done,shared-libraries=[
36501 @{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"@}]@},
36502 @{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"@}]@}]
36508 @subheading The @code{-file-list-symbol-files} Command
36509 @findex -file-list-symbol-files
36511 @subsubheading Synopsis
36514 -file-list-symbol-files
36519 @subsubheading @value{GDBN} Command
36521 The corresponding @value{GDBN} command is @samp{info file} (part of it).
36523 @subsubheading Example
36528 @subheading The @code{-file-symbol-file} Command
36529 @findex -file-symbol-file
36531 @subsubheading Synopsis
36534 -file-symbol-file @var{file}
36537 Read symbol table info from the specified @var{file} argument. When
36538 used without arguments, clears @value{GDBN}'s symbol table info. No output is
36539 produced, except for a completion notification.
36541 @subsubheading @value{GDBN} Command
36543 The corresponding @value{GDBN} command is @samp{symbol-file}.
36545 @subsubheading Example
36549 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
36555 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
36556 @node GDB/MI Memory Overlay Commands
36557 @section @sc{gdb/mi} Memory Overlay Commands
36559 The memory overlay commands are not implemented.
36561 @c @subheading -overlay-auto
36563 @c @subheading -overlay-list-mapping-state
36565 @c @subheading -overlay-list-overlays
36567 @c @subheading -overlay-map
36569 @c @subheading -overlay-off
36571 @c @subheading -overlay-on
36573 @c @subheading -overlay-unmap
36575 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
36576 @node GDB/MI Signal Handling Commands
36577 @section @sc{gdb/mi} Signal Handling Commands
36579 Signal handling commands are not implemented.
36581 @c @subheading -signal-handle
36583 @c @subheading -signal-list-handle-actions
36585 @c @subheading -signal-list-signal-types
36589 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
36590 @node GDB/MI Target Manipulation
36591 @section @sc{gdb/mi} Target Manipulation Commands
36594 @subheading The @code{-target-attach} Command
36595 @findex -target-attach
36597 @subsubheading Synopsis
36600 -target-attach @var{pid} | @var{gid} | @var{file}
36603 Attach to a process @var{pid} or a file @var{file} outside of
36604 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
36605 group, the id previously returned by
36606 @samp{-list-thread-groups --available} must be used.
36608 @subsubheading @value{GDBN} Command
36610 The corresponding @value{GDBN} command is @samp{attach}.
36612 @subsubheading Example
36616 =thread-created,id="1"
36617 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
36623 @subheading The @code{-target-compare-sections} Command
36624 @findex -target-compare-sections
36626 @subsubheading Synopsis
36629 -target-compare-sections [ @var{section} ]
36632 Compare data of section @var{section} on target to the exec file.
36633 Without the argument, all sections are compared.
36635 @subsubheading @value{GDBN} Command
36637 The @value{GDBN} equivalent is @samp{compare-sections}.
36639 @subsubheading Example
36644 @subheading The @code{-target-detach} Command
36645 @findex -target-detach
36647 @subsubheading Synopsis
36650 -target-detach [ @var{pid} | @var{gid} ]
36653 Detach from the remote target which normally resumes its execution.
36654 If either @var{pid} or @var{gid} is specified, detaches from either
36655 the specified process, or specified thread group. There's no output.
36657 @subsubheading @value{GDBN} Command
36659 The corresponding @value{GDBN} command is @samp{detach}.
36661 @subsubheading Example
36671 @subheading The @code{-target-disconnect} Command
36672 @findex -target-disconnect
36674 @subsubheading Synopsis
36680 Disconnect from the remote target. There's no output and the target is
36681 generally not resumed.
36683 @subsubheading @value{GDBN} Command
36685 The corresponding @value{GDBN} command is @samp{disconnect}.
36687 @subsubheading Example
36697 @subheading The @code{-target-download} Command
36698 @findex -target-download
36700 @subsubheading Synopsis
36706 Loads the executable onto the remote target.
36707 It prints out an update message every half second, which includes the fields:
36711 The name of the section.
36713 The size of what has been sent so far for that section.
36715 The size of the section.
36717 The total size of what was sent so far (the current and the previous sections).
36719 The size of the overall executable to download.
36723 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
36724 @sc{gdb/mi} Output Syntax}).
36726 In addition, it prints the name and size of the sections, as they are
36727 downloaded. These messages include the following fields:
36731 The name of the section.
36733 The size of the section.
36735 The size of the overall executable to download.
36739 At the end, a summary is printed.
36741 @subsubheading @value{GDBN} Command
36743 The corresponding @value{GDBN} command is @samp{load}.
36745 @subsubheading Example
36747 Note: each status message appears on a single line. Here the messages
36748 have been broken down so that they can fit onto a page.
36753 +download,@{section=".text",section-size="6668",total-size="9880"@}
36754 +download,@{section=".text",section-sent="512",section-size="6668",
36755 total-sent="512",total-size="9880"@}
36756 +download,@{section=".text",section-sent="1024",section-size="6668",
36757 total-sent="1024",total-size="9880"@}
36758 +download,@{section=".text",section-sent="1536",section-size="6668",
36759 total-sent="1536",total-size="9880"@}
36760 +download,@{section=".text",section-sent="2048",section-size="6668",
36761 total-sent="2048",total-size="9880"@}
36762 +download,@{section=".text",section-sent="2560",section-size="6668",
36763 total-sent="2560",total-size="9880"@}
36764 +download,@{section=".text",section-sent="3072",section-size="6668",
36765 total-sent="3072",total-size="9880"@}
36766 +download,@{section=".text",section-sent="3584",section-size="6668",
36767 total-sent="3584",total-size="9880"@}
36768 +download,@{section=".text",section-sent="4096",section-size="6668",
36769 total-sent="4096",total-size="9880"@}
36770 +download,@{section=".text",section-sent="4608",section-size="6668",
36771 total-sent="4608",total-size="9880"@}
36772 +download,@{section=".text",section-sent="5120",section-size="6668",
36773 total-sent="5120",total-size="9880"@}
36774 +download,@{section=".text",section-sent="5632",section-size="6668",
36775 total-sent="5632",total-size="9880"@}
36776 +download,@{section=".text",section-sent="6144",section-size="6668",
36777 total-sent="6144",total-size="9880"@}
36778 +download,@{section=".text",section-sent="6656",section-size="6668",
36779 total-sent="6656",total-size="9880"@}
36780 +download,@{section=".init",section-size="28",total-size="9880"@}
36781 +download,@{section=".fini",section-size="28",total-size="9880"@}
36782 +download,@{section=".data",section-size="3156",total-size="9880"@}
36783 +download,@{section=".data",section-sent="512",section-size="3156",
36784 total-sent="7236",total-size="9880"@}
36785 +download,@{section=".data",section-sent="1024",section-size="3156",
36786 total-sent="7748",total-size="9880"@}
36787 +download,@{section=".data",section-sent="1536",section-size="3156",
36788 total-sent="8260",total-size="9880"@}
36789 +download,@{section=".data",section-sent="2048",section-size="3156",
36790 total-sent="8772",total-size="9880"@}
36791 +download,@{section=".data",section-sent="2560",section-size="3156",
36792 total-sent="9284",total-size="9880"@}
36793 +download,@{section=".data",section-sent="3072",section-size="3156",
36794 total-sent="9796",total-size="9880"@}
36795 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
36802 @subheading The @code{-target-exec-status} Command
36803 @findex -target-exec-status
36805 @subsubheading Synopsis
36808 -target-exec-status
36811 Provide information on the state of the target (whether it is running or
36812 not, for instance).
36814 @subsubheading @value{GDBN} Command
36816 There's no equivalent @value{GDBN} command.
36818 @subsubheading Example
36822 @subheading The @code{-target-list-available-targets} Command
36823 @findex -target-list-available-targets
36825 @subsubheading Synopsis
36828 -target-list-available-targets
36831 List the possible targets to connect to.
36833 @subsubheading @value{GDBN} Command
36835 The corresponding @value{GDBN} command is @samp{help target}.
36837 @subsubheading Example
36841 @subheading The @code{-target-list-current-targets} Command
36842 @findex -target-list-current-targets
36844 @subsubheading Synopsis
36847 -target-list-current-targets
36850 Describe the current target.
36852 @subsubheading @value{GDBN} Command
36854 The corresponding information is printed by @samp{info file} (among
36857 @subsubheading Example
36861 @subheading The @code{-target-list-parameters} Command
36862 @findex -target-list-parameters
36864 @subsubheading Synopsis
36867 -target-list-parameters
36873 @subsubheading @value{GDBN} Command
36877 @subsubheading Example
36880 @subheading The @code{-target-flash-erase} Command
36881 @findex -target-flash-erase
36883 @subsubheading Synopsis
36886 -target-flash-erase
36889 Erases all known flash memory regions on the target.
36891 The corresponding @value{GDBN} command is @samp{flash-erase}.
36893 The output is a list of flash regions that have been erased, with starting
36894 addresses and memory region sizes.
36898 -target-flash-erase
36899 ^done,erased-regions=@{address="0x0",size="0x40000"@}
36903 @subheading The @code{-target-select} Command
36904 @findex -target-select
36906 @subsubheading Synopsis
36909 -target-select @var{type} @var{parameters @dots{}}
36912 Connect @value{GDBN} to the remote target. This command takes two args:
36916 The type of target, for instance @samp{remote}, etc.
36917 @item @var{parameters}
36918 Device names, host names and the like. @xref{Target Commands, ,
36919 Commands for Managing Targets}, for more details.
36922 The output is a connection notification, followed by the address at
36923 which the target program is, in the following form:
36926 ^connected,addr="@var{address}",func="@var{function name}",
36927 args=[@var{arg list}]
36930 @subsubheading @value{GDBN} Command
36932 The corresponding @value{GDBN} command is @samp{target}.
36934 @subsubheading Example
36938 -target-select remote /dev/ttya
36939 ^connected,addr="0xfe00a300",func="??",args=[]
36943 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
36944 @node GDB/MI File Transfer Commands
36945 @section @sc{gdb/mi} File Transfer Commands
36948 @subheading The @code{-target-file-put} Command
36949 @findex -target-file-put
36951 @subsubheading Synopsis
36954 -target-file-put @var{hostfile} @var{targetfile}
36957 Copy file @var{hostfile} from the host system (the machine running
36958 @value{GDBN}) to @var{targetfile} on the target system.
36960 @subsubheading @value{GDBN} Command
36962 The corresponding @value{GDBN} command is @samp{remote put}.
36964 @subsubheading Example
36968 -target-file-put localfile remotefile
36974 @subheading The @code{-target-file-get} Command
36975 @findex -target-file-get
36977 @subsubheading Synopsis
36980 -target-file-get @var{targetfile} @var{hostfile}
36983 Copy file @var{targetfile} from the target system to @var{hostfile}
36984 on the host system.
36986 @subsubheading @value{GDBN} Command
36988 The corresponding @value{GDBN} command is @samp{remote get}.
36990 @subsubheading Example
36994 -target-file-get remotefile localfile
37000 @subheading The @code{-target-file-delete} Command
37001 @findex -target-file-delete
37003 @subsubheading Synopsis
37006 -target-file-delete @var{targetfile}
37009 Delete @var{targetfile} from the target system.
37011 @subsubheading @value{GDBN} Command
37013 The corresponding @value{GDBN} command is @samp{remote delete}.
37015 @subsubheading Example
37019 -target-file-delete remotefile
37025 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
37026 @node GDB/MI Ada Exceptions Commands
37027 @section Ada Exceptions @sc{gdb/mi} Commands
37029 @subheading The @code{-info-ada-exceptions} Command
37030 @findex -info-ada-exceptions
37032 @subsubheading Synopsis
37035 -info-ada-exceptions [ @var{regexp}]
37038 List all Ada exceptions defined within the program being debugged.
37039 With a regular expression @var{regexp}, only those exceptions whose
37040 names match @var{regexp} are listed.
37042 @subsubheading @value{GDBN} Command
37044 The corresponding @value{GDBN} command is @samp{info exceptions}.
37046 @subsubheading Result
37048 The result is a table of Ada exceptions. The following columns are
37049 defined for each exception:
37053 The name of the exception.
37056 The address of the exception.
37060 @subsubheading Example
37063 -info-ada-exceptions aint
37064 ^done,ada-exceptions=@{nr_rows="2",nr_cols="2",
37065 hdr=[@{width="1",alignment="-1",col_name="name",colhdr="Name"@},
37066 @{width="1",alignment="-1",col_name="address",colhdr="Address"@}],
37067 body=[@{name="constraint_error",address="0x0000000000613da0"@},
37068 @{name="const.aint_global_e",address="0x0000000000613b00"@}]@}
37071 @subheading Catching Ada Exceptions
37073 The commands describing how to ask @value{GDBN} to stop when a program
37074 raises an exception are described at @ref{Ada Exception GDB/MI
37075 Catchpoint Commands}.
37078 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
37079 @node GDB/MI Support Commands
37080 @section @sc{gdb/mi} Support Commands
37082 Since new commands and features get regularly added to @sc{gdb/mi},
37083 some commands are available to help front-ends query the debugger
37084 about support for these capabilities. Similarly, it is also possible
37085 to query @value{GDBN} about target support of certain features.
37087 @subheading The @code{-info-gdb-mi-command} Command
37088 @cindex @code{-info-gdb-mi-command}
37089 @findex -info-gdb-mi-command
37091 @subsubheading Synopsis
37094 -info-gdb-mi-command @var{cmd_name}
37097 Query support for the @sc{gdb/mi} command named @var{cmd_name}.
37099 Note that the dash (@code{-}) starting all @sc{gdb/mi} commands
37100 is technically not part of the command name (@pxref{GDB/MI Input
37101 Syntax}), and thus should be omitted in @var{cmd_name}. However,
37102 for ease of use, this command also accepts the form with the leading
37105 @subsubheading @value{GDBN} Command
37107 There is no corresponding @value{GDBN} command.
37109 @subsubheading Result
37111 The result is a tuple. There is currently only one field:
37115 This field is equal to @code{"true"} if the @sc{gdb/mi} command exists,
37116 @code{"false"} otherwise.
37120 @subsubheading Example
37122 Here is an example where the @sc{gdb/mi} command does not exist:
37125 -info-gdb-mi-command unsupported-command
37126 ^done,command=@{exists="false"@}
37130 And here is an example where the @sc{gdb/mi} command is known
37134 -info-gdb-mi-command symbol-list-lines
37135 ^done,command=@{exists="true"@}
37138 @subheading The @code{-list-features} Command
37139 @findex -list-features
37140 @cindex supported @sc{gdb/mi} features, list
37142 Returns a list of particular features of the MI protocol that
37143 this version of gdb implements. A feature can be a command,
37144 or a new field in an output of some command, or even an
37145 important bugfix. While a frontend can sometimes detect presence
37146 of a feature at runtime, it is easier to perform detection at debugger
37149 The command returns a list of strings, with each string naming an
37150 available feature. Each returned string is just a name, it does not
37151 have any internal structure. The list of possible feature names
37157 (gdb) -list-features
37158 ^done,result=["feature1","feature2"]
37161 The current list of features is:
37164 @item frozen-varobjs
37165 Indicates support for the @code{-var-set-frozen} command, as well
37166 as possible presence of the @code{frozen} field in the output
37167 of @code{-varobj-create}.
37168 @item pending-breakpoints
37169 Indicates support for the @option{-f} option to the @code{-break-insert}
37172 Indicates Python scripting support, Python-based
37173 pretty-printing commands, and possible presence of the
37174 @samp{display_hint} field in the output of @code{-var-list-children}
37176 Indicates support for the @code{-thread-info} command.
37177 @item data-read-memory-bytes
37178 Indicates support for the @code{-data-read-memory-bytes} and the
37179 @code{-data-write-memory-bytes} commands.
37180 @item breakpoint-notifications
37181 Indicates that changes to breakpoints and breakpoints created via the
37182 CLI will be announced via async records.
37183 @item ada-task-info
37184 Indicates support for the @code{-ada-task-info} command.
37185 @item language-option
37186 Indicates that all @sc{gdb/mi} commands accept the @option{--language}
37187 option (@pxref{Context management}).
37188 @item info-gdb-mi-command
37189 Indicates support for the @code{-info-gdb-mi-command} command.
37190 @item undefined-command-error-code
37191 Indicates support for the "undefined-command" error code in error result
37192 records, produced when trying to execute an undefined @sc{gdb/mi} command
37193 (@pxref{GDB/MI Result Records}).
37194 @item exec-run-start-option
37195 Indicates that the @code{-exec-run} command supports the @option{--start}
37196 option (@pxref{GDB/MI Program Execution}).
37197 @item data-disassemble-a-option
37198 Indicates that the @code{-data-disassemble} command supports the @option{-a}
37199 option (@pxref{GDB/MI Data Manipulation}).
37202 @subheading The @code{-list-target-features} Command
37203 @findex -list-target-features
37205 Returns a list of particular features that are supported by the
37206 target. Those features affect the permitted MI commands, but
37207 unlike the features reported by the @code{-list-features} command, the
37208 features depend on which target GDB is using at the moment. Whenever
37209 a target can change, due to commands such as @code{-target-select},
37210 @code{-target-attach} or @code{-exec-run}, the list of target features
37211 may change, and the frontend should obtain it again.
37215 (gdb) -list-target-features
37216 ^done,result=["async"]
37219 The current list of features is:
37223 Indicates that the target is capable of asynchronous command
37224 execution, which means that @value{GDBN} will accept further commands
37225 while the target is running.
37228 Indicates that the target is capable of reverse execution.
37229 @xref{Reverse Execution}, for more information.
37233 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
37234 @node GDB/MI Miscellaneous Commands
37235 @section Miscellaneous @sc{gdb/mi} Commands
37237 @c @subheading -gdb-complete
37239 @subheading The @code{-gdb-exit} Command
37242 @subsubheading Synopsis
37248 Exit @value{GDBN} immediately.
37250 @subsubheading @value{GDBN} Command
37252 Approximately corresponds to @samp{quit}.
37254 @subsubheading Example
37264 @subheading The @code{-exec-abort} Command
37265 @findex -exec-abort
37267 @subsubheading Synopsis
37273 Kill the inferior running program.
37275 @subsubheading @value{GDBN} Command
37277 The corresponding @value{GDBN} command is @samp{kill}.
37279 @subsubheading Example
37284 @subheading The @code{-gdb-set} Command
37287 @subsubheading Synopsis
37293 Set an internal @value{GDBN} variable.
37294 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
37296 @subsubheading @value{GDBN} Command
37298 The corresponding @value{GDBN} command is @samp{set}.
37300 @subsubheading Example
37310 @subheading The @code{-gdb-show} Command
37313 @subsubheading Synopsis
37319 Show the current value of a @value{GDBN} variable.
37321 @subsubheading @value{GDBN} Command
37323 The corresponding @value{GDBN} command is @samp{show}.
37325 @subsubheading Example
37334 @c @subheading -gdb-source
37337 @subheading The @code{-gdb-version} Command
37338 @findex -gdb-version
37340 @subsubheading Synopsis
37346 Show version information for @value{GDBN}. Used mostly in testing.
37348 @subsubheading @value{GDBN} Command
37350 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
37351 default shows this information when you start an interactive session.
37353 @subsubheading Example
37355 @c This example modifies the actual output from GDB to avoid overfull
37361 ~Copyright 2000 Free Software Foundation, Inc.
37362 ~GDB is free software, covered by the GNU General Public License, and
37363 ~you are welcome to change it and/or distribute copies of it under
37364 ~ certain conditions.
37365 ~Type "show copying" to see the conditions.
37366 ~There is absolutely no warranty for GDB. Type "show warranty" for
37368 ~This GDB was configured as
37369 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
37374 @subheading The @code{-list-thread-groups} Command
37375 @findex -list-thread-groups
37377 @subheading Synopsis
37380 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
37383 Lists thread groups (@pxref{Thread groups}). When a single thread
37384 group is passed as the argument, lists the children of that group.
37385 When several thread group are passed, lists information about those
37386 thread groups. Without any parameters, lists information about all
37387 top-level thread groups.
37389 Normally, thread groups that are being debugged are reported.
37390 With the @samp{--available} option, @value{GDBN} reports thread groups
37391 available on the target.
37393 The output of this command may have either a @samp{threads} result or
37394 a @samp{groups} result. The @samp{thread} result has a list of tuples
37395 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
37396 Information}). The @samp{groups} result has a list of tuples as value,
37397 each tuple describing a thread group. If top-level groups are
37398 requested (that is, no parameter is passed), or when several groups
37399 are passed, the output always has a @samp{groups} result. The format
37400 of the @samp{group} result is described below.
37402 To reduce the number of roundtrips it's possible to list thread groups
37403 together with their children, by passing the @samp{--recurse} option
37404 and the recursion depth. Presently, only recursion depth of 1 is
37405 permitted. If this option is present, then every reported thread group
37406 will also include its children, either as @samp{group} or
37407 @samp{threads} field.
37409 In general, any combination of option and parameters is permitted, with
37410 the following caveats:
37414 When a single thread group is passed, the output will typically
37415 be the @samp{threads} result. Because threads may not contain
37416 anything, the @samp{recurse} option will be ignored.
37419 When the @samp{--available} option is passed, limited information may
37420 be available. In particular, the list of threads of a process might
37421 be inaccessible. Further, specifying specific thread groups might
37422 not give any performance advantage over listing all thread groups.
37423 The frontend should assume that @samp{-list-thread-groups --available}
37424 is always an expensive operation and cache the results.
37428 The @samp{groups} result is a list of tuples, where each tuple may
37429 have the following fields:
37433 Identifier of the thread group. This field is always present.
37434 The identifier is an opaque string; frontends should not try to
37435 convert it to an integer, even though it might look like one.
37438 The type of the thread group. At present, only @samp{process} is a
37442 The target-specific process identifier. This field is only present
37443 for thread groups of type @samp{process} and only if the process exists.
37446 The exit code of this group's last exited thread, formatted in octal.
37447 This field is only present for thread groups of type @samp{process} and
37448 only if the process is not running.
37451 The number of children this thread group has. This field may be
37452 absent for an available thread group.
37455 This field has a list of tuples as value, each tuple describing a
37456 thread. It may be present if the @samp{--recurse} option is
37457 specified, and it's actually possible to obtain the threads.
37460 This field is a list of integers, each identifying a core that one
37461 thread of the group is running on. This field may be absent if
37462 such information is not available.
37465 The name of the executable file that corresponds to this thread group.
37466 The field is only present for thread groups of type @samp{process},
37467 and only if there is a corresponding executable file.
37471 @subheading Example
37475 -list-thread-groups
37476 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
37477 -list-thread-groups 17
37478 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
37479 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
37480 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
37481 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
37482 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},state="running"@}]]
37483 -list-thread-groups --available
37484 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
37485 -list-thread-groups --available --recurse 1
37486 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
37487 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
37488 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
37489 -list-thread-groups --available --recurse 1 17 18
37490 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
37491 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
37492 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
37495 @subheading The @code{-info-os} Command
37498 @subsubheading Synopsis
37501 -info-os [ @var{type} ]
37504 If no argument is supplied, the command returns a table of available
37505 operating-system-specific information types. If one of these types is
37506 supplied as an argument @var{type}, then the command returns a table
37507 of data of that type.
37509 The types of information available depend on the target operating
37512 @subsubheading @value{GDBN} Command
37514 The corresponding @value{GDBN} command is @samp{info os}.
37516 @subsubheading Example
37518 When run on a @sc{gnu}/Linux system, the output will look something
37524 ^done,OSDataTable=@{nr_rows="10",nr_cols="3",
37525 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
37526 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
37527 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
37528 body=[item=@{col0="cpus",col1="Listing of all cpus/cores on the system",
37530 item=@{col0="files",col1="Listing of all file descriptors",
37531 col2="File descriptors"@},
37532 item=@{col0="modules",col1="Listing of all loaded kernel modules",
37533 col2="Kernel modules"@},
37534 item=@{col0="msg",col1="Listing of all message queues",
37535 col2="Message queues"@},
37536 item=@{col0="processes",col1="Listing of all processes",
37537 col2="Processes"@},
37538 item=@{col0="procgroups",col1="Listing of all process groups",
37539 col2="Process groups"@},
37540 item=@{col0="semaphores",col1="Listing of all semaphores",
37541 col2="Semaphores"@},
37542 item=@{col0="shm",col1="Listing of all shared-memory regions",
37543 col2="Shared-memory regions"@},
37544 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
37546 item=@{col0="threads",col1="Listing of all threads",
37550 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
37551 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
37552 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
37553 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
37554 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
37555 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
37556 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
37557 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
37559 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
37560 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
37564 (Note that the MI output here includes a @code{"Title"} column that
37565 does not appear in command-line @code{info os}; this column is useful
37566 for MI clients that want to enumerate the types of data, such as in a
37567 popup menu, but is needless clutter on the command line, and
37568 @code{info os} omits it.)
37570 @subheading The @code{-add-inferior} Command
37571 @findex -add-inferior
37573 @subheading Synopsis
37579 Creates a new inferior (@pxref{Inferiors and Programs}). The created
37580 inferior is not associated with any executable. Such association may
37581 be established with the @samp{-file-exec-and-symbols} command
37582 (@pxref{GDB/MI File Commands}). The command response has a single
37583 field, @samp{inferior}, whose value is the identifier of the
37584 thread group corresponding to the new inferior.
37586 @subheading Example
37591 ^done,inferior="i3"
37594 @subheading The @code{-interpreter-exec} Command
37595 @findex -interpreter-exec
37597 @subheading Synopsis
37600 -interpreter-exec @var{interpreter} @var{command}
37602 @anchor{-interpreter-exec}
37604 Execute the specified @var{command} in the given @var{interpreter}.
37606 @subheading @value{GDBN} Command
37608 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
37610 @subheading Example
37614 -interpreter-exec console "break main"
37615 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
37616 &"During symbol reading, bad structure-type format.\n"
37617 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
37622 @subheading The @code{-inferior-tty-set} Command
37623 @findex -inferior-tty-set
37625 @subheading Synopsis
37628 -inferior-tty-set /dev/pts/1
37631 Set terminal for future runs of the program being debugged.
37633 @subheading @value{GDBN} Command
37635 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
37637 @subheading Example
37641 -inferior-tty-set /dev/pts/1
37646 @subheading The @code{-inferior-tty-show} Command
37647 @findex -inferior-tty-show
37649 @subheading Synopsis
37655 Show terminal for future runs of program being debugged.
37657 @subheading @value{GDBN} Command
37659 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
37661 @subheading Example
37665 -inferior-tty-set /dev/pts/1
37669 ^done,inferior_tty_terminal="/dev/pts/1"
37673 @subheading The @code{-enable-timings} Command
37674 @findex -enable-timings
37676 @subheading Synopsis
37679 -enable-timings [yes | no]
37682 Toggle the printing of the wallclock, user and system times for an MI
37683 command as a field in its output. This command is to help frontend
37684 developers optimize the performance of their code. No argument is
37685 equivalent to @samp{yes}.
37687 @subheading @value{GDBN} Command
37691 @subheading Example
37699 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
37700 addr="0x080484ed",func="main",file="myprog.c",
37701 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
37703 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
37711 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
37712 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
37713 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
37714 fullname="/home/nickrob/myprog.c",line="73",arch="i386:x86_64"@}
37718 @subheading The @code{-complete} Command
37721 @subheading Synopsis
37724 -complete @var{command}
37727 Show a list of completions for partially typed CLI @var{command}.
37729 This command is intended for @sc{gdb/mi} frontends that cannot use two separate
37730 CLI and MI channels --- for example: because of lack of PTYs like on Windows or
37731 because @value{GDBN} is used remotely via a SSH connection.
37735 The result consists of two or three fields:
37739 This field contains the completed @var{command}. If @var{command}
37740 has no known completions, this field is omitted.
37743 This field contains a (possibly empty) array of matches. It is always present.
37745 @item max_completions_reached
37746 This field contains @code{1} if number of known completions is above
37747 @code{max-completions} limit (@pxref{Completion}), otherwise it contains
37748 @code{0}. It is always present.
37752 @subheading @value{GDBN} Command
37754 The corresponding @value{GDBN} command is @samp{complete}.
37756 @subheading Example
37761 ^done,completion="break",
37762 matches=["break","break-range"],
37763 max_completions_reached="0"
37766 ^done,completion="b ma",
37767 matches=["b madvise","b main"],max_completions_reached="0"
37769 -complete "b push_b"
37770 ^done,completion="b push_back(",
37772 "b A::push_back(void*)",
37773 "b std::string::push_back(char)",
37774 "b std::vector<int, std::allocator<int> >::push_back(int&&)"],
37775 max_completions_reached="0"
37777 -complete "nonexist"
37778 ^done,matches=[],max_completions_reached="0"
37784 @chapter @value{GDBN} Annotations
37786 This chapter describes annotations in @value{GDBN}. Annotations were
37787 designed to interface @value{GDBN} to graphical user interfaces or other
37788 similar programs which want to interact with @value{GDBN} at a
37789 relatively high level.
37791 The annotation mechanism has largely been superseded by @sc{gdb/mi}
37795 This is Edition @value{EDITION}, @value{DATE}.
37799 * Annotations Overview:: What annotations are; the general syntax.
37800 * Server Prefix:: Issuing a command without affecting user state.
37801 * Prompting:: Annotations marking @value{GDBN}'s need for input.
37802 * Errors:: Annotations for error messages.
37803 * Invalidation:: Some annotations describe things now invalid.
37804 * Annotations for Running::
37805 Whether the program is running, how it stopped, etc.
37806 * Source Annotations:: Annotations describing source code.
37809 @node Annotations Overview
37810 @section What is an Annotation?
37811 @cindex annotations
37813 Annotations start with a newline character, two @samp{control-z}
37814 characters, and the name of the annotation. If there is no additional
37815 information associated with this annotation, the name of the annotation
37816 is followed immediately by a newline. If there is additional
37817 information, the name of the annotation is followed by a space, the
37818 additional information, and a newline. The additional information
37819 cannot contain newline characters.
37821 Any output not beginning with a newline and two @samp{control-z}
37822 characters denotes literal output from @value{GDBN}. Currently there is
37823 no need for @value{GDBN} to output a newline followed by two
37824 @samp{control-z} characters, but if there was such a need, the
37825 annotations could be extended with an @samp{escape} annotation which
37826 means those three characters as output.
37828 The annotation @var{level}, which is specified using the
37829 @option{--annotate} command line option (@pxref{Mode Options}), controls
37830 how much information @value{GDBN} prints together with its prompt,
37831 values of expressions, source lines, and other types of output. Level 0
37832 is for no annotations, level 1 is for use when @value{GDBN} is run as a
37833 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
37834 for programs that control @value{GDBN}, and level 2 annotations have
37835 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
37836 Interface, annotate, GDB's Obsolete Annotations}).
37839 @kindex set annotate
37840 @item set annotate @var{level}
37841 The @value{GDBN} command @code{set annotate} sets the level of
37842 annotations to the specified @var{level}.
37844 @item show annotate
37845 @kindex show annotate
37846 Show the current annotation level.
37849 This chapter describes level 3 annotations.
37851 A simple example of starting up @value{GDBN} with annotations is:
37854 $ @kbd{gdb --annotate=3}
37856 Copyright 2003 Free Software Foundation, Inc.
37857 GDB is free software, covered by the GNU General Public License,
37858 and you are welcome to change it and/or distribute copies of it
37859 under certain conditions.
37860 Type "show copying" to see the conditions.
37861 There is absolutely no warranty for GDB. Type "show warranty"
37863 This GDB was configured as "i386-pc-linux-gnu"
37874 Here @samp{quit} is input to @value{GDBN}; the rest is output from
37875 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
37876 denotes a @samp{control-z} character) are annotations; the rest is
37877 output from @value{GDBN}.
37879 @node Server Prefix
37880 @section The Server Prefix
37881 @cindex server prefix
37883 If you prefix a command with @samp{server } then it will not affect
37884 the command history, nor will it affect @value{GDBN}'s notion of which
37885 command to repeat if @key{RET} is pressed on a line by itself. This
37886 means that commands can be run behind a user's back by a front-end in
37887 a transparent manner.
37889 The @code{server } prefix does not affect the recording of values into
37890 the value history; to print a value without recording it into the
37891 value history, use the @code{output} command instead of the
37892 @code{print} command.
37894 Using this prefix also disables confirmation requests
37895 (@pxref{confirmation requests}).
37898 @section Annotation for @value{GDBN} Input
37900 @cindex annotations for prompts
37901 When @value{GDBN} prompts for input, it annotates this fact so it is possible
37902 to know when to send output, when the output from a given command is
37905 Different kinds of input each have a different @dfn{input type}. Each
37906 input type has three annotations: a @code{pre-} annotation, which
37907 denotes the beginning of any prompt which is being output, a plain
37908 annotation, which denotes the end of the prompt, and then a @code{post-}
37909 annotation which denotes the end of any echo which may (or may not) be
37910 associated with the input. For example, the @code{prompt} input type
37911 features the following annotations:
37919 The input types are
37922 @findex pre-prompt annotation
37923 @findex prompt annotation
37924 @findex post-prompt annotation
37926 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
37928 @findex pre-commands annotation
37929 @findex commands annotation
37930 @findex post-commands annotation
37932 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
37933 command. The annotations are repeated for each command which is input.
37935 @findex pre-overload-choice annotation
37936 @findex overload-choice annotation
37937 @findex post-overload-choice annotation
37938 @item overload-choice
37939 When @value{GDBN} wants the user to select between various overloaded functions.
37941 @findex pre-query annotation
37942 @findex query annotation
37943 @findex post-query annotation
37945 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
37947 @findex pre-prompt-for-continue annotation
37948 @findex prompt-for-continue annotation
37949 @findex post-prompt-for-continue annotation
37950 @item prompt-for-continue
37951 When @value{GDBN} is asking the user to press return to continue. Note: Don't
37952 expect this to work well; instead use @code{set height 0} to disable
37953 prompting. This is because the counting of lines is buggy in the
37954 presence of annotations.
37959 @cindex annotations for errors, warnings and interrupts
37961 @findex quit annotation
37966 This annotation occurs right before @value{GDBN} responds to an interrupt.
37968 @findex error annotation
37973 This annotation occurs right before @value{GDBN} responds to an error.
37975 Quit and error annotations indicate that any annotations which @value{GDBN} was
37976 in the middle of may end abruptly. For example, if a
37977 @code{value-history-begin} annotation is followed by a @code{error}, one
37978 cannot expect to receive the matching @code{value-history-end}. One
37979 cannot expect not to receive it either, however; an error annotation
37980 does not necessarily mean that @value{GDBN} is immediately returning all the way
37983 @findex error-begin annotation
37984 A quit or error annotation may be preceded by
37990 Any output between that and the quit or error annotation is the error
37993 Warning messages are not yet annotated.
37994 @c If we want to change that, need to fix warning(), type_error(),
37995 @c range_error(), and possibly other places.
37998 @section Invalidation Notices
38000 @cindex annotations for invalidation messages
38001 The following annotations say that certain pieces of state may have
38005 @findex frames-invalid annotation
38006 @item ^Z^Zframes-invalid
38008 The frames (for example, output from the @code{backtrace} command) may
38011 @findex breakpoints-invalid annotation
38012 @item ^Z^Zbreakpoints-invalid
38014 The breakpoints may have changed. For example, the user just added or
38015 deleted a breakpoint.
38018 @node Annotations for Running
38019 @section Running the Program
38020 @cindex annotations for running programs
38022 @findex starting annotation
38023 @findex stopping annotation
38024 When the program starts executing due to a @value{GDBN} command such as
38025 @code{step} or @code{continue},
38031 is output. When the program stops,
38037 is output. Before the @code{stopped} annotation, a variety of
38038 annotations describe how the program stopped.
38041 @findex exited annotation
38042 @item ^Z^Zexited @var{exit-status}
38043 The program exited, and @var{exit-status} is the exit status (zero for
38044 successful exit, otherwise nonzero).
38046 @findex signalled annotation
38047 @findex signal-name annotation
38048 @findex signal-name-end annotation
38049 @findex signal-string annotation
38050 @findex signal-string-end annotation
38051 @item ^Z^Zsignalled
38052 The program exited with a signal. After the @code{^Z^Zsignalled}, the
38053 annotation continues:
38059 ^Z^Zsignal-name-end
38063 ^Z^Zsignal-string-end
38068 where @var{name} is the name of the signal, such as @code{SIGILL} or
38069 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
38070 as @code{Illegal Instruction} or @code{Segmentation fault}. The arguments
38071 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
38072 user's benefit and have no particular format.
38074 @findex signal annotation
38076 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
38077 just saying that the program received the signal, not that it was
38078 terminated with it.
38080 @findex breakpoint annotation
38081 @item ^Z^Zbreakpoint @var{number}
38082 The program hit breakpoint number @var{number}.
38084 @findex watchpoint annotation
38085 @item ^Z^Zwatchpoint @var{number}
38086 The program hit watchpoint number @var{number}.
38089 @node Source Annotations
38090 @section Displaying Source
38091 @cindex annotations for source display
38093 @findex source annotation
38094 The following annotation is used instead of displaying source code:
38097 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
38100 where @var{filename} is an absolute file name indicating which source
38101 file, @var{line} is the line number within that file (where 1 is the
38102 first line in the file), @var{character} is the character position
38103 within the file (where 0 is the first character in the file) (for most
38104 debug formats this will necessarily point to the beginning of a line),
38105 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
38106 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
38107 @var{addr} is the address in the target program associated with the
38108 source which is being displayed. The @var{addr} is in the form @samp{0x}
38109 followed by one or more lowercase hex digits (note that this does not
38110 depend on the language).
38112 @node JIT Interface
38113 @chapter JIT Compilation Interface
38114 @cindex just-in-time compilation
38115 @cindex JIT compilation interface
38117 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
38118 interface. A JIT compiler is a program or library that generates native
38119 executable code at runtime and executes it, usually in order to achieve good
38120 performance while maintaining platform independence.
38122 Programs that use JIT compilation are normally difficult to debug because
38123 portions of their code are generated at runtime, instead of being loaded from
38124 object files, which is where @value{GDBN} normally finds the program's symbols
38125 and debug information. In order to debug programs that use JIT compilation,
38126 @value{GDBN} has an interface that allows the program to register in-memory
38127 symbol files with @value{GDBN} at runtime.
38129 If you are using @value{GDBN} to debug a program that uses this interface, then
38130 it should work transparently so long as you have not stripped the binary. If
38131 you are developing a JIT compiler, then the interface is documented in the rest
38132 of this chapter. At this time, the only known client of this interface is the
38135 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
38136 JIT compiler communicates with @value{GDBN} by writing data into a global
38137 variable and calling a function at a well-known symbol. When @value{GDBN}
38138 attaches, it reads a linked list of symbol files from the global variable to
38139 find existing code, and puts a breakpoint in the function so that it can find
38140 out about additional code.
38143 * Declarations:: Relevant C struct declarations
38144 * Registering Code:: Steps to register code
38145 * Unregistering Code:: Steps to unregister code
38146 * Custom Debug Info:: Emit debug information in a custom format
38150 @section JIT Declarations
38152 These are the relevant struct declarations that a C program should include to
38153 implement the interface:
38163 struct jit_code_entry
38165 struct jit_code_entry *next_entry;
38166 struct jit_code_entry *prev_entry;
38167 const char *symfile_addr;
38168 uint64_t symfile_size;
38171 struct jit_descriptor
38174 /* This type should be jit_actions_t, but we use uint32_t
38175 to be explicit about the bitwidth. */
38176 uint32_t action_flag;
38177 struct jit_code_entry *relevant_entry;
38178 struct jit_code_entry *first_entry;
38181 /* GDB puts a breakpoint in this function. */
38182 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
38184 /* Make sure to specify the version statically, because the
38185 debugger may check the version before we can set it. */
38186 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
38189 If the JIT is multi-threaded, then it is important that the JIT synchronize any
38190 modifications to this global data properly, which can easily be done by putting
38191 a global mutex around modifications to these structures.
38193 @node Registering Code
38194 @section Registering Code
38196 To register code with @value{GDBN}, the JIT should follow this protocol:
38200 Generate an object file in memory with symbols and other desired debug
38201 information. The file must include the virtual addresses of the sections.
38204 Create a code entry for the file, which gives the start and size of the symbol
38208 Add it to the linked list in the JIT descriptor.
38211 Point the relevant_entry field of the descriptor at the entry.
38214 Set @code{action_flag} to @code{JIT_REGISTER} and call
38215 @code{__jit_debug_register_code}.
38218 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
38219 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
38220 new code. However, the linked list must still be maintained in order to allow
38221 @value{GDBN} to attach to a running process and still find the symbol files.
38223 @node Unregistering Code
38224 @section Unregistering Code
38226 If code is freed, then the JIT should use the following protocol:
38230 Remove the code entry corresponding to the code from the linked list.
38233 Point the @code{relevant_entry} field of the descriptor at the code entry.
38236 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
38237 @code{__jit_debug_register_code}.
38240 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
38241 and the JIT will leak the memory used for the associated symbol files.
38243 @node Custom Debug Info
38244 @section Custom Debug Info
38245 @cindex custom JIT debug info
38246 @cindex JIT debug info reader
38248 Generating debug information in platform-native file formats (like ELF
38249 or COFF) may be an overkill for JIT compilers; especially if all the
38250 debug info is used for is displaying a meaningful backtrace. The
38251 issue can be resolved by having the JIT writers decide on a debug info
38252 format and also provide a reader that parses the debug info generated
38253 by the JIT compiler. This section gives a brief overview on writing
38254 such a parser. More specific details can be found in the source file
38255 @file{gdb/jit-reader.in}, which is also installed as a header at
38256 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
38258 The reader is implemented as a shared object (so this functionality is
38259 not available on platforms which don't allow loading shared objects at
38260 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
38261 @code{jit-reader-unload} are provided, to be used to load and unload
38262 the readers from a preconfigured directory. Once loaded, the shared
38263 object is used the parse the debug information emitted by the JIT
38267 * Using JIT Debug Info Readers:: How to use supplied readers correctly
38268 * Writing JIT Debug Info Readers:: Creating a debug-info reader
38271 @node Using JIT Debug Info Readers
38272 @subsection Using JIT Debug Info Readers
38273 @kindex jit-reader-load
38274 @kindex jit-reader-unload
38276 Readers can be loaded and unloaded using the @code{jit-reader-load}
38277 and @code{jit-reader-unload} commands.
38280 @item jit-reader-load @var{reader}
38281 Load the JIT reader named @var{reader}, which is a shared
38282 object specified as either an absolute or a relative file name. In
38283 the latter case, @value{GDBN} will try to load the reader from a
38284 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
38285 system (here @var{libdir} is the system library directory, often
38286 @file{/usr/local/lib}).
38288 Only one reader can be active at a time; trying to load a second
38289 reader when one is already loaded will result in @value{GDBN}
38290 reporting an error. A new JIT reader can be loaded by first unloading
38291 the current one using @code{jit-reader-unload} and then invoking
38292 @code{jit-reader-load}.
38294 @item jit-reader-unload
38295 Unload the currently loaded JIT reader.
38299 @node Writing JIT Debug Info Readers
38300 @subsection Writing JIT Debug Info Readers
38301 @cindex writing JIT debug info readers
38303 As mentioned, a reader is essentially a shared object conforming to a
38304 certain ABI. This ABI is described in @file{jit-reader.h}.
38306 @file{jit-reader.h} defines the structures, macros and functions
38307 required to write a reader. It is installed (along with
38308 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
38309 the system include directory.
38311 Readers need to be released under a GPL compatible license. A reader
38312 can be declared as released under such a license by placing the macro
38313 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
38315 The entry point for readers is the symbol @code{gdb_init_reader},
38316 which is expected to be a function with the prototype
38318 @findex gdb_init_reader
38320 extern struct gdb_reader_funcs *gdb_init_reader (void);
38323 @cindex @code{struct gdb_reader_funcs}
38325 @code{struct gdb_reader_funcs} contains a set of pointers to callback
38326 functions. These functions are executed to read the debug info
38327 generated by the JIT compiler (@code{read}), to unwind stack frames
38328 (@code{unwind}) and to create canonical frame IDs
38329 (@code{get_frame_id}). It also has a callback that is called when the
38330 reader is being unloaded (@code{destroy}). The struct looks like this
38333 struct gdb_reader_funcs
38335 /* Must be set to GDB_READER_INTERFACE_VERSION. */
38336 int reader_version;
38338 /* For use by the reader. */
38341 gdb_read_debug_info *read;
38342 gdb_unwind_frame *unwind;
38343 gdb_get_frame_id *get_frame_id;
38344 gdb_destroy_reader *destroy;
38348 @cindex @code{struct gdb_symbol_callbacks}
38349 @cindex @code{struct gdb_unwind_callbacks}
38351 The callbacks are provided with another set of callbacks by
38352 @value{GDBN} to do their job. For @code{read}, these callbacks are
38353 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
38354 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
38355 @code{struct gdb_symbol_callbacks} has callbacks to create new object
38356 files and new symbol tables inside those object files. @code{struct
38357 gdb_unwind_callbacks} has callbacks to read registers off the current
38358 frame and to write out the values of the registers in the previous
38359 frame. Both have a callback (@code{target_read}) to read bytes off the
38360 target's address space.
38362 @node In-Process Agent
38363 @chapter In-Process Agent
38364 @cindex debugging agent
38365 The traditional debugging model is conceptually low-speed, but works fine,
38366 because most bugs can be reproduced in debugging-mode execution. However,
38367 as multi-core or many-core processors are becoming mainstream, and
38368 multi-threaded programs become more and more popular, there should be more
38369 and more bugs that only manifest themselves at normal-mode execution, for
38370 example, thread races, because debugger's interference with the program's
38371 timing may conceal the bugs. On the other hand, in some applications,
38372 it is not feasible for the debugger to interrupt the program's execution
38373 long enough for the developer to learn anything helpful about its behavior.
38374 If the program's correctness depends on its real-time behavior, delays
38375 introduced by a debugger might cause the program to fail, even when the
38376 code itself is correct. It is useful to be able to observe the program's
38377 behavior without interrupting it.
38379 Therefore, traditional debugging model is too intrusive to reproduce
38380 some bugs. In order to reduce the interference with the program, we can
38381 reduce the number of operations performed by debugger. The
38382 @dfn{In-Process Agent}, a shared library, is running within the same
38383 process with inferior, and is able to perform some debugging operations
38384 itself. As a result, debugger is only involved when necessary, and
38385 performance of debugging can be improved accordingly. Note that
38386 interference with program can be reduced but can't be removed completely,
38387 because the in-process agent will still stop or slow down the program.
38389 The in-process agent can interpret and execute Agent Expressions
38390 (@pxref{Agent Expressions}) during performing debugging operations. The
38391 agent expressions can be used for different purposes, such as collecting
38392 data in tracepoints, and condition evaluation in breakpoints.
38394 @anchor{Control Agent}
38395 You can control whether the in-process agent is used as an aid for
38396 debugging with the following commands:
38399 @kindex set agent on
38401 Causes the in-process agent to perform some operations on behalf of the
38402 debugger. Just which operations requested by the user will be done
38403 by the in-process agent depends on the its capabilities. For example,
38404 if you request to evaluate breakpoint conditions in the in-process agent,
38405 and the in-process agent has such capability as well, then breakpoint
38406 conditions will be evaluated in the in-process agent.
38408 @kindex set agent off
38409 @item set agent off
38410 Disables execution of debugging operations by the in-process agent. All
38411 of the operations will be performed by @value{GDBN}.
38415 Display the current setting of execution of debugging operations by
38416 the in-process agent.
38420 * In-Process Agent Protocol::
38423 @node In-Process Agent Protocol
38424 @section In-Process Agent Protocol
38425 @cindex in-process agent protocol
38427 The in-process agent is able to communicate with both @value{GDBN} and
38428 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
38429 used for communications between @value{GDBN} or GDBserver and the IPA.
38430 In general, @value{GDBN} or GDBserver sends commands
38431 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
38432 in-process agent replies back with the return result of the command, or
38433 some other information. The data sent to in-process agent is composed
38434 of primitive data types, such as 4-byte or 8-byte type, and composite
38435 types, which are called objects (@pxref{IPA Protocol Objects}).
38438 * IPA Protocol Objects::
38439 * IPA Protocol Commands::
38442 @node IPA Protocol Objects
38443 @subsection IPA Protocol Objects
38444 @cindex ipa protocol objects
38446 The commands sent to and results received from agent may contain some
38447 complex data types called @dfn{objects}.
38449 The in-process agent is running on the same machine with @value{GDBN}
38450 or GDBserver, so it doesn't have to handle as much differences between
38451 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
38452 However, there are still some differences of two ends in two processes:
38456 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
38457 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
38459 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
38460 GDBserver is compiled with one, and in-process agent is compiled with
38464 Here are the IPA Protocol Objects:
38468 agent expression object. It represents an agent expression
38469 (@pxref{Agent Expressions}).
38470 @anchor{agent expression object}
38472 tracepoint action object. It represents a tracepoint action
38473 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
38474 memory, static trace data and to evaluate expression.
38475 @anchor{tracepoint action object}
38477 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
38478 @anchor{tracepoint object}
38482 The following table describes important attributes of each IPA protocol
38485 @multitable @columnfractions .30 .20 .50
38486 @headitem Name @tab Size @tab Description
38487 @item @emph{agent expression object} @tab @tab
38488 @item length @tab 4 @tab length of bytes code
38489 @item byte code @tab @var{length} @tab contents of byte code
38490 @item @emph{tracepoint action for collecting memory} @tab @tab
38491 @item 'M' @tab 1 @tab type of tracepoint action
38492 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
38493 address of the lowest byte to collect, otherwise @var{addr} is the offset
38494 of @var{basereg} for memory collecting.
38495 @item len @tab 8 @tab length of memory for collecting
38496 @item basereg @tab 4 @tab the register number containing the starting
38497 memory address for collecting.
38498 @item @emph{tracepoint action for collecting registers} @tab @tab
38499 @item 'R' @tab 1 @tab type of tracepoint action
38500 @item @emph{tracepoint action for collecting static trace data} @tab @tab
38501 @item 'L' @tab 1 @tab type of tracepoint action
38502 @item @emph{tracepoint action for expression evaluation} @tab @tab
38503 @item 'X' @tab 1 @tab type of tracepoint action
38504 @item agent expression @tab length of @tab @ref{agent expression object}
38505 @item @emph{tracepoint object} @tab @tab
38506 @item number @tab 4 @tab number of tracepoint
38507 @item address @tab 8 @tab address of tracepoint inserted on
38508 @item type @tab 4 @tab type of tracepoint
38509 @item enabled @tab 1 @tab enable or disable of tracepoint
38510 @item step_count @tab 8 @tab step
38511 @item pass_count @tab 8 @tab pass
38512 @item numactions @tab 4 @tab number of tracepoint actions
38513 @item hit count @tab 8 @tab hit count
38514 @item trace frame usage @tab 8 @tab trace frame usage
38515 @item compiled_cond @tab 8 @tab compiled condition
38516 @item orig_size @tab 8 @tab orig size
38517 @item condition @tab 4 if condition is NULL otherwise length of
38518 @ref{agent expression object}
38519 @tab zero if condition is NULL, otherwise is
38520 @ref{agent expression object}
38521 @item actions @tab variable
38522 @tab numactions number of @ref{tracepoint action object}
38525 @node IPA Protocol Commands
38526 @subsection IPA Protocol Commands
38527 @cindex ipa protocol commands
38529 The spaces in each command are delimiters to ease reading this commands
38530 specification. They don't exist in real commands.
38534 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
38535 Installs a new fast tracepoint described by @var{tracepoint_object}
38536 (@pxref{tracepoint object}). The @var{gdb_jump_pad_head}, 8-byte long, is the
38537 head of @dfn{jumppad}, which is used to jump to data collection routine
38542 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
38543 @var{target_address} is address of tracepoint in the inferior.
38544 The @var{gdb_jump_pad_head} is updated head of jumppad. Both of
38545 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
38546 The @var{fjump} contains a sequence of instructions jump to jumppad entry.
38547 The @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
38554 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
38555 is about to kill inferiors.
38563 @item probe_marker_at:@var{address}
38564 Asks in-process agent to probe the marker at @var{address}.
38571 @item unprobe_marker_at:@var{address}
38572 Asks in-process agent to unprobe the marker at @var{address}.
38576 @chapter Reporting Bugs in @value{GDBN}
38577 @cindex bugs in @value{GDBN}
38578 @cindex reporting bugs in @value{GDBN}
38580 Your bug reports play an essential role in making @value{GDBN} reliable.
38582 Reporting a bug may help you by bringing a solution to your problem, or it
38583 may not. But in any case the principal function of a bug report is to help
38584 the entire community by making the next version of @value{GDBN} work better. Bug
38585 reports are your contribution to the maintenance of @value{GDBN}.
38587 In order for a bug report to serve its purpose, you must include the
38588 information that enables us to fix the bug.
38591 * Bug Criteria:: Have you found a bug?
38592 * Bug Reporting:: How to report bugs
38596 @section Have You Found a Bug?
38597 @cindex bug criteria
38599 If you are not sure whether you have found a bug, here are some guidelines:
38602 @cindex fatal signal
38603 @cindex debugger crash
38604 @cindex crash of debugger
38606 If the debugger gets a fatal signal, for any input whatever, that is a
38607 @value{GDBN} bug. Reliable debuggers never crash.
38609 @cindex error on valid input
38611 If @value{GDBN} produces an error message for valid input, that is a
38612 bug. (Note that if you're cross debugging, the problem may also be
38613 somewhere in the connection to the target.)
38615 @cindex invalid input
38617 If @value{GDBN} does not produce an error message for invalid input,
38618 that is a bug. However, you should note that your idea of
38619 ``invalid input'' might be our idea of ``an extension'' or ``support
38620 for traditional practice''.
38623 If you are an experienced user of debugging tools, your suggestions
38624 for improvement of @value{GDBN} are welcome in any case.
38627 @node Bug Reporting
38628 @section How to Report Bugs
38629 @cindex bug reports
38630 @cindex @value{GDBN} bugs, reporting
38632 A number of companies and individuals offer support for @sc{gnu} products.
38633 If you obtained @value{GDBN} from a support organization, we recommend you
38634 contact that organization first.
38636 You can find contact information for many support companies and
38637 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
38639 @c should add a web page ref...
38642 @ifset BUGURL_DEFAULT
38643 In any event, we also recommend that you submit bug reports for
38644 @value{GDBN}. The preferred method is to submit them directly using
38645 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
38646 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
38649 @strong{Do not send bug reports to @samp{info-gdb}, or to
38650 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
38651 not want to receive bug reports. Those that do have arranged to receive
38654 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
38655 serves as a repeater. The mailing list and the newsgroup carry exactly
38656 the same messages. Often people think of posting bug reports to the
38657 newsgroup instead of mailing them. This appears to work, but it has one
38658 problem which can be crucial: a newsgroup posting often lacks a mail
38659 path back to the sender. Thus, if we need to ask for more information,
38660 we may be unable to reach you. For this reason, it is better to send
38661 bug reports to the mailing list.
38663 @ifclear BUGURL_DEFAULT
38664 In any event, we also recommend that you submit bug reports for
38665 @value{GDBN} to @value{BUGURL}.
38669 The fundamental principle of reporting bugs usefully is this:
38670 @strong{report all the facts}. If you are not sure whether to state a
38671 fact or leave it out, state it!
38673 Often people omit facts because they think they know what causes the
38674 problem and assume that some details do not matter. Thus, you might
38675 assume that the name of the variable you use in an example does not matter.
38676 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
38677 stray memory reference which happens to fetch from the location where that
38678 name is stored in memory; perhaps, if the name were different, the contents
38679 of that location would fool the debugger into doing the right thing despite
38680 the bug. Play it safe and give a specific, complete example. That is the
38681 easiest thing for you to do, and the most helpful.
38683 Keep in mind that the purpose of a bug report is to enable us to fix the
38684 bug. It may be that the bug has been reported previously, but neither
38685 you nor we can know that unless your bug report is complete and
38688 Sometimes people give a few sketchy facts and ask, ``Does this ring a
38689 bell?'' Those bug reports are useless, and we urge everyone to
38690 @emph{refuse to respond to them} except to chide the sender to report
38693 To enable us to fix the bug, you should include all these things:
38697 The version of @value{GDBN}. @value{GDBN} announces it if you start
38698 with no arguments; you can also print it at any time using @code{show
38701 Without this, we will not know whether there is any point in looking for
38702 the bug in the current version of @value{GDBN}.
38705 The type of machine you are using, and the operating system name and
38709 The details of the @value{GDBN} build-time configuration.
38710 @value{GDBN} shows these details if you invoke it with the
38711 @option{--configuration} command-line option, or if you type
38712 @code{show configuration} at @value{GDBN}'s prompt.
38715 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
38716 ``@value{GCC}--2.8.1''.
38719 What compiler (and its version) was used to compile the program you are
38720 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
38721 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
38722 to get this information; for other compilers, see the documentation for
38726 The command arguments you gave the compiler to compile your example and
38727 observe the bug. For example, did you use @samp{-O}? To guarantee
38728 you will not omit something important, list them all. A copy of the
38729 Makefile (or the output from make) is sufficient.
38731 If we were to try to guess the arguments, we would probably guess wrong
38732 and then we might not encounter the bug.
38735 A complete input script, and all necessary source files, that will
38739 A description of what behavior you observe that you believe is
38740 incorrect. For example, ``It gets a fatal signal.''
38742 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
38743 will certainly notice it. But if the bug is incorrect output, we might
38744 not notice unless it is glaringly wrong. You might as well not give us
38745 a chance to make a mistake.
38747 Even if the problem you experience is a fatal signal, you should still
38748 say so explicitly. Suppose something strange is going on, such as, your
38749 copy of @value{GDBN} is out of synch, or you have encountered a bug in
38750 the C library on your system. (This has happened!) Your copy might
38751 crash and ours would not. If you told us to expect a crash, then when
38752 ours fails to crash, we would know that the bug was not happening for
38753 us. If you had not told us to expect a crash, then we would not be able
38754 to draw any conclusion from our observations.
38757 @cindex recording a session script
38758 To collect all this information, you can use a session recording program
38759 such as @command{script}, which is available on many Unix systems.
38760 Just run your @value{GDBN} session inside @command{script} and then
38761 include the @file{typescript} file with your bug report.
38763 Another way to record a @value{GDBN} session is to run @value{GDBN}
38764 inside Emacs and then save the entire buffer to a file.
38767 If you wish to suggest changes to the @value{GDBN} source, send us context
38768 diffs. If you even discuss something in the @value{GDBN} source, refer to
38769 it by context, not by line number.
38771 The line numbers in our development sources will not match those in your
38772 sources. Your line numbers would convey no useful information to us.
38776 Here are some things that are not necessary:
38780 A description of the envelope of the bug.
38782 Often people who encounter a bug spend a lot of time investigating
38783 which changes to the input file will make the bug go away and which
38784 changes will not affect it.
38786 This is often time consuming and not very useful, because the way we
38787 will find the bug is by running a single example under the debugger
38788 with breakpoints, not by pure deduction from a series of examples.
38789 We recommend that you save your time for something else.
38791 Of course, if you can find a simpler example to report @emph{instead}
38792 of the original one, that is a convenience for us. Errors in the
38793 output will be easier to spot, running under the debugger will take
38794 less time, and so on.
38796 However, simplification is not vital; if you do not want to do this,
38797 report the bug anyway and send us the entire test case you used.
38800 A patch for the bug.
38802 A patch for the bug does help us if it is a good one. But do not omit
38803 the necessary information, such as the test case, on the assumption that
38804 a patch is all we need. We might see problems with your patch and decide
38805 to fix the problem another way, or we might not understand it at all.
38807 Sometimes with a program as complicated as @value{GDBN} it is very hard to
38808 construct an example that will make the program follow a certain path
38809 through the code. If you do not send us the example, we will not be able
38810 to construct one, so we will not be able to verify that the bug is fixed.
38812 And if we cannot understand what bug you are trying to fix, or why your
38813 patch should be an improvement, we will not install it. A test case will
38814 help us to understand.
38817 A guess about what the bug is or what it depends on.
38819 Such guesses are usually wrong. Even we cannot guess right about such
38820 things without first using the debugger to find the facts.
38823 @c The readline documentation is distributed with the readline code
38824 @c and consists of the two following files:
38827 @c Use -I with makeinfo to point to the appropriate directory,
38828 @c environment var TEXINPUTS with TeX.
38829 @ifclear SYSTEM_READLINE
38830 @include rluser.texi
38831 @include hsuser.texi
38835 @appendix In Memoriam
38837 The @value{GDBN} project mourns the loss of the following long-time
38842 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
38843 to Free Software in general. Outside of @value{GDBN}, he was known in
38844 the Amiga world for his series of Fish Disks, and the GeekGadget project.
38846 @item Michael Snyder
38847 Michael was one of the Global Maintainers of the @value{GDBN} project,
38848 with contributions recorded as early as 1996, until 2011. In addition
38849 to his day to day participation, he was a large driving force behind
38850 adding Reverse Debugging to @value{GDBN}.
38853 Beyond their technical contributions to the project, they were also
38854 enjoyable members of the Free Software Community. We will miss them.
38856 @node Formatting Documentation
38857 @appendix Formatting Documentation
38859 @cindex @value{GDBN} reference card
38860 @cindex reference card
38861 The @value{GDBN} 4 release includes an already-formatted reference card, ready
38862 for printing with PostScript or Ghostscript, in the @file{gdb}
38863 subdirectory of the main source directory@footnote{In
38864 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
38865 release.}. If you can use PostScript or Ghostscript with your printer,
38866 you can print the reference card immediately with @file{refcard.ps}.
38868 The release also includes the source for the reference card. You
38869 can format it, using @TeX{}, by typing:
38875 The @value{GDBN} reference card is designed to print in @dfn{landscape}
38876 mode on US ``letter'' size paper;
38877 that is, on a sheet 11 inches wide by 8.5 inches
38878 high. You will need to specify this form of printing as an option to
38879 your @sc{dvi} output program.
38881 @cindex documentation
38883 All the documentation for @value{GDBN} comes as part of the machine-readable
38884 distribution. The documentation is written in Texinfo format, which is
38885 a documentation system that uses a single source file to produce both
38886 on-line information and a printed manual. You can use one of the Info
38887 formatting commands to create the on-line version of the documentation
38888 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
38890 @value{GDBN} includes an already formatted copy of the on-line Info
38891 version of this manual in the @file{gdb} subdirectory. The main Info
38892 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
38893 subordinate files matching @samp{gdb.info*} in the same directory. If
38894 necessary, you can print out these files, or read them with any editor;
38895 but they are easier to read using the @code{info} subsystem in @sc{gnu}
38896 Emacs or the standalone @code{info} program, available as part of the
38897 @sc{gnu} Texinfo distribution.
38899 If you want to format these Info files yourself, you need one of the
38900 Info formatting programs, such as @code{texinfo-format-buffer} or
38903 If you have @code{makeinfo} installed, and are in the top level
38904 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
38905 version @value{GDBVN}), you can make the Info file by typing:
38912 If you want to typeset and print copies of this manual, you need @TeX{},
38913 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
38914 Texinfo definitions file.
38916 @TeX{} is a typesetting program; it does not print files directly, but
38917 produces output files called @sc{dvi} files. To print a typeset
38918 document, you need a program to print @sc{dvi} files. If your system
38919 has @TeX{} installed, chances are it has such a program. The precise
38920 command to use depends on your system; @kbd{lpr -d} is common; another
38921 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
38922 require a file name without any extension or a @samp{.dvi} extension.
38924 @TeX{} also requires a macro definitions file called
38925 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
38926 written in Texinfo format. On its own, @TeX{} cannot either read or
38927 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
38928 and is located in the @file{gdb-@var{version-number}/texinfo}
38931 If you have @TeX{} and a @sc{dvi} printer program installed, you can
38932 typeset and print this manual. First switch to the @file{gdb}
38933 subdirectory of the main source directory (for example, to
38934 @file{gdb-@value{GDBVN}/gdb}) and type:
38940 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
38942 @node Installing GDB
38943 @appendix Installing @value{GDBN}
38944 @cindex installation
38947 * Requirements:: Requirements for building @value{GDBN}
38948 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
38949 * Separate Objdir:: Compiling @value{GDBN} in another directory
38950 * Config Names:: Specifying names for hosts and targets
38951 * Configure Options:: Summary of options for configure
38952 * System-wide configuration:: Having a system-wide init file
38956 @section Requirements for Building @value{GDBN}
38957 @cindex building @value{GDBN}, requirements for
38959 Building @value{GDBN} requires various tools and packages to be available.
38960 Other packages will be used only if they are found.
38962 @heading Tools/Packages Necessary for Building @value{GDBN}
38964 @item C@t{++}11 compiler
38965 @value{GDBN} is written in C@t{++}11. It should be buildable with any
38966 recent C@t{++}11 compiler, e.g.@: GCC.
38969 @value{GDBN}'s build system relies on features only found in the GNU
38970 make program. Other variants of @code{make} will not work.
38973 @heading Tools/Packages Optional for Building @value{GDBN}
38977 @value{GDBN} can use the Expat XML parsing library. This library may be
38978 included with your operating system distribution; if it is not, you
38979 can get the latest version from @url{http://expat.sourceforge.net}.
38980 The @file{configure} script will search for this library in several
38981 standard locations; if it is installed in an unusual path, you can
38982 use the @option{--with-libexpat-prefix} option to specify its location.
38988 Remote protocol memory maps (@pxref{Memory Map Format})
38990 Target descriptions (@pxref{Target Descriptions})
38992 Remote shared library lists (@xref{Library List Format},
38993 or alternatively @pxref{Library List Format for SVR4 Targets})
38995 MS-Windows shared libraries (@pxref{Shared Libraries})
38997 Traceframe info (@pxref{Traceframe Info Format})
38999 Branch trace (@pxref{Branch Trace Format},
39000 @pxref{Branch Trace Configuration Format})
39004 @value{GDBN} can be scripted using GNU Guile. @xref{Guile}. By
39005 default, @value{GDBN} will be compiled if the Guile libraries are
39006 installed and are found by @file{configure}. You can use the
39007 @code{--with-guile} option to request Guile, and pass either the Guile
39008 version number or the file name of the relevant @code{pkg-config}
39009 program to choose a particular version of Guile.
39012 @value{GDBN}'s features related to character sets (@pxref{Character
39013 Sets}) require a functioning @code{iconv} implementation. If you are
39014 on a GNU system, then this is provided by the GNU C Library. Some
39015 other systems also provide a working @code{iconv}.
39017 If @value{GDBN} is using the @code{iconv} program which is installed
39018 in a non-standard place, you will need to tell @value{GDBN} where to
39019 find it. This is done with @option{--with-iconv-bin} which specifies
39020 the directory that contains the @code{iconv} program. This program is
39021 run in order to make a list of the available character sets.
39023 On systems without @code{iconv}, you can install GNU Libiconv. If
39024 Libiconv is installed in a standard place, @value{GDBN} will
39025 automatically use it if it is needed. If you have previously
39026 installed Libiconv in a non-standard place, you can use the
39027 @option{--with-libiconv-prefix} option to @file{configure}.
39029 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
39030 arrange to build Libiconv if a directory named @file{libiconv} appears
39031 in the top-most source directory. If Libiconv is built this way, and
39032 if the operating system does not provide a suitable @code{iconv}
39033 implementation, then the just-built library will automatically be used
39034 by @value{GDBN}. One easy way to set this up is to download GNU
39035 Libiconv, unpack it inside the top-level directory of the @value{GDBN}
39036 source tree, and then rename the directory holding the Libiconv source
39037 code to @samp{libiconv}.
39040 @value{GDBN} can support debugging sections that are compressed with
39041 the LZMA library. @xref{MiniDebugInfo}. If this library is not
39042 included with your operating system, you can find it in the xz package
39043 at @url{http://tukaani.org/xz/}. If the LZMA library is available in
39044 the usual place, then the @file{configure} script will use it
39045 automatically. If it is installed in an unusual path, you can use the
39046 @option{--with-lzma-prefix} option to specify its location.
39050 @value{GDBN} can use the GNU MPFR multiple-precision floating-point
39051 library. This library may be included with your operating system
39052 distribution; if it is not, you can get the latest version from
39053 @url{http://www.mpfr.org}. The @file{configure} script will search
39054 for this library in several standard locations; if it is installed
39055 in an unusual path, you can use the @option{--with-libmpfr-prefix}
39056 option to specify its location.
39058 GNU MPFR is used to emulate target floating-point arithmetic during
39059 expression evaluation when the target uses different floating-point
39060 formats than the host. If GNU MPFR it is not available, @value{GDBN}
39061 will fall back to using host floating-point arithmetic.
39064 @value{GDBN} can be scripted using Python language. @xref{Python}.
39065 By default, @value{GDBN} will be compiled if the Python libraries are
39066 installed and are found by @file{configure}. You can use the
39067 @code{--with-python} option to request Python, and pass either the
39068 file name of the relevant @code{python} executable, or the name of the
39069 directory in which Python is installed, to choose a particular
39070 installation of Python.
39073 @cindex compressed debug sections
39074 @value{GDBN} will use the @samp{zlib} library, if available, to read
39075 compressed debug sections. Some linkers, such as GNU gold, are capable
39076 of producing binaries with compressed debug sections. If @value{GDBN}
39077 is compiled with @samp{zlib}, it will be able to read the debug
39078 information in such binaries.
39080 The @samp{zlib} library is likely included with your operating system
39081 distribution; if it is not, you can get the latest version from
39082 @url{http://zlib.net}.
39085 @node Running Configure
39086 @section Invoking the @value{GDBN} @file{configure} Script
39087 @cindex configuring @value{GDBN}
39088 @value{GDBN} comes with a @file{configure} script that automates the process
39089 of preparing @value{GDBN} for installation; you can then use @code{make} to
39090 build the @code{gdb} program.
39092 @c irrelevant in info file; it's as current as the code it lives with.
39093 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
39094 look at the @file{README} file in the sources; we may have improved the
39095 installation procedures since publishing this manual.}
39098 The @value{GDBN} distribution includes all the source code you need for
39099 @value{GDBN} in a single directory, whose name is usually composed by
39100 appending the version number to @samp{gdb}.
39102 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
39103 @file{gdb-@value{GDBVN}} directory. That directory contains:
39106 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
39107 script for configuring @value{GDBN} and all its supporting libraries
39109 @item gdb-@value{GDBVN}/gdb
39110 the source specific to @value{GDBN} itself
39112 @item gdb-@value{GDBVN}/bfd
39113 source for the Binary File Descriptor library
39115 @item gdb-@value{GDBVN}/include
39116 @sc{gnu} include files
39118 @item gdb-@value{GDBVN}/libiberty
39119 source for the @samp{-liberty} free software library
39121 @item gdb-@value{GDBVN}/opcodes
39122 source for the library of opcode tables and disassemblers
39124 @item gdb-@value{GDBVN}/readline
39125 source for the @sc{gnu} command-line interface
39128 There may be other subdirectories as well.
39130 The simplest way to configure and build @value{GDBN} is to run @file{configure}
39131 from the @file{gdb-@var{version-number}} source directory, which in
39132 this example is the @file{gdb-@value{GDBVN}} directory.
39134 First switch to the @file{gdb-@var{version-number}} source directory
39135 if you are not already in it; then run @file{configure}. Pass the
39136 identifier for the platform on which @value{GDBN} will run as an
39142 cd gdb-@value{GDBVN}
39147 Running @samp{configure} and then running @code{make} builds the
39148 included supporting libraries, then @code{gdb} itself. The configured
39149 source files, and the binaries, are left in the corresponding source
39153 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
39154 system does not recognize this automatically when you run a different
39155 shell, you may need to run @code{sh} on it explicitly:
39161 You should run the @file{configure} script from the top directory in the
39162 source tree, the @file{gdb-@var{version-number}} directory. If you run
39163 @file{configure} from one of the subdirectories, you will configure only
39164 that subdirectory. That is usually not what you want. In particular,
39165 if you run the first @file{configure} from the @file{gdb} subdirectory
39166 of the @file{gdb-@var{version-number}} directory, you will omit the
39167 configuration of @file{bfd}, @file{readline}, and other sibling
39168 directories of the @file{gdb} subdirectory. This leads to build errors
39169 about missing include files such as @file{bfd/bfd.h}.
39171 You can install @code{@value{GDBN}} anywhere. The best way to do this
39172 is to pass the @code{--prefix} option to @code{configure}, and then
39173 install it with @code{make install}.
39175 @node Separate Objdir
39176 @section Compiling @value{GDBN} in Another Directory
39178 If you want to run @value{GDBN} versions for several host or target machines,
39179 you need a different @code{gdb} compiled for each combination of
39180 host and target. @file{configure} is designed to make this easy by
39181 allowing you to generate each configuration in a separate subdirectory,
39182 rather than in the source directory. If your @code{make} program
39183 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
39184 @code{make} in each of these directories builds the @code{gdb}
39185 program specified there.
39187 To build @code{gdb} in a separate directory, run @file{configure}
39188 with the @samp{--srcdir} option to specify where to find the source.
39189 (You also need to specify a path to find @file{configure}
39190 itself from your working directory. If the path to @file{configure}
39191 would be the same as the argument to @samp{--srcdir}, you can leave out
39192 the @samp{--srcdir} option; it is assumed.)
39194 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
39195 separate directory for a Sun 4 like this:
39199 cd gdb-@value{GDBVN}
39202 ../gdb-@value{GDBVN}/configure
39207 When @file{configure} builds a configuration using a remote source
39208 directory, it creates a tree for the binaries with the same structure
39209 (and using the same names) as the tree under the source directory. In
39210 the example, you'd find the Sun 4 library @file{libiberty.a} in the
39211 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
39212 @file{gdb-sun4/gdb}.
39214 Make sure that your path to the @file{configure} script has just one
39215 instance of @file{gdb} in it. If your path to @file{configure} looks
39216 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
39217 one subdirectory of @value{GDBN}, not the whole package. This leads to
39218 build errors about missing include files such as @file{bfd/bfd.h}.
39220 One popular reason to build several @value{GDBN} configurations in separate
39221 directories is to configure @value{GDBN} for cross-compiling (where
39222 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
39223 programs that run on another machine---the @dfn{target}).
39224 You specify a cross-debugging target by
39225 giving the @samp{--target=@var{target}} option to @file{configure}.
39227 When you run @code{make} to build a program or library, you must run
39228 it in a configured directory---whatever directory you were in when you
39229 called @file{configure} (or one of its subdirectories).
39231 The @code{Makefile} that @file{configure} generates in each source
39232 directory also runs recursively. If you type @code{make} in a source
39233 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
39234 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
39235 will build all the required libraries, and then build GDB.
39237 When you have multiple hosts or targets configured in separate
39238 directories, you can run @code{make} on them in parallel (for example,
39239 if they are NFS-mounted on each of the hosts); they will not interfere
39243 @section Specifying Names for Hosts and Targets
39245 The specifications used for hosts and targets in the @file{configure}
39246 script are based on a three-part naming scheme, but some short predefined
39247 aliases are also supported. The full naming scheme encodes three pieces
39248 of information in the following pattern:
39251 @var{architecture}-@var{vendor}-@var{os}
39254 For example, you can use the alias @code{sun4} as a @var{host} argument,
39255 or as the value for @var{target} in a @code{--target=@var{target}}
39256 option. The equivalent full name is @samp{sparc-sun-sunos4}.
39258 The @file{configure} script accompanying @value{GDBN} does not provide
39259 any query facility to list all supported host and target names or
39260 aliases. @file{configure} calls the Bourne shell script
39261 @code{config.sub} to map abbreviations to full names; you can read the
39262 script, if you wish, or you can use it to test your guesses on
39263 abbreviations---for example:
39266 % sh config.sub i386-linux
39268 % sh config.sub alpha-linux
39269 alpha-unknown-linux-gnu
39270 % sh config.sub hp9k700
39272 % sh config.sub sun4
39273 sparc-sun-sunos4.1.1
39274 % sh config.sub sun3
39275 m68k-sun-sunos4.1.1
39276 % sh config.sub i986v
39277 Invalid configuration `i986v': machine `i986v' not recognized
39281 @code{config.sub} is also distributed in the @value{GDBN} source
39282 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
39284 @node Configure Options
39285 @section @file{configure} Options
39287 Here is a summary of the @file{configure} options and arguments that
39288 are most often useful for building @value{GDBN}. @file{configure}
39289 also has several other options not listed here. @inforef{Running
39290 configure scripts,,autoconf.info}, for a full
39291 explanation of @file{configure}.
39294 configure @r{[}--help@r{]}
39295 @r{[}--prefix=@var{dir}@r{]}
39296 @r{[}--exec-prefix=@var{dir}@r{]}
39297 @r{[}--srcdir=@var{dirname}@r{]}
39298 @r{[}--target=@var{target}@r{]}
39302 You may introduce options with a single @samp{-} rather than
39303 @samp{--} if you prefer; but you may abbreviate option names if you use
39308 Display a quick summary of how to invoke @file{configure}.
39310 @item --prefix=@var{dir}
39311 Configure the source to install programs and files under directory
39314 @item --exec-prefix=@var{dir}
39315 Configure the source to install programs under directory
39318 @c avoid splitting the warning from the explanation:
39320 @item --srcdir=@var{dirname}
39321 Use this option to make configurations in directories separate from the
39322 @value{GDBN} source directories. Among other things, you can use this to
39323 build (or maintain) several configurations simultaneously, in separate
39324 directories. @file{configure} writes configuration-specific files in
39325 the current directory, but arranges for them to use the source in the
39326 directory @var{dirname}. @file{configure} creates directories under
39327 the working directory in parallel to the source directories below
39330 @item --target=@var{target}
39331 Configure @value{GDBN} for cross-debugging programs running on the specified
39332 @var{target}. Without this option, @value{GDBN} is configured to debug
39333 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
39335 There is no convenient way to generate a list of all available
39336 targets. Also see the @code{--enable-targets} option, below.
39339 There are many other options that are specific to @value{GDBN}. This
39340 lists just the most common ones; there are some very specialized
39341 options not described here.
39344 @item --enable-targets=@r{[}@var{target}@r{]}@dots{}
39345 @itemx --enable-targets=all
39346 Configure @value{GDBN} for cross-debugging programs running on the
39347 specified list of targets. The special value @samp{all} configures
39348 @value{GDBN} for debugging programs running on any target it supports.
39350 @item --with-gdb-datadir=@var{path}
39351 Set the @value{GDBN}-specific data directory. @value{GDBN} will look
39352 here for certain supporting files or scripts. This defaults to the
39353 @file{gdb} subdirectory of @samp{datadir} (which can be set using
39356 @item --with-relocated-sources=@var{dir}
39357 Sets up the default source path substitution rule so that directory
39358 names recorded in debug information will be automatically adjusted for
39359 any directory under @var{dir}. @var{dir} should be a subdirectory of
39360 @value{GDBN}'s configured prefix, the one mentioned in the
39361 @code{--prefix} or @code{--exec-prefix} options to configure. This
39362 option is useful if GDB is supposed to be moved to a different place
39365 @item --enable-64-bit-bfd
39366 Enable 64-bit support in BFD on 32-bit hosts.
39368 @item --disable-gdbmi
39369 Build @value{GDBN} without the GDB/MI machine interface
39373 Build @value{GDBN} with the text-mode full-screen user interface
39374 (TUI). Requires a curses library (ncurses and cursesX are also
39377 @item --with-curses
39378 Use the curses library instead of the termcap library, for text-mode
39379 terminal operations.
39381 @item --with-libunwind-ia64
39382 Use the libunwind library for unwinding function call stack on ia64
39383 target platforms. See http://www.nongnu.org/libunwind/index.html for
39386 @item --with-system-readline
39387 Use the readline library installed on the host, rather than the
39388 library supplied as part of @value{GDBN}. Readline 7 or newer is
39389 required; this is enforced by the build system.
39391 @item --with-system-zlib
39392 Use the zlib library installed on the host, rather than the library
39393 supplied as part of @value{GDBN}.
39396 Build @value{GDBN} with Expat, a library for XML parsing. (Done by
39397 default if libexpat is installed and found at configure time.) This
39398 library is used to read XML files supplied with @value{GDBN}. If it
39399 is unavailable, some features, such as remote protocol memory maps,
39400 target descriptions, and shared library lists, that are based on XML
39401 files, will not be available in @value{GDBN}. If your host does not
39402 have libexpat installed, you can get the latest version from
39403 `http://expat.sourceforge.net'.
39405 @item --with-libiconv-prefix@r{[}=@var{dir}@r{]}
39407 Build @value{GDBN} with GNU libiconv, a character set encoding
39408 conversion library. This is not done by default, as on GNU systems
39409 the @code{iconv} that is built in to the C library is sufficient. If
39410 your host does not have a working @code{iconv}, you can get the latest
39411 version of GNU iconv from `https://www.gnu.org/software/libiconv/'.
39413 @value{GDBN}'s build system also supports building GNU libiconv as
39414 part of the overall build. @xref{Requirements}.
39417 Build @value{GDBN} with LZMA, a compression library. (Done by default
39418 if liblzma is installed and found at configure time.) LZMA is used by
39419 @value{GDBN}'s "mini debuginfo" feature, which is only useful on
39420 platforms using the ELF object file format. If your host does not
39421 have liblzma installed, you can get the latest version from
39422 `https://tukaani.org/xz/'.
39425 Build @value{GDBN} with GNU MPFR, a library for multiple-precision
39426 floating-point computation with correct rounding. (Done by default if
39427 GNU MPFR is installed and found at configure time.) This library is
39428 used to emulate target floating-point arithmetic during expression
39429 evaluation when the target uses different floating-point formats than
39430 the host. If GNU MPFR is not available, @value{GDBN} will fall back
39431 to using host floating-point arithmetic. If your host does not have
39432 GNU MPFR installed, you can get the latest version from
39433 `http://www.mpfr.org'.
39435 @item --with-python@r{[}=@var{python}@r{]}
39436 Build @value{GDBN} with Python scripting support. (Done by default if
39437 libpython is present and found at configure time.) Python makes
39438 @value{GDBN} scripting much more powerful than the restricted CLI
39439 scripting language. If your host does not have Python installed, you
39440 can find it on `http://www.python.org/download/'. The oldest version
39441 of Python supported by GDB is 2.6. The optional argument @var{python}
39442 is used to find the Python headers and libraries. It can be either
39443 the name of a Python executable, or the name of the directory in which
39444 Python is installed.
39446 @item --with-guile[=GUILE]'
39447 Build @value{GDBN} with GNU Guile scripting support. (Done by default
39448 if libguile is present and found at configure time.) If your host
39449 does not have Guile installed, you can find it at
39450 `https://www.gnu.org/software/guile/'. The optional argument GUILE
39451 can be a version number, which will cause @code{configure} to try to
39452 use that version of Guile; or the file name of a @code{pkg-config}
39453 executable, which will be queried to find the information needed to
39454 compile and link against Guile.
39456 @item --without-included-regex
39457 Don't use the regex library included with @value{GDBN} (as part of the
39458 libiberty library). This is the default on hosts with version 2 of
39461 @item --with-sysroot=@var{dir}
39462 Use @var{dir} as the default system root directory for libraries whose
39463 file names begin with @file{/lib}' or @file{/usr/lib'}. (The value of
39464 @var{dir} can be modified at run time by using the @command{set
39465 sysroot} command.) If @var{dir} is under the @value{GDBN} configured
39466 prefix (set with @code{--prefix} or @code{--exec-prefix options}, the
39467 default system root will be automatically adjusted if and when
39468 @value{GDBN} is moved to a different location.
39470 @item --with-system-gdbinit=@var{file}
39471 Configure @value{GDBN} to automatically load a system-wide init file.
39472 @var{file} should be an absolute file name. If @var{file} is in a
39473 directory under the configured prefix, and @value{GDBN} is moved to
39474 another location after being built, the location of the system-wide
39475 init file will be adjusted accordingly.
39477 @item --with-system-gdbinit-dir=@var{directory}
39478 Configure @value{GDBN} to automatically load init files from a
39479 system-wide directory. @var{directory} should be an absolute directory
39480 name. If @var{directory} is in a directory under the configured
39481 prefix, and @value{GDBN} is moved to another location after being
39482 built, the location of the system-wide init directory will be
39483 adjusted accordingly.
39485 @item --enable-build-warnings
39486 When building the @value{GDBN} sources, ask the compiler to warn about
39487 any code which looks even vaguely suspicious. It passes many
39488 different warning flags, depending on the exact version of the
39489 compiler you are using.
39491 @item --enable-werror
39492 Treat compiler warnings as werrors. It adds the @code{-Werror} flag
39493 to the compiler, which will fail the compilation if the compiler
39494 outputs any warning messages.
39496 @item --enable-ubsan
39497 Enable the GCC undefined behavior sanitizer. This is disabled by
39498 default, but passing @code{--enable-ubsan=yes} or
39499 @code{--enable-ubsan=auto} to @code{configure} will enable it. The
39500 undefined behavior sanitizer checks for C@t{++} undefined behavior.
39501 It has a performance cost, so if you are looking at @value{GDBN}'s
39502 performance, you should disable it. The undefined behavior sanitizer
39503 was first introduced in GCC 4.9.
39506 @node System-wide configuration
39507 @section System-wide configuration and settings
39508 @cindex system-wide init file
39510 @value{GDBN} can be configured to have a system-wide init file and a
39511 system-wide init file directory; this file and files in that directory
39512 (if they have a recognized file extension) will be read and executed at
39513 startup (@pxref{Startup, , What @value{GDBN} does during startup}).
39515 Here are the corresponding configure options:
39518 @item --with-system-gdbinit=@var{file}
39519 Specify that the default location of the system-wide init file is
39521 @item --with-system-gdbinit-dir=@var{directory}
39522 Specify that the default location of the system-wide init file directory
39523 is @var{directory}.
39526 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
39527 they may be subject to relocation. Two possible cases:
39531 If the default location of this init file/directory contains @file{$prefix},
39532 it will be subject to relocation. Suppose that the configure options
39533 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
39534 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
39535 init file is looked for as @file{$install/etc/gdbinit} instead of
39536 @file{$prefix/etc/gdbinit}.
39539 By contrast, if the default location does not contain the prefix,
39540 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
39541 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
39542 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
39543 wherever @value{GDBN} is installed.
39546 If the configured location of the system-wide init file (as given by the
39547 @option{--with-system-gdbinit} option at configure time) is in the
39548 data-directory (as specified by @option{--with-gdb-datadir} at configure
39549 time) or in one of its subdirectories, then @value{GDBN} will look for the
39550 system-wide init file in the directory specified by the
39551 @option{--data-directory} command-line option.
39552 Note that the system-wide init file is only read once, during @value{GDBN}
39553 initialization. If the data-directory is changed after @value{GDBN} has
39554 started with the @code{set data-directory} command, the file will not be
39557 This applies similarly to the system-wide directory specified in
39558 @option{--with-system-gdbinit-dir}.
39560 Any supported scripting language can be used for these init files, as long
39561 as the file extension matches the scripting language. To be interpreted
39562 as regular @value{GDBN} commands, the files needs to have a @file{.gdb}
39566 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
39569 @node System-wide Configuration Scripts
39570 @subsection Installed System-wide Configuration Scripts
39571 @cindex system-wide configuration scripts
39573 The @file{system-gdbinit} directory, located inside the data-directory
39574 (as specified by @option{--with-gdb-datadir} at configure time) contains
39575 a number of scripts which can be used as system-wide init files. To
39576 automatically source those scripts at startup, @value{GDBN} should be
39577 configured with @option{--with-system-gdbinit}. Otherwise, any user
39578 should be able to source them by hand as needed.
39580 The following scripts are currently available:
39583 @item @file{elinos.py}
39585 @cindex ELinOS system-wide configuration script
39586 This script is useful when debugging a program on an ELinOS target.
39587 It takes advantage of the environment variables defined in a standard
39588 ELinOS environment in order to determine the location of the system
39589 shared libraries, and then sets the @samp{solib-absolute-prefix}
39590 and @samp{solib-search-path} variables appropriately.
39592 @item @file{wrs-linux.py}
39593 @pindex wrs-linux.py
39594 @cindex Wind River Linux system-wide configuration script
39595 This script is useful when debugging a program on a target running
39596 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
39597 the host-side sysroot used by the target system.
39601 @node Maintenance Commands
39602 @appendix Maintenance Commands
39603 @cindex maintenance commands
39604 @cindex internal commands
39606 In addition to commands intended for @value{GDBN} users, @value{GDBN}
39607 includes a number of commands intended for @value{GDBN} developers,
39608 that are not documented elsewhere in this manual. These commands are
39609 provided here for reference. (For commands that turn on debugging
39610 messages, see @ref{Debugging Output}.)
39613 @kindex maint agent
39614 @kindex maint agent-eval
39615 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
39616 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
39617 Translate the given @var{expression} into remote agent bytecodes.
39618 This command is useful for debugging the Agent Expression mechanism
39619 (@pxref{Agent Expressions}). The @samp{agent} version produces an
39620 expression useful for data collection, such as by tracepoints, while
39621 @samp{maint agent-eval} produces an expression that evaluates directly
39622 to a result. For instance, a collection expression for @code{globa +
39623 globb} will include bytecodes to record four bytes of memory at each
39624 of the addresses of @code{globa} and @code{globb}, while discarding
39625 the result of the addition, while an evaluation expression will do the
39626 addition and return the sum.
39627 If @code{-at} is given, generate remote agent bytecode for @var{location}.
39628 If not, generate remote agent bytecode for current frame PC address.
39630 @kindex maint agent-printf
39631 @item maint agent-printf @var{format},@var{expr},...
39632 Translate the given format string and list of argument expressions
39633 into remote agent bytecodes and display them as a disassembled list.
39634 This command is useful for debugging the agent version of dynamic
39635 printf (@pxref{Dynamic Printf}).
39637 @kindex maint info breakpoints
39638 @item @anchor{maint info breakpoints}maint info breakpoints
39639 Using the same format as @samp{info breakpoints}, display both the
39640 breakpoints you've set explicitly, and those @value{GDBN} is using for
39641 internal purposes. Internal breakpoints are shown with negative
39642 breakpoint numbers. The type column identifies what kind of breakpoint
39647 Normal, explicitly set breakpoint.
39650 Normal, explicitly set watchpoint.
39653 Internal breakpoint, used to handle correctly stepping through
39654 @code{longjmp} calls.
39656 @item longjmp resume
39657 Internal breakpoint at the target of a @code{longjmp}.
39660 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
39663 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
39666 Shared library events.
39670 @kindex maint info btrace
39671 @item maint info btrace
39672 Pint information about raw branch tracing data.
39674 @kindex maint btrace packet-history
39675 @item maint btrace packet-history
39676 Print the raw branch trace packets that are used to compute the
39677 execution history for the @samp{record btrace} command. Both the
39678 information and the format in which it is printed depend on the btrace
39683 For the BTS recording format, print a list of blocks of sequential
39684 code. For each block, the following information is printed:
39688 Newer blocks have higher numbers. The oldest block has number zero.
39689 @item Lowest @samp{PC}
39690 @item Highest @samp{PC}
39694 For the Intel Processor Trace recording format, print a list of
39695 Intel Processor Trace packets. For each packet, the following
39696 information is printed:
39699 @item Packet number
39700 Newer packets have higher numbers. The oldest packet has number zero.
39702 The packet's offset in the trace stream.
39703 @item Packet opcode and payload
39707 @kindex maint btrace clear-packet-history
39708 @item maint btrace clear-packet-history
39709 Discards the cached packet history printed by the @samp{maint btrace
39710 packet-history} command. The history will be computed again when
39713 @kindex maint btrace clear
39714 @item maint btrace clear
39715 Discard the branch trace data. The data will be fetched anew and the
39716 branch trace will be recomputed when needed.
39718 This implicitly truncates the branch trace to a single branch trace
39719 buffer. When updating branch trace incrementally, the branch trace
39720 available to @value{GDBN} may be bigger than a single branch trace
39723 @kindex maint set btrace pt skip-pad
39724 @item maint set btrace pt skip-pad
39725 @kindex maint show btrace pt skip-pad
39726 @item maint show btrace pt skip-pad
39727 Control whether @value{GDBN} will skip PAD packets when computing the
39730 @kindex set displaced-stepping
39731 @kindex show displaced-stepping
39732 @cindex displaced stepping support
39733 @cindex out-of-line single-stepping
39734 @item set displaced-stepping
39735 @itemx show displaced-stepping
39736 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
39737 if the target supports it. Displaced stepping is a way to single-step
39738 over breakpoints without removing them from the inferior, by executing
39739 an out-of-line copy of the instruction that was originally at the
39740 breakpoint location. It is also known as out-of-line single-stepping.
39743 @item set displaced-stepping on
39744 If the target architecture supports it, @value{GDBN} will use
39745 displaced stepping to step over breakpoints.
39747 @item set displaced-stepping off
39748 @value{GDBN} will not use displaced stepping to step over breakpoints,
39749 even if such is supported by the target architecture.
39751 @cindex non-stop mode, and @samp{set displaced-stepping}
39752 @item set displaced-stepping auto
39753 This is the default mode. @value{GDBN} will use displaced stepping
39754 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
39755 architecture supports displaced stepping.
39758 @kindex maint check-psymtabs
39759 @item maint check-psymtabs
39760 Check the consistency of currently expanded psymtabs versus symtabs.
39761 Use this to check, for example, whether a symbol is in one but not the other.
39763 @kindex maint check-symtabs
39764 @item maint check-symtabs
39765 Check the consistency of currently expanded symtabs.
39767 @kindex maint expand-symtabs
39768 @item maint expand-symtabs [@var{regexp}]
39769 Expand symbol tables.
39770 If @var{regexp} is specified, only expand symbol tables for file
39771 names matching @var{regexp}.
39773 @kindex maint set catch-demangler-crashes
39774 @kindex maint show catch-demangler-crashes
39775 @cindex demangler crashes
39776 @item maint set catch-demangler-crashes [on|off]
39777 @itemx maint show catch-demangler-crashes
39778 Control whether @value{GDBN} should attempt to catch crashes in the
39779 symbol name demangler. The default is to attempt to catch crashes.
39780 If enabled, the first time a crash is caught, a core file is created,
39781 the offending symbol is displayed and the user is presented with the
39782 option to terminate the current session.
39784 @kindex maint cplus first_component
39785 @item maint cplus first_component @var{name}
39786 Print the first C@t{++} class/namespace component of @var{name}.
39788 @kindex maint cplus namespace
39789 @item maint cplus namespace
39790 Print the list of possible C@t{++} namespaces.
39792 @kindex maint deprecate
39793 @kindex maint undeprecate
39794 @cindex deprecated commands
39795 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
39796 @itemx maint undeprecate @var{command}
39797 Deprecate or undeprecate the named @var{command}. Deprecated commands
39798 cause @value{GDBN} to issue a warning when you use them. The optional
39799 argument @var{replacement} says which newer command should be used in
39800 favor of the deprecated one; if it is given, @value{GDBN} will mention
39801 the replacement as part of the warning.
39803 @kindex maint dump-me
39804 @item maint dump-me
39805 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
39806 Cause a fatal signal in the debugger and force it to dump its core.
39807 This is supported only on systems which support aborting a program
39808 with the @code{SIGQUIT} signal.
39810 @kindex maint internal-error
39811 @kindex maint internal-warning
39812 @kindex maint demangler-warning
39813 @cindex demangler crashes
39814 @item maint internal-error @r{[}@var{message-text}@r{]}
39815 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
39816 @itemx maint demangler-warning @r{[}@var{message-text}@r{]}
39818 Cause @value{GDBN} to call the internal function @code{internal_error},
39819 @code{internal_warning} or @code{demangler_warning} and hence behave
39820 as though an internal problem has been detected. In addition to
39821 reporting the internal problem, these functions give the user the
39822 opportunity to either quit @value{GDBN} or (for @code{internal_error}
39823 and @code{internal_warning}) create a core file of the current
39824 @value{GDBN} session.
39826 These commands take an optional parameter @var{message-text} that is
39827 used as the text of the error or warning message.
39829 Here's an example of using @code{internal-error}:
39832 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
39833 @dots{}/maint.c:121: internal-error: testing, 1, 2
39834 A problem internal to GDB has been detected. Further
39835 debugging may prove unreliable.
39836 Quit this debugging session? (y or n) @kbd{n}
39837 Create a core file? (y or n) @kbd{n}
39841 @cindex @value{GDBN} internal error
39842 @cindex internal errors, control of @value{GDBN} behavior
39843 @cindex demangler crashes
39845 @kindex maint set internal-error
39846 @kindex maint show internal-error
39847 @kindex maint set internal-warning
39848 @kindex maint show internal-warning
39849 @kindex maint set demangler-warning
39850 @kindex maint show demangler-warning
39851 @item maint set internal-error @var{action} [ask|yes|no]
39852 @itemx maint show internal-error @var{action}
39853 @itemx maint set internal-warning @var{action} [ask|yes|no]
39854 @itemx maint show internal-warning @var{action}
39855 @itemx maint set demangler-warning @var{action} [ask|yes|no]
39856 @itemx maint show demangler-warning @var{action}
39857 When @value{GDBN} reports an internal problem (error or warning) it
39858 gives the user the opportunity to both quit @value{GDBN} and create a
39859 core file of the current @value{GDBN} session. These commands let you
39860 override the default behaviour for each particular @var{action},
39861 described in the table below.
39865 You can specify that @value{GDBN} should always (yes) or never (no)
39866 quit. The default is to ask the user what to do.
39869 You can specify that @value{GDBN} should always (yes) or never (no)
39870 create a core file. The default is to ask the user what to do. Note
39871 that there is no @code{corefile} option for @code{demangler-warning}:
39872 demangler warnings always create a core file and this cannot be
39876 @kindex maint packet
39877 @item maint packet @var{text}
39878 If @value{GDBN} is talking to an inferior via the serial protocol,
39879 then this command sends the string @var{text} to the inferior, and
39880 displays the response packet. @value{GDBN} supplies the initial
39881 @samp{$} character, the terminating @samp{#} character, and the
39884 @kindex maint print architecture
39885 @item maint print architecture @r{[}@var{file}@r{]}
39886 Print the entire architecture configuration. The optional argument
39887 @var{file} names the file where the output goes.
39889 @kindex maint print c-tdesc @r{[}@var{file}@r{]}
39890 @item maint print c-tdesc
39891 Print the target description (@pxref{Target Descriptions}) as
39892 a C source file. By default, the target description is for the current
39893 target, but if the optional argument @var{file} is provided, that file
39894 is used to produce the description. The @var{file} should be an XML
39895 document, of the form described in @ref{Target Description Format}.
39896 The created source file is built into @value{GDBN} when @value{GDBN} is
39897 built again. This command is used by developers after they add or
39898 modify XML target descriptions.
39900 @kindex maint check xml-descriptions
39901 @item maint check xml-descriptions @var{dir}
39902 Check that the target descriptions dynamically created by @value{GDBN}
39903 equal the descriptions created from XML files found in @var{dir}.
39905 @anchor{maint check libthread-db}
39906 @kindex maint check libthread-db
39907 @item maint check libthread-db
39908 Run integrity checks on the current inferior's thread debugging
39909 library. This exercises all @code{libthread_db} functionality used by
39910 @value{GDBN} on GNU/Linux systems, and by extension also exercises the
39911 @code{proc_service} functions provided by @value{GDBN} that
39912 @code{libthread_db} uses. Note that parts of the test may be skipped
39913 on some platforms when debugging core files.
39915 @kindex maint print dummy-frames
39916 @item maint print dummy-frames
39917 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
39920 (@value{GDBP}) @kbd{b add}
39922 (@value{GDBP}) @kbd{print add(2,3)}
39923 Breakpoint 2, add (a=2, b=3) at @dots{}
39925 The program being debugged stopped while in a function called from GDB.
39927 (@value{GDBP}) @kbd{maint print dummy-frames}
39928 0xa8206d8: id=@{stack=0xbfffe734,code=0xbfffe73f,!special@}, ptid=process 9353
39932 Takes an optional file parameter.
39934 @kindex maint print registers
39935 @kindex maint print raw-registers
39936 @kindex maint print cooked-registers
39937 @kindex maint print register-groups
39938 @kindex maint print remote-registers
39939 @item maint print registers @r{[}@var{file}@r{]}
39940 @itemx maint print raw-registers @r{[}@var{file}@r{]}
39941 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
39942 @itemx maint print register-groups @r{[}@var{file}@r{]}
39943 @itemx maint print remote-registers @r{[}@var{file}@r{]}
39944 Print @value{GDBN}'s internal register data structures.
39946 The command @code{maint print raw-registers} includes the contents of
39947 the raw register cache; the command @code{maint print
39948 cooked-registers} includes the (cooked) value of all registers,
39949 including registers which aren't available on the target nor visible
39950 to user; the command @code{maint print register-groups} includes the
39951 groups that each register is a member of; and the command @code{maint
39952 print remote-registers} includes the remote target's register numbers
39953 and offsets in the `G' packets.
39955 These commands take an optional parameter, a file name to which to
39956 write the information.
39958 @kindex maint print reggroups
39959 @item maint print reggroups @r{[}@var{file}@r{]}
39960 Print @value{GDBN}'s internal register group data structures. The
39961 optional argument @var{file} tells to what file to write the
39964 The register groups info looks like this:
39967 (@value{GDBP}) @kbd{maint print reggroups}
39980 This command forces @value{GDBN} to flush its internal register cache.
39982 @kindex maint print address-spaces
39983 @item maint print address-spaces @r{[}@var{file}@r{]}
39984 Print @value{GDBN}'s internal address space data structures. The
39985 optional argument @var{file} tells to what file to write the
39986 information. @xref{maint print address-spaces,, @code{maint print
39989 @kindex maint print objfiles
39990 @cindex info for known object files
39991 @item maint print objfiles @r{[}@var{regexp}@r{]}
39992 Print a dump of all known object files.
39993 If @var{regexp} is specified, only print object files whose names
39994 match @var{regexp}. For each object file, this command prints its name,
39995 address in memory, and all of its psymtabs and symtabs.
39997 @kindex maint print user-registers
39998 @cindex user registers
39999 @item maint print user-registers
40000 List all currently available @dfn{user registers}. User registers
40001 typically provide alternate names for actual hardware registers. They
40002 include the four ``standard'' registers @code{$fp}, @code{$pc},
40003 @code{$sp}, and @code{$ps}. @xref{standard registers}. User
40004 registers can be used in expressions in the same way as the canonical
40005 register names, but only the latter are listed by the @code{info
40006 registers} and @code{maint print registers} commands.
40008 @kindex maint print section-scripts
40009 @cindex info for known .debug_gdb_scripts-loaded scripts
40010 @item maint print section-scripts [@var{regexp}]
40011 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
40012 If @var{regexp} is specified, only print scripts loaded by object files
40013 matching @var{regexp}.
40014 For each script, this command prints its name as specified in the objfile,
40015 and the full path if known.
40016 @xref{dotdebug_gdb_scripts section}.
40018 @kindex maint print statistics
40019 @cindex bcache statistics
40020 @item maint print statistics
40021 This command prints, for each object file in the program, various data
40022 about that object file followed by the byte cache (@dfn{bcache})
40023 statistics for the object file. The objfile data includes the number
40024 of minimal, partial, full, and stabs symbols, the number of types
40025 defined by the objfile, the number of as yet unexpanded psym tables,
40026 the number of line tables and string tables, and the amount of memory
40027 used by the various tables. The bcache statistics include the counts,
40028 sizes, and counts of duplicates of all and unique objects, max,
40029 average, and median entry size, total memory used and its overhead and
40030 savings, and various measures of the hash table size and chain
40033 @kindex maint print target-stack
40034 @cindex target stack description
40035 @item maint print target-stack
40036 A @dfn{target} is an interface between the debugger and a particular
40037 kind of file or process. Targets can be stacked in @dfn{strata},
40038 so that more than one target can potentially respond to a request.
40039 In particular, memory accesses will walk down the stack of targets
40040 until they find a target that is interested in handling that particular
40043 This command prints a short description of each layer that was pushed on
40044 the @dfn{target stack}, starting from the top layer down to the bottom one.
40046 @kindex maint print type
40047 @cindex type chain of a data type
40048 @item maint print type @var{expr}
40049 Print the type chain for a type specified by @var{expr}. The argument
40050 can be either a type name or a symbol. If it is a symbol, the type of
40051 that symbol is described. The type chain produced by this command is
40052 a recursive definition of the data type as stored in @value{GDBN}'s
40053 data structures, including its flags and contained types.
40055 @kindex maint selftest
40057 @item maint selftest @r{[}@var{filter}@r{]}
40058 Run any self tests that were compiled in to @value{GDBN}. This will
40059 print a message showing how many tests were run, and how many failed.
40060 If a @var{filter} is passed, only the tests with @var{filter} in their
40063 @kindex maint info selftests
40065 @item maint info selftests
40066 List the selftests compiled in to @value{GDBN}.
40068 @kindex maint set dwarf always-disassemble
40069 @kindex maint show dwarf always-disassemble
40070 @item maint set dwarf always-disassemble
40071 @item maint show dwarf always-disassemble
40072 Control the behavior of @code{info address} when using DWARF debugging
40075 The default is @code{off}, which means that @value{GDBN} should try to
40076 describe a variable's location in an easily readable format. When
40077 @code{on}, @value{GDBN} will instead display the DWARF location
40078 expression in an assembly-like format. Note that some locations are
40079 too complex for @value{GDBN} to describe simply; in this case you will
40080 always see the disassembly form.
40082 Here is an example of the resulting disassembly:
40085 (@value{GDBP}) info addr argc
40086 Symbol "argc" is a complex DWARF expression:
40090 For more information on these expressions, see
40091 @uref{http://www.dwarfstd.org/, the DWARF standard}.
40093 @kindex maint set dwarf max-cache-age
40094 @kindex maint show dwarf max-cache-age
40095 @item maint set dwarf max-cache-age
40096 @itemx maint show dwarf max-cache-age
40097 Control the DWARF compilation unit cache.
40099 @cindex DWARF compilation units cache
40100 In object files with inter-compilation-unit references, such as those
40101 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF
40102 reader needs to frequently refer to previously read compilation units.
40103 This setting controls how long a compilation unit will remain in the
40104 cache if it is not referenced. A higher limit means that cached
40105 compilation units will be stored in memory longer, and more total
40106 memory will be used. Setting it to zero disables caching, which will
40107 slow down @value{GDBN} startup, but reduce memory consumption.
40109 @kindex maint set dwarf unwinders
40110 @kindex maint show dwarf unwinders
40111 @item maint set dwarf unwinders
40112 @itemx maint show dwarf unwinders
40113 Control use of the DWARF frame unwinders.
40115 @cindex DWARF frame unwinders
40116 Many targets that support DWARF debugging use @value{GDBN}'s DWARF
40117 frame unwinders to build the backtrace. Many of these targets will
40118 also have a second mechanism for building the backtrace for use in
40119 cases where DWARF information is not available, this second mechanism
40120 is often an analysis of a function's prologue.
40122 In order to extend testing coverage of the second level stack
40123 unwinding mechanisms it is helpful to be able to disable the DWARF
40124 stack unwinders, this can be done with this switch.
40126 In normal use of @value{GDBN} disabling the DWARF unwinders is not
40127 advisable, there are cases that are better handled through DWARF than
40128 prologue analysis, and the debug experience is likely to be better
40129 with the DWARF frame unwinders enabled.
40131 If DWARF frame unwinders are not supported for a particular target
40132 architecture, then enabling this flag does not cause them to be used.
40134 @kindex maint set worker-threads
40135 @kindex maint show worker-threads
40136 @item maint set worker-threads
40137 @item maint show worker-threads
40138 Control the number of worker threads that may be used by @value{GDBN}.
40139 On capable hosts, @value{GDBN} may use multiple threads to speed up
40140 certain CPU-intensive operations, such as demangling symbol names.
40141 While the number of threads used by @value{GDBN} may vary, this
40142 command can be used to set an upper bound on this number. The default
40143 is @code{0} (disabled). A value of @code{unlimited} lets @value{GDBN} choose a
40144 reasonable number. Note that this only controls worker threads started by
40145 @value{GDBN} itself; libraries used by @value{GDBN} may start threads of their
40148 @kindex maint set profile
40149 @kindex maint show profile
40150 @cindex profiling GDB
40151 @item maint set profile
40152 @itemx maint show profile
40153 Control profiling of @value{GDBN}.
40155 Profiling will be disabled until you use the @samp{maint set profile}
40156 command to enable it. When you enable profiling, the system will begin
40157 collecting timing and execution count data; when you disable profiling or
40158 exit @value{GDBN}, the results will be written to a log file. Remember that
40159 if you use profiling, @value{GDBN} will overwrite the profiling log file
40160 (often called @file{gmon.out}). If you have a record of important profiling
40161 data in a @file{gmon.out} file, be sure to move it to a safe location.
40163 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
40164 compiled with the @samp{-pg} compiler option.
40166 @kindex maint set show-debug-regs
40167 @kindex maint show show-debug-regs
40168 @cindex hardware debug registers
40169 @item maint set show-debug-regs
40170 @itemx maint show show-debug-regs
40171 Control whether to show variables that mirror the hardware debug
40172 registers. Use @code{on} to enable, @code{off} to disable. If
40173 enabled, the debug registers values are shown when @value{GDBN} inserts or
40174 removes a hardware breakpoint or watchpoint, and when the inferior
40175 triggers a hardware-assisted breakpoint or watchpoint.
40177 @kindex maint set show-all-tib
40178 @kindex maint show show-all-tib
40179 @item maint set show-all-tib
40180 @itemx maint show show-all-tib
40181 Control whether to show all non zero areas within a 1k block starting
40182 at thread local base, when using the @samp{info w32 thread-information-block}
40185 @kindex maint set target-async
40186 @kindex maint show target-async
40187 @item maint set target-async
40188 @itemx maint show target-async
40189 This controls whether @value{GDBN} targets operate in synchronous or
40190 asynchronous mode (@pxref{Background Execution}). Normally the
40191 default is asynchronous, if it is available; but this can be changed
40192 to more easily debug problems occurring only in synchronous mode.
40194 @kindex maint set target-non-stop @var{mode} [on|off|auto]
40195 @kindex maint show target-non-stop
40196 @item maint set target-non-stop
40197 @itemx maint show target-non-stop
40199 This controls whether @value{GDBN} targets always operate in non-stop
40200 mode even if @code{set non-stop} is @code{off} (@pxref{Non-Stop
40201 Mode}). The default is @code{auto}, meaning non-stop mode is enabled
40202 if supported by the target.
40205 @item maint set target-non-stop auto
40206 This is the default mode. @value{GDBN} controls the target in
40207 non-stop mode if the target supports it.
40209 @item maint set target-non-stop on
40210 @value{GDBN} controls the target in non-stop mode even if the target
40211 does not indicate support.
40213 @item maint set target-non-stop off
40214 @value{GDBN} does not control the target in non-stop mode even if the
40215 target supports it.
40218 @kindex maint set tui-resize-message
40219 @kindex maint show tui-resize-message
40220 @item maint set tui-resize-message
40221 @item maint show tui-resize-message
40222 Control whether @value{GDBN} displays a message each time the terminal
40223 is resized when in TUI mode. The default is @code{off}, which means
40224 that @value{GDBN} is silent during resizes. When @code{on},
40225 @value{GDBN} will display a message after a resize is completed; the
40226 message will include a number indicating how many times the terminal
40227 has been resized. This setting is intended for use by the test suite,
40228 where it would otherwise be difficult to determine when a resize and
40229 refresh has been completed.
40231 @kindex maint set per-command
40232 @kindex maint show per-command
40233 @item maint set per-command
40234 @itemx maint show per-command
40235 @cindex resources used by commands
40237 @value{GDBN} can display the resources used by each command.
40238 This is useful in debugging performance problems.
40241 @item maint set per-command space [on|off]
40242 @itemx maint show per-command space
40243 Enable or disable the printing of the memory used by GDB for each command.
40244 If enabled, @value{GDBN} will display how much memory each command
40245 took, following the command's own output.
40246 This can also be requested by invoking @value{GDBN} with the
40247 @option{--statistics} command-line switch (@pxref{Mode Options}).
40249 @item maint set per-command time [on|off]
40250 @itemx maint show per-command time
40251 Enable or disable the printing of the execution time of @value{GDBN}
40253 If enabled, @value{GDBN} will display how much time it
40254 took to execute each command, following the command's own output.
40255 Both CPU time and wallclock time are printed.
40256 Printing both is useful when trying to determine whether the cost is
40257 CPU or, e.g., disk/network latency.
40258 Note that the CPU time printed is for @value{GDBN} only, it does not include
40259 the execution time of the inferior because there's no mechanism currently
40260 to compute how much time was spent by @value{GDBN} and how much time was
40261 spent by the program been debugged.
40262 This can also be requested by invoking @value{GDBN} with the
40263 @option{--statistics} command-line switch (@pxref{Mode Options}).
40265 @item maint set per-command symtab [on|off]
40266 @itemx maint show per-command symtab
40267 Enable or disable the printing of basic symbol table statistics
40269 If enabled, @value{GDBN} will display the following information:
40273 number of symbol tables
40275 number of primary symbol tables
40277 number of blocks in the blockvector
40281 @kindex maint set check-libthread-db
40282 @kindex maint show check-libthread-db
40283 @item maint set check-libthread-db [on|off]
40284 @itemx maint show check-libthread-db
40285 Control whether @value{GDBN} should run integrity checks on inferior
40286 specific thread debugging libraries as they are loaded. The default
40287 is not to perform such checks. If any check fails @value{GDBN} will
40288 unload the library and continue searching for a suitable candidate as
40289 described in @ref{set libthread-db-search-path}. For more information
40290 about the tests, see @ref{maint check libthread-db}.
40292 @kindex maint space
40293 @cindex memory used by commands
40294 @item maint space @var{value}
40295 An alias for @code{maint set per-command space}.
40296 A non-zero value enables it, zero disables it.
40299 @cindex time of command execution
40300 @item maint time @var{value}
40301 An alias for @code{maint set per-command time}.
40302 A non-zero value enables it, zero disables it.
40304 @kindex maint translate-address
40305 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
40306 Find the symbol stored at the location specified by the address
40307 @var{addr} and an optional section name @var{section}. If found,
40308 @value{GDBN} prints the name of the closest symbol and an offset from
40309 the symbol's location to the specified address. This is similar to
40310 the @code{info address} command (@pxref{Symbols}), except that this
40311 command also allows to find symbols in other sections.
40313 If section was not specified, the section in which the symbol was found
40314 is also printed. For dynamically linked executables, the name of
40315 executable or shared library containing the symbol is printed as well.
40317 @kindex maint test-options
40318 @item maint test-options require-delimiter
40319 @itemx maint test-options unknown-is-error
40320 @itemx maint test-options unknown-is-operand
40321 These commands are used by the testsuite to validate the command
40322 options framework. The @code{require-delimiter} variant requires a
40323 double-dash delimiter to indicate end of options. The
40324 @code{unknown-is-error} and @code{unknown-is-operand} do not. The
40325 @code{unknown-is-error} variant throws an error on unknown option,
40326 while @code{unknown-is-operand} treats unknown options as the start of
40327 the command's operands. When run, the commands output the result of
40328 the processed options. When completed, the commands store the
40329 internal result of completion in a variable exposed by the @code{maint
40330 show test-options-completion-result} command.
40332 @kindex maint show test-options-completion-result
40333 @item maint show test-options-completion-result
40334 Shows the result of completing the @code{maint test-options}
40335 subcommands. This is used by the testsuite to validate completion
40336 support in the command options framework.
40338 @kindex maint set test-settings
40339 @kindex maint show test-settings
40340 @item maint set test-settings @var{kind}
40341 @itemx maint show test-settings @var{kind}
40342 These are representative commands for each @var{kind} of setting type
40343 @value{GDBN} supports. They are used by the testsuite for exercising
40344 the settings infrastructure.
40347 @item maint with @var{setting} [@var{value}] [-- @var{command}]
40348 Like the @code{with} command, but works with @code{maintenance set}
40349 variables. This is used by the testsuite to exercise the @code{with}
40350 command's infrastructure.
40354 The following command is useful for non-interactive invocations of
40355 @value{GDBN}, such as in the test suite.
40358 @item set watchdog @var{nsec}
40359 @kindex set watchdog
40360 @cindex watchdog timer
40361 @cindex timeout for commands
40362 Set the maximum number of seconds @value{GDBN} will wait for the
40363 target operation to finish. If this time expires, @value{GDBN}
40364 reports and error and the command is aborted.
40366 @item show watchdog
40367 Show the current setting of the target wait timeout.
40370 @node Remote Protocol
40371 @appendix @value{GDBN} Remote Serial Protocol
40376 * Stop Reply Packets::
40377 * General Query Packets::
40378 * Architecture-Specific Protocol Details::
40379 * Tracepoint Packets::
40380 * Host I/O Packets::
40382 * Notification Packets::
40383 * Remote Non-Stop::
40384 * Packet Acknowledgment::
40386 * File-I/O Remote Protocol Extension::
40387 * Library List Format::
40388 * Library List Format for SVR4 Targets::
40389 * Memory Map Format::
40390 * Thread List Format::
40391 * Traceframe Info Format::
40392 * Branch Trace Format::
40393 * Branch Trace Configuration Format::
40399 There may be occasions when you need to know something about the
40400 protocol---for example, if there is only one serial port to your target
40401 machine, you might want your program to do something special if it
40402 recognizes a packet meant for @value{GDBN}.
40404 In the examples below, @samp{->} and @samp{<-} are used to indicate
40405 transmitted and received data, respectively.
40407 @cindex protocol, @value{GDBN} remote serial
40408 @cindex serial protocol, @value{GDBN} remote
40409 @cindex remote serial protocol
40410 All @value{GDBN} commands and responses (other than acknowledgments
40411 and notifications, see @ref{Notification Packets}) are sent as a
40412 @var{packet}. A @var{packet} is introduced with the character
40413 @samp{$}, the actual @var{packet-data}, and the terminating character
40414 @samp{#} followed by a two-digit @var{checksum}:
40417 @code{$}@var{packet-data}@code{#}@var{checksum}
40421 @cindex checksum, for @value{GDBN} remote
40423 The two-digit @var{checksum} is computed as the modulo 256 sum of all
40424 characters between the leading @samp{$} and the trailing @samp{#} (an
40425 eight bit unsigned checksum).
40427 Implementors should note that prior to @value{GDBN} 5.0 the protocol
40428 specification also included an optional two-digit @var{sequence-id}:
40431 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
40434 @cindex sequence-id, for @value{GDBN} remote
40436 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
40437 has never output @var{sequence-id}s. Stubs that handle packets added
40438 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
40440 When either the host or the target machine receives a packet, the first
40441 response expected is an acknowledgment: either @samp{+} (to indicate
40442 the package was received correctly) or @samp{-} (to request
40446 -> @code{$}@var{packet-data}@code{#}@var{checksum}
40451 The @samp{+}/@samp{-} acknowledgments can be disabled
40452 once a connection is established.
40453 @xref{Packet Acknowledgment}, for details.
40455 The host (@value{GDBN}) sends @var{command}s, and the target (the
40456 debugging stub incorporated in your program) sends a @var{response}. In
40457 the case of step and continue @var{command}s, the response is only sent
40458 when the operation has completed, and the target has again stopped all
40459 threads in all attached processes. This is the default all-stop mode
40460 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
40461 execution mode; see @ref{Remote Non-Stop}, for details.
40463 @var{packet-data} consists of a sequence of characters with the
40464 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
40467 @cindex remote protocol, field separator
40468 Fields within the packet should be separated using @samp{,} @samp{;} or
40469 @samp{:}. Except where otherwise noted all numbers are represented in
40470 @sc{hex} with leading zeros suppressed.
40472 Implementors should note that prior to @value{GDBN} 5.0, the character
40473 @samp{:} could not appear as the third character in a packet (as it
40474 would potentially conflict with the @var{sequence-id}).
40476 @cindex remote protocol, binary data
40477 @anchor{Binary Data}
40478 Binary data in most packets is encoded either as two hexadecimal
40479 digits per byte of binary data. This allowed the traditional remote
40480 protocol to work over connections which were only seven-bit clean.
40481 Some packets designed more recently assume an eight-bit clean
40482 connection, and use a more efficient encoding to send and receive
40485 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
40486 as an escape character. Any escaped byte is transmitted as the escape
40487 character followed by the original character XORed with @code{0x20}.
40488 For example, the byte @code{0x7d} would be transmitted as the two
40489 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
40490 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
40491 @samp{@}}) must always be escaped. Responses sent by the stub
40492 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
40493 is not interpreted as the start of a run-length encoded sequence
40496 Response @var{data} can be run-length encoded to save space.
40497 Run-length encoding replaces runs of identical characters with one
40498 instance of the repeated character, followed by a @samp{*} and a
40499 repeat count. The repeat count is itself sent encoded, to avoid
40500 binary characters in @var{data}: a value of @var{n} is sent as
40501 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
40502 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
40503 code 32) for a repeat count of 3. (This is because run-length
40504 encoding starts to win for counts 3 or more.) Thus, for example,
40505 @samp{0* } is a run-length encoding of ``0000'': the space character
40506 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
40509 The printable characters @samp{#} and @samp{$} or with a numeric value
40510 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
40511 seven repeats (@samp{$}) can be expanded using a repeat count of only
40512 five (@samp{"}). For example, @samp{00000000} can be encoded as
40515 The error response returned for some packets includes a two character
40516 error number. That number is not well defined.
40518 @cindex empty response, for unsupported packets
40519 For any @var{command} not supported by the stub, an empty response
40520 (@samp{$#00}) should be returned. That way it is possible to extend the
40521 protocol. A newer @value{GDBN} can tell if a packet is supported based
40524 At a minimum, a stub is required to support the @samp{g} and @samp{G}
40525 commands for register access, and the @samp{m} and @samp{M} commands
40526 for memory access. Stubs that only control single-threaded targets
40527 can implement run control with the @samp{c} (continue), and @samp{s}
40528 (step) commands. Stubs that support multi-threading targets should
40529 support the @samp{vCont} command. All other commands are optional.
40534 The following table provides a complete list of all currently defined
40535 @var{command}s and their corresponding response @var{data}.
40536 @xref{File-I/O Remote Protocol Extension}, for details about the File
40537 I/O extension of the remote protocol.
40539 Each packet's description has a template showing the packet's overall
40540 syntax, followed by an explanation of the packet's meaning. We
40541 include spaces in some of the templates for clarity; these are not
40542 part of the packet's syntax. No @value{GDBN} packet uses spaces to
40543 separate its components. For example, a template like @samp{foo
40544 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
40545 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
40546 @var{baz}. @value{GDBN} does not transmit a space character between the
40547 @samp{foo} and the @var{bar}, or between the @var{bar} and the
40550 @cindex @var{thread-id}, in remote protocol
40551 @anchor{thread-id syntax}
40552 Several packets and replies include a @var{thread-id} field to identify
40553 a thread. Normally these are positive numbers with a target-specific
40554 interpretation, formatted as big-endian hex strings. A @var{thread-id}
40555 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
40558 In addition, the remote protocol supports a multiprocess feature in
40559 which the @var{thread-id} syntax is extended to optionally include both
40560 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
40561 The @var{pid} (process) and @var{tid} (thread) components each have the
40562 format described above: a positive number with target-specific
40563 interpretation formatted as a big-endian hex string, literal @samp{-1}
40564 to indicate all processes or threads (respectively), or @samp{0} to
40565 indicate an arbitrary process or thread. Specifying just a process, as
40566 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
40567 error to specify all processes but a specific thread, such as
40568 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
40569 for those packets and replies explicitly documented to include a process
40570 ID, rather than a @var{thread-id}.
40572 The multiprocess @var{thread-id} syntax extensions are only used if both
40573 @value{GDBN} and the stub report support for the @samp{multiprocess}
40574 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
40577 Note that all packet forms beginning with an upper- or lower-case
40578 letter, other than those described here, are reserved for future use.
40580 Here are the packet descriptions.
40585 @cindex @samp{!} packet
40586 @anchor{extended mode}
40587 Enable extended mode. In extended mode, the remote server is made
40588 persistent. The @samp{R} packet is used to restart the program being
40594 The remote target both supports and has enabled extended mode.
40598 @cindex @samp{?} packet
40600 Indicate the reason the target halted. The reply is the same as for
40601 step and continue. This packet has a special interpretation when the
40602 target is in non-stop mode; see @ref{Remote Non-Stop}.
40605 @xref{Stop Reply Packets}, for the reply specifications.
40607 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
40608 @cindex @samp{A} packet
40609 Initialized @code{argv[]} array passed into program. @var{arglen}
40610 specifies the number of bytes in the hex encoded byte stream
40611 @var{arg}. See @code{gdbserver} for more details.
40616 The arguments were set.
40622 @cindex @samp{b} packet
40623 (Don't use this packet; its behavior is not well-defined.)
40624 Change the serial line speed to @var{baud}.
40626 JTC: @emph{When does the transport layer state change? When it's
40627 received, or after the ACK is transmitted. In either case, there are
40628 problems if the command or the acknowledgment packet is dropped.}
40630 Stan: @emph{If people really wanted to add something like this, and get
40631 it working for the first time, they ought to modify ser-unix.c to send
40632 some kind of out-of-band message to a specially-setup stub and have the
40633 switch happen "in between" packets, so that from remote protocol's point
40634 of view, nothing actually happened.}
40636 @item B @var{addr},@var{mode}
40637 @cindex @samp{B} packet
40638 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
40639 breakpoint at @var{addr}.
40641 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
40642 (@pxref{insert breakpoint or watchpoint packet}).
40644 @cindex @samp{bc} packet
40647 Backward continue. Execute the target system in reverse. No parameter.
40648 @xref{Reverse Execution}, for more information.
40651 @xref{Stop Reply Packets}, for the reply specifications.
40653 @cindex @samp{bs} packet
40656 Backward single step. Execute one instruction in reverse. No parameter.
40657 @xref{Reverse Execution}, for more information.
40660 @xref{Stop Reply Packets}, for the reply specifications.
40662 @item c @r{[}@var{addr}@r{]}
40663 @cindex @samp{c} packet
40664 Continue at @var{addr}, which is the address to resume. If @var{addr}
40665 is omitted, resume at current address.
40667 This packet is deprecated for multi-threading support. @xref{vCont
40671 @xref{Stop Reply Packets}, for the reply specifications.
40673 @item C @var{sig}@r{[};@var{addr}@r{]}
40674 @cindex @samp{C} packet
40675 Continue with signal @var{sig} (hex signal number). If
40676 @samp{;@var{addr}} is omitted, resume at same address.
40678 This packet is deprecated for multi-threading support. @xref{vCont
40682 @xref{Stop Reply Packets}, for the reply specifications.
40685 @cindex @samp{d} packet
40688 Don't use this packet; instead, define a general set packet
40689 (@pxref{General Query Packets}).
40693 @cindex @samp{D} packet
40694 The first form of the packet is used to detach @value{GDBN} from the
40695 remote system. It is sent to the remote target
40696 before @value{GDBN} disconnects via the @code{detach} command.
40698 The second form, including a process ID, is used when multiprocess
40699 protocol extensions are enabled (@pxref{multiprocess extensions}), to
40700 detach only a specific process. The @var{pid} is specified as a
40701 big-endian hex string.
40711 @item F @var{RC},@var{EE},@var{CF};@var{XX}
40712 @cindex @samp{F} packet
40713 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
40714 This is part of the File-I/O protocol extension. @xref{File-I/O
40715 Remote Protocol Extension}, for the specification.
40718 @anchor{read registers packet}
40719 @cindex @samp{g} packet
40720 Read general registers.
40724 @item @var{XX@dots{}}
40725 Each byte of register data is described by two hex digits. The bytes
40726 with the register are transmitted in target byte order. The size of
40727 each register and their position within the @samp{g} packet are
40728 determined by the @value{GDBN} internal gdbarch functions
40729 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}.
40731 When reading registers from a trace frame (@pxref{Analyze Collected
40732 Data,,Using the Collected Data}), the stub may also return a string of
40733 literal @samp{x}'s in place of the register data digits, to indicate
40734 that the corresponding register has not been collected, thus its value
40735 is unavailable. For example, for an architecture with 4 registers of
40736 4 bytes each, the following reply indicates to @value{GDBN} that
40737 registers 0 and 2 have not been collected, while registers 1 and 3
40738 have been collected, and both have zero value:
40742 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
40749 @item G @var{XX@dots{}}
40750 @cindex @samp{G} packet
40751 Write general registers. @xref{read registers packet}, for a
40752 description of the @var{XX@dots{}} data.
40762 @item H @var{op} @var{thread-id}
40763 @cindex @samp{H} packet
40764 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
40765 @samp{G}, et.al.). Depending on the operation to be performed, @var{op}
40766 should be @samp{c} for step and continue operations (note that this
40767 is deprecated, supporting the @samp{vCont} command is a better
40768 option), and @samp{g} for other operations. The thread designator
40769 @var{thread-id} has the format and interpretation described in
40770 @ref{thread-id syntax}.
40781 @c 'H': How restrictive (or permissive) is the thread model. If a
40782 @c thread is selected and stopped, are other threads allowed
40783 @c to continue to execute? As I mentioned above, I think the
40784 @c semantics of each command when a thread is selected must be
40785 @c described. For example:
40787 @c 'g': If the stub supports threads and a specific thread is
40788 @c selected, returns the register block from that thread;
40789 @c otherwise returns current registers.
40791 @c 'G' If the stub supports threads and a specific thread is
40792 @c selected, sets the registers of the register block of
40793 @c that thread; otherwise sets current registers.
40795 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
40796 @anchor{cycle step packet}
40797 @cindex @samp{i} packet
40798 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
40799 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
40800 step starting at that address.
40803 @cindex @samp{I} packet
40804 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
40808 @cindex @samp{k} packet
40811 The exact effect of this packet is not specified.
40813 For a bare-metal target, it may power cycle or reset the target
40814 system. For that reason, the @samp{k} packet has no reply.
40816 For a single-process target, it may kill that process if possible.
40818 A multiple-process target may choose to kill just one process, or all
40819 that are under @value{GDBN}'s control. For more precise control, use
40820 the vKill packet (@pxref{vKill packet}).
40822 If the target system immediately closes the connection in response to
40823 @samp{k}, @value{GDBN} does not consider the lack of packet
40824 acknowledgment to be an error, and assumes the kill was successful.
40826 If connected using @kbd{target extended-remote}, and the target does
40827 not close the connection in response to a kill request, @value{GDBN}
40828 probes the target state as if a new connection was opened
40829 (@pxref{? packet}).
40831 @item m @var{addr},@var{length}
40832 @cindex @samp{m} packet
40833 Read @var{length} addressable memory units starting at address @var{addr}
40834 (@pxref{addressable memory unit}). Note that @var{addr} may not be aligned to
40835 any particular boundary.
40837 The stub need not use any particular size or alignment when gathering
40838 data from memory for the response; even if @var{addr} is word-aligned
40839 and @var{length} is a multiple of the word size, the stub is free to
40840 use byte accesses, or not. For this reason, this packet may not be
40841 suitable for accessing memory-mapped I/O devices.
40842 @cindex alignment of remote memory accesses
40843 @cindex size of remote memory accesses
40844 @cindex memory, alignment and size of remote accesses
40848 @item @var{XX@dots{}}
40849 Memory contents; each byte is transmitted as a two-digit hexadecimal number.
40850 The reply may contain fewer addressable memory units than requested if the
40851 server was able to read only part of the region of memory.
40856 @item M @var{addr},@var{length}:@var{XX@dots{}}
40857 @cindex @samp{M} packet
40858 Write @var{length} addressable memory units starting at address @var{addr}
40859 (@pxref{addressable memory unit}). The data is given by @var{XX@dots{}}; each
40860 byte is transmitted as a two-digit hexadecimal number.
40867 for an error (this includes the case where only part of the data was
40872 @cindex @samp{p} packet
40873 Read the value of register @var{n}; @var{n} is in hex.
40874 @xref{read registers packet}, for a description of how the returned
40875 register value is encoded.
40879 @item @var{XX@dots{}}
40880 the register's value
40884 Indicating an unrecognized @var{query}.
40887 @item P @var{n@dots{}}=@var{r@dots{}}
40888 @anchor{write register packet}
40889 @cindex @samp{P} packet
40890 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
40891 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
40892 digits for each byte in the register (target byte order).
40902 @item q @var{name} @var{params}@dots{}
40903 @itemx Q @var{name} @var{params}@dots{}
40904 @cindex @samp{q} packet
40905 @cindex @samp{Q} packet
40906 General query (@samp{q}) and set (@samp{Q}). These packets are
40907 described fully in @ref{General Query Packets}.
40910 @cindex @samp{r} packet
40911 Reset the entire system.
40913 Don't use this packet; use the @samp{R} packet instead.
40916 @cindex @samp{R} packet
40917 Restart the program being debugged. The @var{XX}, while needed, is ignored.
40918 This packet is only available in extended mode (@pxref{extended mode}).
40920 The @samp{R} packet has no reply.
40922 @item s @r{[}@var{addr}@r{]}
40923 @cindex @samp{s} packet
40924 Single step, resuming at @var{addr}. If
40925 @var{addr} is omitted, resume at same address.
40927 This packet is deprecated for multi-threading support. @xref{vCont
40931 @xref{Stop Reply Packets}, for the reply specifications.
40933 @item S @var{sig}@r{[};@var{addr}@r{]}
40934 @anchor{step with signal packet}
40935 @cindex @samp{S} packet
40936 Step with signal. This is analogous to the @samp{C} packet, but
40937 requests a single-step, rather than a normal resumption of execution.
40939 This packet is deprecated for multi-threading support. @xref{vCont
40943 @xref{Stop Reply Packets}, for the reply specifications.
40945 @item t @var{addr}:@var{PP},@var{MM}
40946 @cindex @samp{t} packet
40947 Search backwards starting at address @var{addr} for a match with pattern
40948 @var{PP} and mask @var{MM}, both of which are are 4 byte long.
40949 There must be at least 3 digits in @var{addr}.
40951 @item T @var{thread-id}
40952 @cindex @samp{T} packet
40953 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
40958 thread is still alive
40964 Packets starting with @samp{v} are identified by a multi-letter name,
40965 up to the first @samp{;} or @samp{?} (or the end of the packet).
40967 @item vAttach;@var{pid}
40968 @cindex @samp{vAttach} packet
40969 Attach to a new process with the specified process ID @var{pid}.
40970 The process ID is a
40971 hexadecimal integer identifying the process. In all-stop mode, all
40972 threads in the attached process are stopped; in non-stop mode, it may be
40973 attached without being stopped if that is supported by the target.
40975 @c In non-stop mode, on a successful vAttach, the stub should set the
40976 @c current thread to a thread of the newly-attached process. After
40977 @c attaching, GDB queries for the attached process's thread ID with qC.
40978 @c Also note that, from a user perspective, whether or not the
40979 @c target is stopped on attach in non-stop mode depends on whether you
40980 @c use the foreground or background version of the attach command, not
40981 @c on what vAttach does; GDB does the right thing with respect to either
40982 @c stopping or restarting threads.
40984 This packet is only available in extended mode (@pxref{extended mode}).
40990 @item @r{Any stop packet}
40991 for success in all-stop mode (@pxref{Stop Reply Packets})
40993 for success in non-stop mode (@pxref{Remote Non-Stop})
40996 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
40997 @cindex @samp{vCont} packet
40998 @anchor{vCont packet}
40999 Resume the inferior, specifying different actions for each thread.
41001 For each inferior thread, the leftmost action with a matching
41002 @var{thread-id} is applied. Threads that don't match any action
41003 remain in their current state. Thread IDs are specified using the
41004 syntax described in @ref{thread-id syntax}. If multiprocess
41005 extensions (@pxref{multiprocess extensions}) are supported, actions
41006 can be specified to match all threads in a process by using the
41007 @samp{p@var{pid}.-1} form of the @var{thread-id}. An action with no
41008 @var{thread-id} matches all threads. Specifying no actions is an
41011 Currently supported actions are:
41017 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
41021 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
41024 @item r @var{start},@var{end}
41025 Step once, and then keep stepping as long as the thread stops at
41026 addresses between @var{start} (inclusive) and @var{end} (exclusive).
41027 The remote stub reports a stop reply when either the thread goes out
41028 of the range or is stopped due to an unrelated reason, such as hitting
41029 a breakpoint. @xref{range stepping}.
41031 If the range is empty (@var{start} == @var{end}), then the action
41032 becomes equivalent to the @samp{s} action. In other words,
41033 single-step once, and report the stop (even if the stepped instruction
41034 jumps to @var{start}).
41036 (A stop reply may be sent at any point even if the PC is still within
41037 the stepping range; for example, it is valid to implement this packet
41038 in a degenerate way as a single instruction step operation.)
41042 The optional argument @var{addr} normally associated with the
41043 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
41044 not supported in @samp{vCont}.
41046 The @samp{t} action is only relevant in non-stop mode
41047 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
41048 A stop reply should be generated for any affected thread not already stopped.
41049 When a thread is stopped by means of a @samp{t} action,
41050 the corresponding stop reply should indicate that the thread has stopped with
41051 signal @samp{0}, regardless of whether the target uses some other signal
41052 as an implementation detail.
41054 The server must ignore @samp{c}, @samp{C}, @samp{s}, @samp{S}, and
41055 @samp{r} actions for threads that are already running. Conversely,
41056 the server must ignore @samp{t} actions for threads that are already
41059 @emph{Note:} In non-stop mode, a thread is considered running until
41060 @value{GDBN} acknowledges an asynchronous stop notification for it with
41061 the @samp{vStopped} packet (@pxref{Remote Non-Stop}).
41063 The stub must support @samp{vCont} if it reports support for
41064 multiprocess extensions (@pxref{multiprocess extensions}).
41067 @xref{Stop Reply Packets}, for the reply specifications.
41070 @cindex @samp{vCont?} packet
41071 Request a list of actions supported by the @samp{vCont} packet.
41075 @item vCont@r{[};@var{action}@dots{}@r{]}
41076 The @samp{vCont} packet is supported. Each @var{action} is a supported
41077 command in the @samp{vCont} packet.
41079 The @samp{vCont} packet is not supported.
41082 @anchor{vCtrlC packet}
41084 @cindex @samp{vCtrlC} packet
41085 Interrupt remote target as if a control-C was pressed on the remote
41086 terminal. This is the equivalent to reacting to the @code{^C}
41087 (@samp{\003}, the control-C character) character in all-stop mode
41088 while the target is running, except this works in non-stop mode.
41089 @xref{interrupting remote targets}, for more info on the all-stop
41100 @item vFile:@var{operation}:@var{parameter}@dots{}
41101 @cindex @samp{vFile} packet
41102 Perform a file operation on the target system. For details,
41103 see @ref{Host I/O Packets}.
41105 @item vFlashErase:@var{addr},@var{length}
41106 @cindex @samp{vFlashErase} packet
41107 Direct the stub to erase @var{length} bytes of flash starting at
41108 @var{addr}. The region may enclose any number of flash blocks, but
41109 its start and end must fall on block boundaries, as indicated by the
41110 flash block size appearing in the memory map (@pxref{Memory Map
41111 Format}). @value{GDBN} groups flash memory programming operations
41112 together, and sends a @samp{vFlashDone} request after each group; the
41113 stub is allowed to delay erase operation until the @samp{vFlashDone}
41114 packet is received.
41124 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
41125 @cindex @samp{vFlashWrite} packet
41126 Direct the stub to write data to flash address @var{addr}. The data
41127 is passed in binary form using the same encoding as for the @samp{X}
41128 packet (@pxref{Binary Data}). The memory ranges specified by
41129 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
41130 not overlap, and must appear in order of increasing addresses
41131 (although @samp{vFlashErase} packets for higher addresses may already
41132 have been received; the ordering is guaranteed only between
41133 @samp{vFlashWrite} packets). If a packet writes to an address that was
41134 neither erased by a preceding @samp{vFlashErase} packet nor by some other
41135 target-specific method, the results are unpredictable.
41143 for vFlashWrite addressing non-flash memory
41149 @cindex @samp{vFlashDone} packet
41150 Indicate to the stub that flash programming operation is finished.
41151 The stub is permitted to delay or batch the effects of a group of
41152 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
41153 @samp{vFlashDone} packet is received. The contents of the affected
41154 regions of flash memory are unpredictable until the @samp{vFlashDone}
41155 request is completed.
41157 @item vKill;@var{pid}
41158 @cindex @samp{vKill} packet
41159 @anchor{vKill packet}
41160 Kill the process with the specified process ID @var{pid}, which is a
41161 hexadecimal integer identifying the process. This packet is used in
41162 preference to @samp{k} when multiprocess protocol extensions are
41163 supported; see @ref{multiprocess extensions}.
41173 @item vMustReplyEmpty
41174 @cindex @samp{vMustReplyEmpty} packet
41175 The correct reply to an unknown @samp{v} packet is to return the empty
41176 string, however, some older versions of @command{gdbserver} would
41177 incorrectly return @samp{OK} for unknown @samp{v} packets.
41179 The @samp{vMustReplyEmpty} is used as a feature test to check how
41180 @command{gdbserver} handles unknown packets, it is important that this
41181 packet be handled in the same way as other unknown @samp{v} packets.
41182 If this packet is handled differently to other unknown @samp{v}
41183 packets then it is possible that @value{GDBN} may run into problems in
41184 other areas, specifically around use of @samp{vFile:setfs:}.
41186 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
41187 @cindex @samp{vRun} packet
41188 Run the program @var{filename}, passing it each @var{argument} on its
41189 command line. The file and arguments are hex-encoded strings. If
41190 @var{filename} is an empty string, the stub may use a default program
41191 (e.g.@: the last program run). The program is created in the stopped
41194 @c FIXME: What about non-stop mode?
41196 This packet is only available in extended mode (@pxref{extended mode}).
41202 @item @r{Any stop packet}
41203 for success (@pxref{Stop Reply Packets})
41207 @cindex @samp{vStopped} packet
41208 @xref{Notification Packets}.
41210 @item X @var{addr},@var{length}:@var{XX@dots{}}
41212 @cindex @samp{X} packet
41213 Write data to memory, where the data is transmitted in binary.
41214 Memory is specified by its address @var{addr} and number of addressable memory
41215 units @var{length} (@pxref{addressable memory unit});
41216 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
41226 @item z @var{type},@var{addr},@var{kind}
41227 @itemx Z @var{type},@var{addr},@var{kind}
41228 @anchor{insert breakpoint or watchpoint packet}
41229 @cindex @samp{z} packet
41230 @cindex @samp{Z} packets
41231 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
41232 watchpoint starting at address @var{address} of kind @var{kind}.
41234 Each breakpoint and watchpoint packet @var{type} is documented
41237 @emph{Implementation notes: A remote target shall return an empty string
41238 for an unrecognized breakpoint or watchpoint packet @var{type}. A
41239 remote target shall support either both or neither of a given
41240 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
41241 avoid potential problems with duplicate packets, the operations should
41242 be implemented in an idempotent way.}
41244 @item z0,@var{addr},@var{kind}
41245 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
41246 @cindex @samp{z0} packet
41247 @cindex @samp{Z0} packet
41248 Insert (@samp{Z0}) or remove (@samp{z0}) a software breakpoint at address
41249 @var{addr} of type @var{kind}.
41251 A software breakpoint is implemented by replacing the instruction at
41252 @var{addr} with a software breakpoint or trap instruction. The
41253 @var{kind} is target-specific and typically indicates the size of the
41254 breakpoint in bytes that should be inserted. E.g., the @sc{arm} and
41255 @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
41256 architectures have additional meanings for @var{kind}
41257 (@pxref{Architecture-Specific Protocol Details}); if no
41258 architecture-specific value is being used, it should be @samp{0}.
41259 @var{kind} is hex-encoded. @var{cond_list} is an optional list of
41260 conditional expressions in bytecode form that should be evaluated on
41261 the target's side. These are the conditions that should be taken into
41262 consideration when deciding if the breakpoint trigger should be
41263 reported back to @value{GDBN}.
41265 See also the @samp{swbreak} stop reason (@pxref{swbreak stop reason})
41266 for how to best report a software breakpoint event to @value{GDBN}.
41268 The @var{cond_list} parameter is comprised of a series of expressions,
41269 concatenated without separators. Each expression has the following form:
41273 @item X @var{len},@var{expr}
41274 @var{len} is the length of the bytecode expression and @var{expr} is the
41275 actual conditional expression in bytecode form.
41279 The optional @var{cmd_list} parameter introduces commands that may be
41280 run on the target, rather than being reported back to @value{GDBN}.
41281 The parameter starts with a numeric flag @var{persist}; if the flag is
41282 nonzero, then the breakpoint may remain active and the commands
41283 continue to be run even when @value{GDBN} disconnects from the target.
41284 Following this flag is a series of expressions concatenated with no
41285 separators. Each expression has the following form:
41289 @item X @var{len},@var{expr}
41290 @var{len} is the length of the bytecode expression and @var{expr} is the
41291 actual commands expression in bytecode form.
41295 @emph{Implementation note: It is possible for a target to copy or move
41296 code that contains software breakpoints (e.g., when implementing
41297 overlays). The behavior of this packet, in the presence of such a
41298 target, is not defined.}
41310 @item z1,@var{addr},@var{kind}
41311 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
41312 @cindex @samp{z1} packet
41313 @cindex @samp{Z1} packet
41314 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
41315 address @var{addr}.
41317 A hardware breakpoint is implemented using a mechanism that is not
41318 dependent on being able to modify the target's memory. The
41319 @var{kind}, @var{cond_list}, and @var{cmd_list} arguments have the
41320 same meaning as in @samp{Z0} packets.
41322 @emph{Implementation note: A hardware breakpoint is not affected by code
41335 @item z2,@var{addr},@var{kind}
41336 @itemx Z2,@var{addr},@var{kind}
41337 @cindex @samp{z2} packet
41338 @cindex @samp{Z2} packet
41339 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
41340 The number of bytes to watch is specified by @var{kind}.
41352 @item z3,@var{addr},@var{kind}
41353 @itemx Z3,@var{addr},@var{kind}
41354 @cindex @samp{z3} packet
41355 @cindex @samp{Z3} packet
41356 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
41357 The number of bytes to watch is specified by @var{kind}.
41369 @item z4,@var{addr},@var{kind}
41370 @itemx Z4,@var{addr},@var{kind}
41371 @cindex @samp{z4} packet
41372 @cindex @samp{Z4} packet
41373 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
41374 The number of bytes to watch is specified by @var{kind}.
41388 @node Stop Reply Packets
41389 @section Stop Reply Packets
41390 @cindex stop reply packets
41392 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
41393 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
41394 receive any of the below as a reply. Except for @samp{?}
41395 and @samp{vStopped}, that reply is only returned
41396 when the target halts. In the below the exact meaning of @dfn{signal
41397 number} is defined by the header @file{include/gdb/signals.h} in the
41398 @value{GDBN} source code.
41400 In non-stop mode, the server will simply reply @samp{OK} to commands
41401 such as @samp{vCont}; any stop will be the subject of a future
41402 notification. @xref{Remote Non-Stop}.
41404 As in the description of request packets, we include spaces in the
41405 reply templates for clarity; these are not part of the reply packet's
41406 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
41412 The program received signal number @var{AA} (a two-digit hexadecimal
41413 number). This is equivalent to a @samp{T} response with no
41414 @var{n}:@var{r} pairs.
41416 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
41417 @cindex @samp{T} packet reply
41418 The program received signal number @var{AA} (a two-digit hexadecimal
41419 number). This is equivalent to an @samp{S} response, except that the
41420 @samp{@var{n}:@var{r}} pairs can carry values of important registers
41421 and other information directly in the stop reply packet, reducing
41422 round-trip latency. Single-step and breakpoint traps are reported
41423 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
41427 If @var{n} is a hexadecimal number, it is a register number, and the
41428 corresponding @var{r} gives that register's value. The data @var{r} is a
41429 series of bytes in target byte order, with each byte given by a
41430 two-digit hex number.
41433 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
41434 the stopped thread, as specified in @ref{thread-id syntax}.
41437 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
41438 the core on which the stop event was detected.
41441 If @var{n} is a recognized @dfn{stop reason}, it describes a more
41442 specific event that stopped the target. The currently defined stop
41443 reasons are listed below. The @var{aa} should be @samp{05}, the trap
41444 signal. At most one stop reason should be present.
41447 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
41448 and go on to the next; this allows us to extend the protocol in the
41452 The currently defined stop reasons are:
41458 The packet indicates a watchpoint hit, and @var{r} is the data address, in
41461 @item syscall_entry
41462 @itemx syscall_return
41463 The packet indicates a syscall entry or return, and @var{r} is the
41464 syscall number, in hex.
41466 @cindex shared library events, remote reply
41468 The packet indicates that the loaded libraries have changed.
41469 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
41470 list of loaded libraries. The @var{r} part is ignored.
41472 @cindex replay log events, remote reply
41474 The packet indicates that the target cannot continue replaying
41475 logged execution events, because it has reached the end (or the
41476 beginning when executing backward) of the log. The value of @var{r}
41477 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
41478 for more information.
41481 @anchor{swbreak stop reason}
41482 The packet indicates a software breakpoint instruction was executed,
41483 irrespective of whether it was @value{GDBN} that planted the
41484 breakpoint or the breakpoint is hardcoded in the program. The @var{r}
41485 part must be left empty.
41487 On some architectures, such as x86, at the architecture level, when a
41488 breakpoint instruction executes the program counter points at the
41489 breakpoint address plus an offset. On such targets, the stub is
41490 responsible for adjusting the PC to point back at the breakpoint
41493 This packet should not be sent by default; older @value{GDBN} versions
41494 did not support it. @value{GDBN} requests it, by supplying an
41495 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
41496 remote stub must also supply the appropriate @samp{qSupported} feature
41497 indicating support.
41499 This packet is required for correct non-stop mode operation.
41502 The packet indicates the target stopped for a hardware breakpoint.
41503 The @var{r} part must be left empty.
41505 The same remarks about @samp{qSupported} and non-stop mode above
41508 @cindex fork events, remote reply
41510 The packet indicates that @code{fork} was called, and @var{r}
41511 is the thread ID of the new child process. Refer to
41512 @ref{thread-id syntax} for the format of the @var{thread-id}
41513 field. This packet is only applicable to targets that support
41516 This packet should not be sent by default; older @value{GDBN} versions
41517 did not support it. @value{GDBN} requests it, by supplying an
41518 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
41519 remote stub must also supply the appropriate @samp{qSupported} feature
41520 indicating support.
41522 @cindex vfork events, remote reply
41524 The packet indicates that @code{vfork} was called, and @var{r}
41525 is the thread ID of the new child process. Refer to
41526 @ref{thread-id syntax} for the format of the @var{thread-id}
41527 field. This packet is only applicable to targets that support
41530 This packet should not be sent by default; older @value{GDBN} versions
41531 did not support it. @value{GDBN} requests it, by supplying an
41532 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
41533 remote stub must also supply the appropriate @samp{qSupported} feature
41534 indicating support.
41536 @cindex vforkdone events, remote reply
41538 The packet indicates that a child process created by a vfork
41539 has either called @code{exec} or terminated, so that the
41540 address spaces of the parent and child process are no longer
41541 shared. The @var{r} part is ignored. This packet is only
41542 applicable to targets that support vforkdone events.
41544 This packet should not be sent by default; older @value{GDBN} versions
41545 did not support it. @value{GDBN} requests it, by supplying an
41546 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
41547 remote stub must also supply the appropriate @samp{qSupported} feature
41548 indicating support.
41550 @cindex exec events, remote reply
41552 The packet indicates that @code{execve} was called, and @var{r}
41553 is the absolute pathname of the file that was executed, in hex.
41554 This packet is only applicable to targets that support exec events.
41556 This packet should not be sent by default; older @value{GDBN} versions
41557 did not support it. @value{GDBN} requests it, by supplying an
41558 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
41559 remote stub must also supply the appropriate @samp{qSupported} feature
41560 indicating support.
41562 @cindex thread create event, remote reply
41563 @anchor{thread create event}
41565 The packet indicates that the thread was just created. The new thread
41566 is stopped until @value{GDBN} sets it running with a resumption packet
41567 (@pxref{vCont packet}). This packet should not be sent by default;
41568 @value{GDBN} requests it with the @ref{QThreadEvents} packet. See
41569 also the @samp{w} (@pxref{thread exit event}) remote reply below. The
41570 @var{r} part is ignored.
41575 @itemx W @var{AA} ; process:@var{pid}
41576 The process exited, and @var{AA} is the exit status. This is only
41577 applicable to certain targets.
41579 The second form of the response, including the process ID of the
41580 exited process, can be used only when @value{GDBN} has reported
41581 support for multiprocess protocol extensions; see @ref{multiprocess
41582 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
41586 @itemx X @var{AA} ; process:@var{pid}
41587 The process terminated with signal @var{AA}.
41589 The second form of the response, including the process ID of the
41590 terminated process, can be used only when @value{GDBN} has reported
41591 support for multiprocess protocol extensions; see @ref{multiprocess
41592 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
41595 @anchor{thread exit event}
41596 @cindex thread exit event, remote reply
41597 @item w @var{AA} ; @var{tid}
41599 The thread exited, and @var{AA} is the exit status. This response
41600 should not be sent by default; @value{GDBN} requests it with the
41601 @ref{QThreadEvents} packet. See also @ref{thread create event} above.
41602 @var{AA} is formatted as a big-endian hex string.
41605 There are no resumed threads left in the target. In other words, even
41606 though the process is alive, the last resumed thread has exited. For
41607 example, say the target process has two threads: thread 1 and thread
41608 2. The client leaves thread 1 stopped, and resumes thread 2, which
41609 subsequently exits. At this point, even though the process is still
41610 alive, and thus no @samp{W} stop reply is sent, no thread is actually
41611 executing either. The @samp{N} stop reply thus informs the client
41612 that it can stop waiting for stop replies. This packet should not be
41613 sent by default; older @value{GDBN} versions did not support it.
41614 @value{GDBN} requests it, by supplying an appropriate
41615 @samp{qSupported} feature (@pxref{qSupported}). The remote stub must
41616 also supply the appropriate @samp{qSupported} feature indicating
41619 @item O @var{XX}@dots{}
41620 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
41621 written as the program's console output. This can happen at any time
41622 while the program is running and the debugger should continue to wait
41623 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
41625 @item F @var{call-id},@var{parameter}@dots{}
41626 @var{call-id} is the identifier which says which host system call should
41627 be called. This is just the name of the function. Translation into the
41628 correct system call is only applicable as it's defined in @value{GDBN}.
41629 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
41632 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
41633 this very system call.
41635 The target replies with this packet when it expects @value{GDBN} to
41636 call a host system call on behalf of the target. @value{GDBN} replies
41637 with an appropriate @samp{F} packet and keeps up waiting for the next
41638 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
41639 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
41640 Protocol Extension}, for more details.
41644 @node General Query Packets
41645 @section General Query Packets
41646 @cindex remote query requests
41648 Packets starting with @samp{q} are @dfn{general query packets};
41649 packets starting with @samp{Q} are @dfn{general set packets}. General
41650 query and set packets are a semi-unified form for retrieving and
41651 sending information to and from the stub.
41653 The initial letter of a query or set packet is followed by a name
41654 indicating what sort of thing the packet applies to. For example,
41655 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
41656 definitions with the stub. These packet names follow some
41661 The name must not contain commas, colons or semicolons.
41663 Most @value{GDBN} query and set packets have a leading upper case
41666 The names of custom vendor packets should use a company prefix, in
41667 lower case, followed by a period. For example, packets designed at
41668 the Acme Corporation might begin with @samp{qacme.foo} (for querying
41669 foos) or @samp{Qacme.bar} (for setting bars).
41672 The name of a query or set packet should be separated from any
41673 parameters by a @samp{:}; the parameters themselves should be
41674 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
41675 full packet name, and check for a separator or the end of the packet,
41676 in case two packet names share a common prefix. New packets should not begin
41677 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
41678 packets predate these conventions, and have arguments without any terminator
41679 for the packet name; we suspect they are in widespread use in places that
41680 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
41681 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
41684 Like the descriptions of the other packets, each description here
41685 has a template showing the packet's overall syntax, followed by an
41686 explanation of the packet's meaning. We include spaces in some of the
41687 templates for clarity; these are not part of the packet's syntax. No
41688 @value{GDBN} packet uses spaces to separate its components.
41690 Here are the currently defined query and set packets:
41696 Turn on or off the agent as a helper to perform some debugging operations
41697 delegated from @value{GDBN} (@pxref{Control Agent}).
41699 @item QAllow:@var{op}:@var{val}@dots{}
41700 @cindex @samp{QAllow} packet
41701 Specify which operations @value{GDBN} expects to request of the
41702 target, as a semicolon-separated list of operation name and value
41703 pairs. Possible values for @var{op} include @samp{WriteReg},
41704 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
41705 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
41706 indicating that @value{GDBN} will not request the operation, or 1,
41707 indicating that it may. (The target can then use this to set up its
41708 own internals optimally, for instance if the debugger never expects to
41709 insert breakpoints, it may not need to install its own trap handler.)
41712 @cindex current thread, remote request
41713 @cindex @samp{qC} packet
41714 Return the current thread ID.
41718 @item QC @var{thread-id}
41719 Where @var{thread-id} is a thread ID as documented in
41720 @ref{thread-id syntax}.
41721 @item @r{(anything else)}
41722 Any other reply implies the old thread ID.
41725 @item qCRC:@var{addr},@var{length}
41726 @cindex CRC of memory block, remote request
41727 @cindex @samp{qCRC} packet
41728 @anchor{qCRC packet}
41729 Compute the CRC checksum of a block of memory using CRC-32 defined in
41730 IEEE 802.3. The CRC is computed byte at a time, taking the most
41731 significant bit of each byte first. The initial pattern code
41732 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
41734 @emph{Note:} This is the same CRC used in validating separate debug
41735 files (@pxref{Separate Debug Files, , Debugging Information in Separate
41736 Files}). However the algorithm is slightly different. When validating
41737 separate debug files, the CRC is computed taking the @emph{least}
41738 significant bit of each byte first, and the final result is inverted to
41739 detect trailing zeros.
41744 An error (such as memory fault)
41745 @item C @var{crc32}
41746 The specified memory region's checksum is @var{crc32}.
41749 @item QDisableRandomization:@var{value}
41750 @cindex disable address space randomization, remote request
41751 @cindex @samp{QDisableRandomization} packet
41752 Some target operating systems will randomize the virtual address space
41753 of the inferior process as a security feature, but provide a feature
41754 to disable such randomization, e.g.@: to allow for a more deterministic
41755 debugging experience. On such systems, this packet with a @var{value}
41756 of 1 directs the target to disable address space randomization for
41757 processes subsequently started via @samp{vRun} packets, while a packet
41758 with a @var{value} of 0 tells the target to enable address space
41761 This packet is only available in extended mode (@pxref{extended mode}).
41766 The request succeeded.
41769 An error occurred. The error number @var{nn} is given as hex digits.
41772 An empty reply indicates that @samp{QDisableRandomization} is not supported
41776 This packet is not probed by default; the remote stub must request it,
41777 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
41778 This should only be done on targets that actually support disabling
41779 address space randomization.
41781 @item QStartupWithShell:@var{value}
41782 @cindex startup with shell, remote request
41783 @cindex @samp{QStartupWithShell} packet
41784 On UNIX-like targets, it is possible to start the inferior using a
41785 shell program. This is the default behavior on both @value{GDBN} and
41786 @command{gdbserver} (@pxref{set startup-with-shell}). This packet is
41787 used to inform @command{gdbserver} whether it should start the
41788 inferior using a shell or not.
41790 If @var{value} is @samp{0}, @command{gdbserver} will not use a shell
41791 to start the inferior. If @var{value} is @samp{1},
41792 @command{gdbserver} will use a shell to start the inferior. All other
41793 values are considered an error.
41795 This packet is only available in extended mode (@pxref{extended
41801 The request succeeded.
41804 An error occurred. The error number @var{nn} is given as hex digits.
41807 This packet is not probed by default; the remote stub must request it,
41808 by supplying an appropriate @samp{qSupported} response
41809 (@pxref{qSupported}). This should only be done on targets that
41810 actually support starting the inferior using a shell.
41812 Use of this packet is controlled by the @code{set startup-with-shell}
41813 command; @pxref{set startup-with-shell}.
41815 @item QEnvironmentHexEncoded:@var{hex-value}
41816 @anchor{QEnvironmentHexEncoded}
41817 @cindex set environment variable, remote request
41818 @cindex @samp{QEnvironmentHexEncoded} packet
41819 On UNIX-like targets, it is possible to set environment variables that
41820 will be passed to the inferior during the startup process. This
41821 packet is used to inform @command{gdbserver} of an environment
41822 variable that has been defined by the user on @value{GDBN} (@pxref{set
41825 The packet is composed by @var{hex-value}, an hex encoded
41826 representation of the @var{name=value} format representing an
41827 environment variable. The name of the environment variable is
41828 represented by @var{name}, and the value to be assigned to the
41829 environment variable is represented by @var{value}. If the variable
41830 has no value (i.e., the value is @code{null}), then @var{value} will
41833 This packet is only available in extended mode (@pxref{extended
41839 The request succeeded.
41842 This packet is not probed by default; the remote stub must request it,
41843 by supplying an appropriate @samp{qSupported} response
41844 (@pxref{qSupported}). This should only be done on targets that
41845 actually support passing environment variables to the starting
41848 This packet is related to the @code{set environment} command;
41849 @pxref{set environment}.
41851 @item QEnvironmentUnset:@var{hex-value}
41852 @anchor{QEnvironmentUnset}
41853 @cindex unset environment variable, remote request
41854 @cindex @samp{QEnvironmentUnset} packet
41855 On UNIX-like targets, it is possible to unset environment variables
41856 before starting the inferior in the remote target. This packet is
41857 used to inform @command{gdbserver} of an environment variable that has
41858 been unset by the user on @value{GDBN} (@pxref{unset environment}).
41860 The packet is composed by @var{hex-value}, an hex encoded
41861 representation of the name of the environment variable to be unset.
41863 This packet is only available in extended mode (@pxref{extended
41869 The request succeeded.
41872 This packet is not probed by default; the remote stub must request it,
41873 by supplying an appropriate @samp{qSupported} response
41874 (@pxref{qSupported}). This should only be done on targets that
41875 actually support passing environment variables to the starting
41878 This packet is related to the @code{unset environment} command;
41879 @pxref{unset environment}.
41881 @item QEnvironmentReset
41882 @anchor{QEnvironmentReset}
41883 @cindex reset environment, remote request
41884 @cindex @samp{QEnvironmentReset} packet
41885 On UNIX-like targets, this packet is used to reset the state of
41886 environment variables in the remote target before starting the
41887 inferior. In this context, reset means unsetting all environment
41888 variables that were previously set by the user (i.e., were not
41889 initially present in the environment). It is sent to
41890 @command{gdbserver} before the @samp{QEnvironmentHexEncoded}
41891 (@pxref{QEnvironmentHexEncoded}) and the @samp{QEnvironmentUnset}
41892 (@pxref{QEnvironmentUnset}) packets.
41894 This packet is only available in extended mode (@pxref{extended
41900 The request succeeded.
41903 This packet is not probed by default; the remote stub must request it,
41904 by supplying an appropriate @samp{qSupported} response
41905 (@pxref{qSupported}). This should only be done on targets that
41906 actually support passing environment variables to the starting
41909 @item QSetWorkingDir:@r{[}@var{directory}@r{]}
41910 @anchor{QSetWorkingDir packet}
41911 @cindex set working directory, remote request
41912 @cindex @samp{QSetWorkingDir} packet
41913 This packet is used to inform the remote server of the intended
41914 current working directory for programs that are going to be executed.
41916 The packet is composed by @var{directory}, an hex encoded
41917 representation of the directory that the remote inferior will use as
41918 its current working directory. If @var{directory} is an empty string,
41919 the remote server should reset the inferior's current working
41920 directory to its original, empty value.
41922 This packet is only available in extended mode (@pxref{extended
41928 The request succeeded.
41932 @itemx qsThreadInfo
41933 @cindex list active threads, remote request
41934 @cindex @samp{qfThreadInfo} packet
41935 @cindex @samp{qsThreadInfo} packet
41936 Obtain a list of all active thread IDs from the target (OS). Since there
41937 may be too many active threads to fit into one reply packet, this query
41938 works iteratively: it may require more than one query/reply sequence to
41939 obtain the entire list of threads. The first query of the sequence will
41940 be the @samp{qfThreadInfo} query; subsequent queries in the
41941 sequence will be the @samp{qsThreadInfo} query.
41943 NOTE: This packet replaces the @samp{qL} query (see below).
41947 @item m @var{thread-id}
41949 @item m @var{thread-id},@var{thread-id}@dots{}
41950 a comma-separated list of thread IDs
41952 (lower case letter @samp{L}) denotes end of list.
41955 In response to each query, the target will reply with a list of one or
41956 more thread IDs, separated by commas.
41957 @value{GDBN} will respond to each reply with a request for more thread
41958 ids (using the @samp{qs} form of the query), until the target responds
41959 with @samp{l} (lower-case ell, for @dfn{last}).
41960 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
41963 @emph{Note: @value{GDBN} will send the @code{qfThreadInfo} query during the
41964 initial connection with the remote target, and the very first thread ID
41965 mentioned in the reply will be stopped by @value{GDBN} in a subsequent
41966 message. Therefore, the stub should ensure that the first thread ID in
41967 the @code{qfThreadInfo} reply is suitable for being stopped by @value{GDBN}.}
41969 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
41970 @cindex get thread-local storage address, remote request
41971 @cindex @samp{qGetTLSAddr} packet
41972 Fetch the address associated with thread local storage specified
41973 by @var{thread-id}, @var{offset}, and @var{lm}.
41975 @var{thread-id} is the thread ID associated with the
41976 thread for which to fetch the TLS address. @xref{thread-id syntax}.
41978 @var{offset} is the (big endian, hex encoded) offset associated with the
41979 thread local variable. (This offset is obtained from the debug
41980 information associated with the variable.)
41982 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
41983 load module associated with the thread local storage. For example,
41984 a @sc{gnu}/Linux system will pass the link map address of the shared
41985 object associated with the thread local storage under consideration.
41986 Other operating environments may choose to represent the load module
41987 differently, so the precise meaning of this parameter will vary.
41991 @item @var{XX}@dots{}
41992 Hex encoded (big endian) bytes representing the address of the thread
41993 local storage requested.
41996 An error occurred. The error number @var{nn} is given as hex digits.
41999 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
42002 @item qGetTIBAddr:@var{thread-id}
42003 @cindex get thread information block address
42004 @cindex @samp{qGetTIBAddr} packet
42005 Fetch address of the Windows OS specific Thread Information Block.
42007 @var{thread-id} is the thread ID associated with the thread.
42011 @item @var{XX}@dots{}
42012 Hex encoded (big endian) bytes representing the linear address of the
42013 thread information block.
42016 An error occured. This means that either the thread was not found, or the
42017 address could not be retrieved.
42020 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
42023 @item qL @var{startflag} @var{threadcount} @var{nextthread}
42024 Obtain thread information from RTOS. Where: @var{startflag} (one hex
42025 digit) is one to indicate the first query and zero to indicate a
42026 subsequent query; @var{threadcount} (two hex digits) is the maximum
42027 number of threads the response packet can contain; and @var{nextthread}
42028 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
42029 returned in the response as @var{argthread}.
42031 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
42035 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
42036 Where: @var{count} (two hex digits) is the number of threads being
42037 returned; @var{done} (one hex digit) is zero to indicate more threads
42038 and one indicates no further threads; @var{argthreadid} (eight hex
42039 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
42040 is a sequence of thread IDs, @var{threadid} (eight hex
42041 digits), from the target. See @code{remote.c:parse_threadlist_response()}.
42045 @cindex section offsets, remote request
42046 @cindex @samp{qOffsets} packet
42047 Get section offsets that the target used when relocating the downloaded
42052 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
42053 Relocate the @code{Text} section by @var{xxx} from its original address.
42054 Relocate the @code{Data} section by @var{yyy} from its original address.
42055 If the object file format provides segment information (e.g.@: @sc{elf}
42056 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
42057 segments by the supplied offsets.
42059 @emph{Note: while a @code{Bss} offset may be included in the response,
42060 @value{GDBN} ignores this and instead applies the @code{Data} offset
42061 to the @code{Bss} section.}
42063 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
42064 Relocate the first segment of the object file, which conventionally
42065 contains program code, to a starting address of @var{xxx}. If
42066 @samp{DataSeg} is specified, relocate the second segment, which
42067 conventionally contains modifiable data, to a starting address of
42068 @var{yyy}. @value{GDBN} will report an error if the object file
42069 does not contain segment information, or does not contain at least
42070 as many segments as mentioned in the reply. Extra segments are
42071 kept at fixed offsets relative to the last relocated segment.
42074 @item qP @var{mode} @var{thread-id}
42075 @cindex thread information, remote request
42076 @cindex @samp{qP} packet
42077 Returns information on @var{thread-id}. Where: @var{mode} is a hex
42078 encoded 32 bit mode; @var{thread-id} is a thread ID
42079 (@pxref{thread-id syntax}).
42081 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
42084 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
42088 @cindex non-stop mode, remote request
42089 @cindex @samp{QNonStop} packet
42091 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
42092 @xref{Remote Non-Stop}, for more information.
42097 The request succeeded.
42100 An error occurred. The error number @var{nn} is given as hex digits.
42103 An empty reply indicates that @samp{QNonStop} is not supported by
42107 This packet is not probed by default; the remote stub must request it,
42108 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
42109 Use of this packet is controlled by the @code{set non-stop} command;
42110 @pxref{Non-Stop Mode}.
42112 @item QCatchSyscalls:1 @r{[};@var{sysno}@r{]}@dots{}
42113 @itemx QCatchSyscalls:0
42114 @cindex catch syscalls from inferior, remote request
42115 @cindex @samp{QCatchSyscalls} packet
42116 @anchor{QCatchSyscalls}
42117 Enable (@samp{QCatchSyscalls:1}) or disable (@samp{QCatchSyscalls:0})
42118 catching syscalls from the inferior process.
42120 For @samp{QCatchSyscalls:1}, each listed syscall @var{sysno} (encoded
42121 in hex) should be reported to @value{GDBN}. If no syscall @var{sysno}
42122 is listed, every system call should be reported.
42124 Note that if a syscall not in the list is reported, @value{GDBN} will
42125 still filter the event according to its own list from all corresponding
42126 @code{catch syscall} commands. However, it is more efficient to only
42127 report the requested syscalls.
42129 Multiple @samp{QCatchSyscalls:1} packets do not combine; any earlier
42130 @samp{QCatchSyscalls:1} list is completely replaced by the new list.
42132 If the inferior process execs, the state of @samp{QCatchSyscalls} is
42133 kept for the new process too. On targets where exec may affect syscall
42134 numbers, for example with exec between 32 and 64-bit processes, the
42135 client should send a new packet with the new syscall list.
42140 The request succeeded.
42143 An error occurred. @var{nn} are hex digits.
42146 An empty reply indicates that @samp{QCatchSyscalls} is not supported by
42150 Use of this packet is controlled by the @code{set remote catch-syscalls}
42151 command (@pxref{Remote Configuration, set remote catch-syscalls}).
42152 This packet is not probed by default; the remote stub must request it,
42153 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
42155 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
42156 @cindex pass signals to inferior, remote request
42157 @cindex @samp{QPassSignals} packet
42158 @anchor{QPassSignals}
42159 Each listed @var{signal} should be passed directly to the inferior process.
42160 Signals are numbered identically to continue packets and stop replies
42161 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
42162 strictly greater than the previous item. These signals do not need to stop
42163 the inferior, or be reported to @value{GDBN}. All other signals should be
42164 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
42165 combine; any earlier @samp{QPassSignals} list is completely replaced by the
42166 new list. This packet improves performance when using @samp{handle
42167 @var{signal} nostop noprint pass}.
42172 The request succeeded.
42175 An error occurred. The error number @var{nn} is given as hex digits.
42178 An empty reply indicates that @samp{QPassSignals} is not supported by
42182 Use of this packet is controlled by the @code{set remote pass-signals}
42183 command (@pxref{Remote Configuration, set remote pass-signals}).
42184 This packet is not probed by default; the remote stub must request it,
42185 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
42187 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
42188 @cindex signals the inferior may see, remote request
42189 @cindex @samp{QProgramSignals} packet
42190 @anchor{QProgramSignals}
42191 Each listed @var{signal} may be delivered to the inferior process.
42192 Others should be silently discarded.
42194 In some cases, the remote stub may need to decide whether to deliver a
42195 signal to the program or not without @value{GDBN} involvement. One
42196 example of that is while detaching --- the program's threads may have
42197 stopped for signals that haven't yet had a chance of being reported to
42198 @value{GDBN}, and so the remote stub can use the signal list specified
42199 by this packet to know whether to deliver or ignore those pending
42202 This does not influence whether to deliver a signal as requested by a
42203 resumption packet (@pxref{vCont packet}).
42205 Signals are numbered identically to continue packets and stop replies
42206 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
42207 strictly greater than the previous item. Multiple
42208 @samp{QProgramSignals} packets do not combine; any earlier
42209 @samp{QProgramSignals} list is completely replaced by the new list.
42214 The request succeeded.
42217 An error occurred. The error number @var{nn} is given as hex digits.
42220 An empty reply indicates that @samp{QProgramSignals} is not supported
42224 Use of this packet is controlled by the @code{set remote program-signals}
42225 command (@pxref{Remote Configuration, set remote program-signals}).
42226 This packet is not probed by default; the remote stub must request it,
42227 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
42229 @anchor{QThreadEvents}
42230 @item QThreadEvents:1
42231 @itemx QThreadEvents:0
42232 @cindex thread create/exit events, remote request
42233 @cindex @samp{QThreadEvents} packet
42235 Enable (@samp{QThreadEvents:1}) or disable (@samp{QThreadEvents:0})
42236 reporting of thread create and exit events. @xref{thread create
42237 event}, for the reply specifications. For example, this is used in
42238 non-stop mode when @value{GDBN} stops a set of threads and
42239 synchronously waits for the their corresponding stop replies. Without
42240 exit events, if one of the threads exits, @value{GDBN} would hang
42241 forever not knowing that it should no longer expect a stop for that
42242 same thread. @value{GDBN} does not enable this feature unless the
42243 stub reports that it supports it by including @samp{QThreadEvents+} in
42244 its @samp{qSupported} reply.
42249 The request succeeded.
42252 An error occurred. The error number @var{nn} is given as hex digits.
42255 An empty reply indicates that @samp{QThreadEvents} is not supported by
42259 Use of this packet is controlled by the @code{set remote thread-events}
42260 command (@pxref{Remote Configuration, set remote thread-events}).
42262 @item qRcmd,@var{command}
42263 @cindex execute remote command, remote request
42264 @cindex @samp{qRcmd} packet
42265 @var{command} (hex encoded) is passed to the local interpreter for
42266 execution. Invalid commands should be reported using the output
42267 string. Before the final result packet, the target may also respond
42268 with a number of intermediate @samp{O@var{output}} console output
42269 packets. @emph{Implementors should note that providing access to a
42270 stubs's interpreter may have security implications}.
42275 A command response with no output.
42277 A command response with the hex encoded output string @var{OUTPUT}.
42279 Indicate a badly formed request.
42281 An empty reply indicates that @samp{qRcmd} is not recognized.
42284 (Note that the @code{qRcmd} packet's name is separated from the
42285 command by a @samp{,}, not a @samp{:}, contrary to the naming
42286 conventions above. Please don't use this packet as a model for new
42289 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
42290 @cindex searching memory, in remote debugging
42292 @cindex @samp{qSearch:memory} packet
42294 @cindex @samp{qSearch memory} packet
42295 @anchor{qSearch memory}
42296 Search @var{length} bytes at @var{address} for @var{search-pattern}.
42297 Both @var{address} and @var{length} are encoded in hex;
42298 @var{search-pattern} is a sequence of bytes, also hex encoded.
42303 The pattern was not found.
42305 The pattern was found at @var{address}.
42307 A badly formed request or an error was encountered while searching memory.
42309 An empty reply indicates that @samp{qSearch:memory} is not recognized.
42312 @item QStartNoAckMode
42313 @cindex @samp{QStartNoAckMode} packet
42314 @anchor{QStartNoAckMode}
42315 Request that the remote stub disable the normal @samp{+}/@samp{-}
42316 protocol acknowledgments (@pxref{Packet Acknowledgment}).
42321 The stub has switched to no-acknowledgment mode.
42322 @value{GDBN} acknowledges this response,
42323 but neither the stub nor @value{GDBN} shall send or expect further
42324 @samp{+}/@samp{-} acknowledgments in the current connection.
42326 An empty reply indicates that the stub does not support no-acknowledgment mode.
42329 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
42330 @cindex supported packets, remote query
42331 @cindex features of the remote protocol
42332 @cindex @samp{qSupported} packet
42333 @anchor{qSupported}
42334 Tell the remote stub about features supported by @value{GDBN}, and
42335 query the stub for features it supports. This packet allows
42336 @value{GDBN} and the remote stub to take advantage of each others'
42337 features. @samp{qSupported} also consolidates multiple feature probes
42338 at startup, to improve @value{GDBN} performance---a single larger
42339 packet performs better than multiple smaller probe packets on
42340 high-latency links. Some features may enable behavior which must not
42341 be on by default, e.g.@: because it would confuse older clients or
42342 stubs. Other features may describe packets which could be
42343 automatically probed for, but are not. These features must be
42344 reported before @value{GDBN} will use them. This ``default
42345 unsupported'' behavior is not appropriate for all packets, but it
42346 helps to keep the initial connection time under control with new
42347 versions of @value{GDBN} which support increasing numbers of packets.
42351 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
42352 The stub supports or does not support each returned @var{stubfeature},
42353 depending on the form of each @var{stubfeature} (see below for the
42356 An empty reply indicates that @samp{qSupported} is not recognized,
42357 or that no features needed to be reported to @value{GDBN}.
42360 The allowed forms for each feature (either a @var{gdbfeature} in the
42361 @samp{qSupported} packet, or a @var{stubfeature} in the response)
42365 @item @var{name}=@var{value}
42366 The remote protocol feature @var{name} is supported, and associated
42367 with the specified @var{value}. The format of @var{value} depends
42368 on the feature, but it must not include a semicolon.
42370 The remote protocol feature @var{name} is supported, and does not
42371 need an associated value.
42373 The remote protocol feature @var{name} is not supported.
42375 The remote protocol feature @var{name} may be supported, and
42376 @value{GDBN} should auto-detect support in some other way when it is
42377 needed. This form will not be used for @var{gdbfeature} notifications,
42378 but may be used for @var{stubfeature} responses.
42381 Whenever the stub receives a @samp{qSupported} request, the
42382 supplied set of @value{GDBN} features should override any previous
42383 request. This allows @value{GDBN} to put the stub in a known
42384 state, even if the stub had previously been communicating with
42385 a different version of @value{GDBN}.
42387 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
42392 This feature indicates whether @value{GDBN} supports multiprocess
42393 extensions to the remote protocol. @value{GDBN} does not use such
42394 extensions unless the stub also reports that it supports them by
42395 including @samp{multiprocess+} in its @samp{qSupported} reply.
42396 @xref{multiprocess extensions}, for details.
42399 This feature indicates that @value{GDBN} supports the XML target
42400 description. If the stub sees @samp{xmlRegisters=} with target
42401 specific strings separated by a comma, it will report register
42405 This feature indicates whether @value{GDBN} supports the
42406 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
42407 instruction reply packet}).
42410 This feature indicates whether @value{GDBN} supports the swbreak stop
42411 reason in stop replies. @xref{swbreak stop reason}, for details.
42414 This feature indicates whether @value{GDBN} supports the hwbreak stop
42415 reason in stop replies. @xref{swbreak stop reason}, for details.
42418 This feature indicates whether @value{GDBN} supports fork event
42419 extensions to the remote protocol. @value{GDBN} does not use such
42420 extensions unless the stub also reports that it supports them by
42421 including @samp{fork-events+} in its @samp{qSupported} reply.
42424 This feature indicates whether @value{GDBN} supports vfork event
42425 extensions to the remote protocol. @value{GDBN} does not use such
42426 extensions unless the stub also reports that it supports them by
42427 including @samp{vfork-events+} in its @samp{qSupported} reply.
42430 This feature indicates whether @value{GDBN} supports exec event
42431 extensions to the remote protocol. @value{GDBN} does not use such
42432 extensions unless the stub also reports that it supports them by
42433 including @samp{exec-events+} in its @samp{qSupported} reply.
42435 @item vContSupported
42436 This feature indicates whether @value{GDBN} wants to know the
42437 supported actions in the reply to @samp{vCont?} packet.
42440 Stubs should ignore any unknown values for
42441 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
42442 packet supports receiving packets of unlimited length (earlier
42443 versions of @value{GDBN} may reject overly long responses). Additional values
42444 for @var{gdbfeature} may be defined in the future to let the stub take
42445 advantage of new features in @value{GDBN}, e.g.@: incompatible
42446 improvements in the remote protocol---the @samp{multiprocess} feature is
42447 an example of such a feature. The stub's reply should be independent
42448 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
42449 describes all the features it supports, and then the stub replies with
42450 all the features it supports.
42452 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
42453 responses, as long as each response uses one of the standard forms.
42455 Some features are flags. A stub which supports a flag feature
42456 should respond with a @samp{+} form response. Other features
42457 require values, and the stub should respond with an @samp{=}
42460 Each feature has a default value, which @value{GDBN} will use if
42461 @samp{qSupported} is not available or if the feature is not mentioned
42462 in the @samp{qSupported} response. The default values are fixed; a
42463 stub is free to omit any feature responses that match the defaults.
42465 Not all features can be probed, but for those which can, the probing
42466 mechanism is useful: in some cases, a stub's internal
42467 architecture may not allow the protocol layer to know some information
42468 about the underlying target in advance. This is especially common in
42469 stubs which may be configured for multiple targets.
42471 These are the currently defined stub features and their properties:
42473 @multitable @columnfractions 0.35 0.2 0.12 0.2
42474 @c NOTE: The first row should be @headitem, but we do not yet require
42475 @c a new enough version of Texinfo (4.7) to use @headitem.
42477 @tab Value Required
42481 @item @samp{PacketSize}
42486 @item @samp{qXfer:auxv:read}
42491 @item @samp{qXfer:btrace:read}
42496 @item @samp{qXfer:btrace-conf:read}
42501 @item @samp{qXfer:exec-file:read}
42506 @item @samp{qXfer:features:read}
42511 @item @samp{qXfer:libraries:read}
42516 @item @samp{qXfer:libraries-svr4:read}
42521 @item @samp{augmented-libraries-svr4-read}
42526 @item @samp{qXfer:memory-map:read}
42531 @item @samp{qXfer:sdata:read}
42536 @item @samp{qXfer:siginfo:read}
42541 @item @samp{qXfer:siginfo:write}
42546 @item @samp{qXfer:threads:read}
42551 @item @samp{qXfer:traceframe-info:read}
42556 @item @samp{qXfer:uib:read}
42561 @item @samp{qXfer:fdpic:read}
42566 @item @samp{Qbtrace:off}
42571 @item @samp{Qbtrace:bts}
42576 @item @samp{Qbtrace:pt}
42581 @item @samp{Qbtrace-conf:bts:size}
42586 @item @samp{Qbtrace-conf:pt:size}
42591 @item @samp{QNonStop}
42596 @item @samp{QCatchSyscalls}
42601 @item @samp{QPassSignals}
42606 @item @samp{QStartNoAckMode}
42611 @item @samp{multiprocess}
42616 @item @samp{ConditionalBreakpoints}
42621 @item @samp{ConditionalTracepoints}
42626 @item @samp{ReverseContinue}
42631 @item @samp{ReverseStep}
42636 @item @samp{TracepointSource}
42641 @item @samp{QAgent}
42646 @item @samp{QAllow}
42651 @item @samp{QDisableRandomization}
42656 @item @samp{EnableDisableTracepoints}
42661 @item @samp{QTBuffer:size}
42666 @item @samp{tracenz}
42671 @item @samp{BreakpointCommands}
42676 @item @samp{swbreak}
42681 @item @samp{hwbreak}
42686 @item @samp{fork-events}
42691 @item @samp{vfork-events}
42696 @item @samp{exec-events}
42701 @item @samp{QThreadEvents}
42706 @item @samp{no-resumed}
42713 These are the currently defined stub features, in more detail:
42716 @cindex packet size, remote protocol
42717 @item PacketSize=@var{bytes}
42718 The remote stub can accept packets up to at least @var{bytes} in
42719 length. @value{GDBN} will send packets up to this size for bulk
42720 transfers, and will never send larger packets. This is a limit on the
42721 data characters in the packet, including the frame and checksum.
42722 There is no trailing NUL byte in a remote protocol packet; if the stub
42723 stores packets in a NUL-terminated format, it should allow an extra
42724 byte in its buffer for the NUL. If this stub feature is not supported,
42725 @value{GDBN} guesses based on the size of the @samp{g} packet response.
42727 @item qXfer:auxv:read
42728 The remote stub understands the @samp{qXfer:auxv:read} packet
42729 (@pxref{qXfer auxiliary vector read}).
42731 @item qXfer:btrace:read
42732 The remote stub understands the @samp{qXfer:btrace:read}
42733 packet (@pxref{qXfer btrace read}).
42735 @item qXfer:btrace-conf:read
42736 The remote stub understands the @samp{qXfer:btrace-conf:read}
42737 packet (@pxref{qXfer btrace-conf read}).
42739 @item qXfer:exec-file:read
42740 The remote stub understands the @samp{qXfer:exec-file:read} packet
42741 (@pxref{qXfer executable filename read}).
42743 @item qXfer:features:read
42744 The remote stub understands the @samp{qXfer:features:read} packet
42745 (@pxref{qXfer target description read}).
42747 @item qXfer:libraries:read
42748 The remote stub understands the @samp{qXfer:libraries:read} packet
42749 (@pxref{qXfer library list read}).
42751 @item qXfer:libraries-svr4:read
42752 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
42753 (@pxref{qXfer svr4 library list read}).
42755 @item augmented-libraries-svr4-read
42756 The remote stub understands the augmented form of the
42757 @samp{qXfer:libraries-svr4:read} packet
42758 (@pxref{qXfer svr4 library list read}).
42760 @item qXfer:memory-map:read
42761 The remote stub understands the @samp{qXfer:memory-map:read} packet
42762 (@pxref{qXfer memory map read}).
42764 @item qXfer:sdata:read
42765 The remote stub understands the @samp{qXfer:sdata:read} packet
42766 (@pxref{qXfer sdata read}).
42768 @item qXfer:siginfo:read
42769 The remote stub understands the @samp{qXfer:siginfo:read} packet
42770 (@pxref{qXfer siginfo read}).
42772 @item qXfer:siginfo:write
42773 The remote stub understands the @samp{qXfer:siginfo:write} packet
42774 (@pxref{qXfer siginfo write}).
42776 @item qXfer:threads:read
42777 The remote stub understands the @samp{qXfer:threads:read} packet
42778 (@pxref{qXfer threads read}).
42780 @item qXfer:traceframe-info:read
42781 The remote stub understands the @samp{qXfer:traceframe-info:read}
42782 packet (@pxref{qXfer traceframe info read}).
42784 @item qXfer:uib:read
42785 The remote stub understands the @samp{qXfer:uib:read}
42786 packet (@pxref{qXfer unwind info block}).
42788 @item qXfer:fdpic:read
42789 The remote stub understands the @samp{qXfer:fdpic:read}
42790 packet (@pxref{qXfer fdpic loadmap read}).
42793 The remote stub understands the @samp{QNonStop} packet
42794 (@pxref{QNonStop}).
42796 @item QCatchSyscalls
42797 The remote stub understands the @samp{QCatchSyscalls} packet
42798 (@pxref{QCatchSyscalls}).
42801 The remote stub understands the @samp{QPassSignals} packet
42802 (@pxref{QPassSignals}).
42804 @item QStartNoAckMode
42805 The remote stub understands the @samp{QStartNoAckMode} packet and
42806 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
42809 @anchor{multiprocess extensions}
42810 @cindex multiprocess extensions, in remote protocol
42811 The remote stub understands the multiprocess extensions to the remote
42812 protocol syntax. The multiprocess extensions affect the syntax of
42813 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
42814 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
42815 replies. Note that reporting this feature indicates support for the
42816 syntactic extensions only, not that the stub necessarily supports
42817 debugging of more than one process at a time. The stub must not use
42818 multiprocess extensions in packet replies unless @value{GDBN} has also
42819 indicated it supports them in its @samp{qSupported} request.
42821 @item qXfer:osdata:read
42822 The remote stub understands the @samp{qXfer:osdata:read} packet
42823 ((@pxref{qXfer osdata read}).
42825 @item ConditionalBreakpoints
42826 The target accepts and implements evaluation of conditional expressions
42827 defined for breakpoints. The target will only report breakpoint triggers
42828 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
42830 @item ConditionalTracepoints
42831 The remote stub accepts and implements conditional expressions defined
42832 for tracepoints (@pxref{Tracepoint Conditions}).
42834 @item ReverseContinue
42835 The remote stub accepts and implements the reverse continue packet
42839 The remote stub accepts and implements the reverse step packet
42842 @item TracepointSource
42843 The remote stub understands the @samp{QTDPsrc} packet that supplies
42844 the source form of tracepoint definitions.
42847 The remote stub understands the @samp{QAgent} packet.
42850 The remote stub understands the @samp{QAllow} packet.
42852 @item QDisableRandomization
42853 The remote stub understands the @samp{QDisableRandomization} packet.
42855 @item StaticTracepoint
42856 @cindex static tracepoints, in remote protocol
42857 The remote stub supports static tracepoints.
42859 @item InstallInTrace
42860 @anchor{install tracepoint in tracing}
42861 The remote stub supports installing tracepoint in tracing.
42863 @item EnableDisableTracepoints
42864 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
42865 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
42866 to be enabled and disabled while a trace experiment is running.
42868 @item QTBuffer:size
42869 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
42870 packet that allows to change the size of the trace buffer.
42873 @cindex string tracing, in remote protocol
42874 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
42875 See @ref{Bytecode Descriptions} for details about the bytecode.
42877 @item BreakpointCommands
42878 @cindex breakpoint commands, in remote protocol
42879 The remote stub supports running a breakpoint's command list itself,
42880 rather than reporting the hit to @value{GDBN}.
42883 The remote stub understands the @samp{Qbtrace:off} packet.
42886 The remote stub understands the @samp{Qbtrace:bts} packet.
42889 The remote stub understands the @samp{Qbtrace:pt} packet.
42891 @item Qbtrace-conf:bts:size
42892 The remote stub understands the @samp{Qbtrace-conf:bts:size} packet.
42894 @item Qbtrace-conf:pt:size
42895 The remote stub understands the @samp{Qbtrace-conf:pt:size} packet.
42898 The remote stub reports the @samp{swbreak} stop reason for memory
42902 The remote stub reports the @samp{hwbreak} stop reason for hardware
42906 The remote stub reports the @samp{fork} stop reason for fork events.
42909 The remote stub reports the @samp{vfork} stop reason for vfork events
42910 and vforkdone events.
42913 The remote stub reports the @samp{exec} stop reason for exec events.
42915 @item vContSupported
42916 The remote stub reports the supported actions in the reply to
42917 @samp{vCont?} packet.
42919 @item QThreadEvents
42920 The remote stub understands the @samp{QThreadEvents} packet.
42923 The remote stub reports the @samp{N} stop reply.
42928 @cindex symbol lookup, remote request
42929 @cindex @samp{qSymbol} packet
42930 Notify the target that @value{GDBN} is prepared to serve symbol lookup
42931 requests. Accept requests from the target for the values of symbols.
42936 The target does not need to look up any (more) symbols.
42937 @item qSymbol:@var{sym_name}
42938 The target requests the value of symbol @var{sym_name} (hex encoded).
42939 @value{GDBN} may provide the value by using the
42940 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
42944 @item qSymbol:@var{sym_value}:@var{sym_name}
42945 Set the value of @var{sym_name} to @var{sym_value}.
42947 @var{sym_name} (hex encoded) is the name of a symbol whose value the
42948 target has previously requested.
42950 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
42951 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
42957 The target does not need to look up any (more) symbols.
42958 @item qSymbol:@var{sym_name}
42959 The target requests the value of a new symbol @var{sym_name} (hex
42960 encoded). @value{GDBN} will continue to supply the values of symbols
42961 (if available), until the target ceases to request them.
42966 @itemx QTDisconnected
42973 @itemx qTMinFTPILen
42975 @xref{Tracepoint Packets}.
42977 @item qThreadExtraInfo,@var{thread-id}
42978 @cindex thread attributes info, remote request
42979 @cindex @samp{qThreadExtraInfo} packet
42980 Obtain from the target OS a printable string description of thread
42981 attributes for the thread @var{thread-id}; see @ref{thread-id syntax},
42982 for the forms of @var{thread-id}. This
42983 string may contain anything that the target OS thinks is interesting
42984 for @value{GDBN} to tell the user about the thread. The string is
42985 displayed in @value{GDBN}'s @code{info threads} display. Some
42986 examples of possible thread extra info strings are @samp{Runnable}, or
42987 @samp{Blocked on Mutex}.
42991 @item @var{XX}@dots{}
42992 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
42993 comprising the printable string containing the extra information about
42994 the thread's attributes.
42997 (Note that the @code{qThreadExtraInfo} packet's name is separated from
42998 the command by a @samp{,}, not a @samp{:}, contrary to the naming
42999 conventions above. Please don't use this packet as a model for new
43018 @xref{Tracepoint Packets}.
43020 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
43021 @cindex read special object, remote request
43022 @cindex @samp{qXfer} packet
43023 @anchor{qXfer read}
43024 Read uninterpreted bytes from the target's special data area
43025 identified by the keyword @var{object}. Request @var{length} bytes
43026 starting at @var{offset} bytes into the data. The content and
43027 encoding of @var{annex} is specific to @var{object}; it can supply
43028 additional details about what data to access.
43033 Data @var{data} (@pxref{Binary Data}) has been read from the
43034 target. There may be more data at a higher address (although
43035 it is permitted to return @samp{m} even for the last valid
43036 block of data, as long as at least one byte of data was read).
43037 It is possible for @var{data} to have fewer bytes than the @var{length} in the
43041 Data @var{data} (@pxref{Binary Data}) has been read from the target.
43042 There is no more data to be read. It is possible for @var{data} to
43043 have fewer bytes than the @var{length} in the request.
43046 The @var{offset} in the request is at the end of the data.
43047 There is no more data to be read.
43050 The request was malformed, or @var{annex} was invalid.
43053 The offset was invalid, or there was an error encountered reading the data.
43054 The @var{nn} part is a hex-encoded @code{errno} value.
43057 An empty reply indicates the @var{object} string was not recognized by
43058 the stub, or that the object does not support reading.
43061 Here are the specific requests of this form defined so far. All the
43062 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
43063 formats, listed above.
43066 @item qXfer:auxv:read::@var{offset},@var{length}
43067 @anchor{qXfer auxiliary vector read}
43068 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
43069 auxiliary vector}. Note @var{annex} must be empty.
43071 This packet is not probed by default; the remote stub must request it,
43072 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43074 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
43075 @anchor{qXfer btrace read}
43077 Return a description of the current branch trace.
43078 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
43079 packet may have one of the following values:
43083 Returns all available branch trace.
43086 Returns all available branch trace if the branch trace changed since
43087 the last read request.
43090 Returns the new branch trace since the last read request. Adds a new
43091 block to the end of the trace that begins at zero and ends at the source
43092 location of the first branch in the trace buffer. This extra block is
43093 used to stitch traces together.
43095 If the trace buffer overflowed, returns an error indicating the overflow.
43098 This packet is not probed by default; the remote stub must request it
43099 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43101 @item qXfer:btrace-conf:read::@var{offset},@var{length}
43102 @anchor{qXfer btrace-conf read}
43104 Return a description of the current branch trace configuration.
43105 @xref{Branch Trace Configuration Format}.
43107 This packet is not probed by default; the remote stub must request it
43108 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43110 @item qXfer:exec-file:read:@var{annex}:@var{offset},@var{length}
43111 @anchor{qXfer executable filename read}
43112 Return the full absolute name of the file that was executed to create
43113 a process running on the remote system. The annex specifies the
43114 numeric process ID of the process to query, encoded as a hexadecimal
43115 number. If the annex part is empty the remote stub should return the
43116 filename corresponding to the currently executing process.
43118 This packet is not probed by default; the remote stub must request it,
43119 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43121 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
43122 @anchor{qXfer target description read}
43123 Access the @dfn{target description}. @xref{Target Descriptions}. The
43124 annex specifies which XML document to access. The main description is
43125 always loaded from the @samp{target.xml} annex.
43127 This packet is not probed by default; the remote stub must request it,
43128 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43130 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
43131 @anchor{qXfer library list read}
43132 Access the target's list of loaded libraries. @xref{Library List Format}.
43133 The annex part of the generic @samp{qXfer} packet must be empty
43134 (@pxref{qXfer read}).
43136 Targets which maintain a list of libraries in the program's memory do
43137 not need to implement this packet; it is designed for platforms where
43138 the operating system manages the list of loaded libraries.
43140 This packet is not probed by default; the remote stub must request it,
43141 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43143 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
43144 @anchor{qXfer svr4 library list read}
43145 Access the target's list of loaded libraries when the target is an SVR4
43146 platform. @xref{Library List Format for SVR4 Targets}. The annex part
43147 of the generic @samp{qXfer} packet must be empty unless the remote
43148 stub indicated it supports the augmented form of this packet
43149 by supplying an appropriate @samp{qSupported} response
43150 (@pxref{qXfer read}, @ref{qSupported}).
43152 This packet is optional for better performance on SVR4 targets.
43153 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
43155 This packet is not probed by default; the remote stub must request it,
43156 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43158 If the remote stub indicates it supports the augmented form of this
43159 packet then the annex part of the generic @samp{qXfer} packet may
43160 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
43161 arguments. The currently supported arguments are:
43164 @item start=@var{address}
43165 A hexadecimal number specifying the address of the @samp{struct
43166 link_map} to start reading the library list from. If unset or zero
43167 then the first @samp{struct link_map} in the library list will be
43168 chosen as the starting point.
43170 @item prev=@var{address}
43171 A hexadecimal number specifying the address of the @samp{struct
43172 link_map} immediately preceding the @samp{struct link_map}
43173 specified by the @samp{start} argument. If unset or zero then
43174 the remote stub will expect that no @samp{struct link_map}
43175 exists prior to the starting point.
43179 Arguments that are not understood by the remote stub will be silently
43182 @item qXfer:memory-map:read::@var{offset},@var{length}
43183 @anchor{qXfer memory map read}
43184 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
43185 annex part of the generic @samp{qXfer} packet must be empty
43186 (@pxref{qXfer read}).
43188 This packet is not probed by default; the remote stub must request it,
43189 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43191 @item qXfer:sdata:read::@var{offset},@var{length}
43192 @anchor{qXfer sdata read}
43194 Read contents of the extra collected static tracepoint marker
43195 information. The annex part of the generic @samp{qXfer} packet must
43196 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
43199 This packet is not probed by default; the remote stub must request it,
43200 by supplying an appropriate @samp{qSupported} response
43201 (@pxref{qSupported}).
43203 @item qXfer:siginfo:read::@var{offset},@var{length}
43204 @anchor{qXfer siginfo read}
43205 Read contents of the extra signal information on the target
43206 system. The annex part of the generic @samp{qXfer} packet must be
43207 empty (@pxref{qXfer read}).
43209 This packet is not probed by default; the remote stub must request it,
43210 by supplying an appropriate @samp{qSupported} response
43211 (@pxref{qSupported}).
43213 @item qXfer:threads:read::@var{offset},@var{length}
43214 @anchor{qXfer threads read}
43215 Access the list of threads on target. @xref{Thread List Format}. The
43216 annex part of the generic @samp{qXfer} packet must be empty
43217 (@pxref{qXfer read}).
43219 This packet is not probed by default; the remote stub must request it,
43220 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43222 @item qXfer:traceframe-info:read::@var{offset},@var{length}
43223 @anchor{qXfer traceframe info read}
43225 Return a description of the current traceframe's contents.
43226 @xref{Traceframe Info Format}. The annex part of the generic
43227 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
43229 This packet is not probed by default; the remote stub must request it,
43230 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43232 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
43233 @anchor{qXfer unwind info block}
43235 Return the unwind information block for @var{pc}. This packet is used
43236 on OpenVMS/ia64 to ask the kernel unwind information.
43238 This packet is not probed by default.
43240 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
43241 @anchor{qXfer fdpic loadmap read}
43242 Read contents of @code{loadmap}s on the target system. The
43243 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
43244 executable @code{loadmap} or interpreter @code{loadmap} to read.
43246 This packet is not probed by default; the remote stub must request it,
43247 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
43249 @item qXfer:osdata:read::@var{offset},@var{length}
43250 @anchor{qXfer osdata read}
43251 Access the target's @dfn{operating system information}.
43252 @xref{Operating System Information}.
43256 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
43257 @cindex write data into object, remote request
43258 @anchor{qXfer write}
43259 Write uninterpreted bytes into the target's special data area
43260 identified by the keyword @var{object}, starting at @var{offset} bytes
43261 into the data. The binary-encoded data (@pxref{Binary Data}) to be
43262 written is given by @var{data}@dots{}. The content and encoding of @var{annex}
43263 is specific to @var{object}; it can supply additional details about what data
43269 @var{nn} (hex encoded) is the number of bytes written.
43270 This may be fewer bytes than supplied in the request.
43273 The request was malformed, or @var{annex} was invalid.
43276 The offset was invalid, or there was an error encountered writing the data.
43277 The @var{nn} part is a hex-encoded @code{errno} value.
43280 An empty reply indicates the @var{object} string was not
43281 recognized by the stub, or that the object does not support writing.
43284 Here are the specific requests of this form defined so far. All the
43285 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
43286 formats, listed above.
43289 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
43290 @anchor{qXfer siginfo write}
43291 Write @var{data} to the extra signal information on the target system.
43292 The annex part of the generic @samp{qXfer} packet must be
43293 empty (@pxref{qXfer write}).
43295 This packet is not probed by default; the remote stub must request it,
43296 by supplying an appropriate @samp{qSupported} response
43297 (@pxref{qSupported}).
43300 @item qXfer:@var{object}:@var{operation}:@dots{}
43301 Requests of this form may be added in the future. When a stub does
43302 not recognize the @var{object} keyword, or its support for
43303 @var{object} does not recognize the @var{operation} keyword, the stub
43304 must respond with an empty packet.
43306 @item qAttached:@var{pid}
43307 @cindex query attached, remote request
43308 @cindex @samp{qAttached} packet
43309 Return an indication of whether the remote server attached to an
43310 existing process or created a new process. When the multiprocess
43311 protocol extensions are supported (@pxref{multiprocess extensions}),
43312 @var{pid} is an integer in hexadecimal format identifying the target
43313 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
43314 the query packet will be simplified as @samp{qAttached}.
43316 This query is used, for example, to know whether the remote process
43317 should be detached or killed when a @value{GDBN} session is ended with
43318 the @code{quit} command.
43323 The remote server attached to an existing process.
43325 The remote server created a new process.
43327 A badly formed request or an error was encountered.
43331 Enable branch tracing for the current thread using Branch Trace Store.
43336 Branch tracing has been enabled.
43338 A badly formed request or an error was encountered.
43342 Enable branch tracing for the current thread using Intel Processor Trace.
43347 Branch tracing has been enabled.
43349 A badly formed request or an error was encountered.
43353 Disable branch tracing for the current thread.
43358 Branch tracing has been disabled.
43360 A badly formed request or an error was encountered.
43363 @item Qbtrace-conf:bts:size=@var{value}
43364 Set the requested ring buffer size for new threads that use the
43365 btrace recording method in bts format.
43370 The ring buffer size has been set.
43372 A badly formed request or an error was encountered.
43375 @item Qbtrace-conf:pt:size=@var{value}
43376 Set the requested ring buffer size for new threads that use the
43377 btrace recording method in pt format.
43382 The ring buffer size has been set.
43384 A badly formed request or an error was encountered.
43389 @node Architecture-Specific Protocol Details
43390 @section Architecture-Specific Protocol Details
43392 This section describes how the remote protocol is applied to specific
43393 target architectures. Also see @ref{Standard Target Features}, for
43394 details of XML target descriptions for each architecture.
43397 * ARM-Specific Protocol Details::
43398 * MIPS-Specific Protocol Details::
43401 @node ARM-Specific Protocol Details
43402 @subsection @acronym{ARM}-specific Protocol Details
43405 * ARM Breakpoint Kinds::
43408 @node ARM Breakpoint Kinds
43409 @subsubsection @acronym{ARM} Breakpoint Kinds
43410 @cindex breakpoint kinds, @acronym{ARM}
43412 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
43417 16-bit Thumb mode breakpoint.
43420 32-bit Thumb mode (Thumb-2) breakpoint.
43423 32-bit @acronym{ARM} mode breakpoint.
43427 @node MIPS-Specific Protocol Details
43428 @subsection @acronym{MIPS}-specific Protocol Details
43431 * MIPS Register packet Format::
43432 * MIPS Breakpoint Kinds::
43435 @node MIPS Register packet Format
43436 @subsubsection @acronym{MIPS} Register Packet Format
43437 @cindex register packet format, @acronym{MIPS}
43439 The following @code{g}/@code{G} packets have previously been defined.
43440 In the below, some thirty-two bit registers are transferred as
43441 sixty-four bits. Those registers should be zero/sign extended (which?)
43442 to fill the space allocated. Register bytes are transferred in target
43443 byte order. The two nibbles within a register byte are transferred
43444 most-significant -- least-significant.
43449 All registers are transferred as thirty-two bit quantities in the order:
43450 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
43451 registers; fsr; fir; fp.
43454 All registers are transferred as sixty-four bit quantities (including
43455 thirty-two bit registers such as @code{sr}). The ordering is the same
43460 @node MIPS Breakpoint Kinds
43461 @subsubsection @acronym{MIPS} Breakpoint Kinds
43462 @cindex breakpoint kinds, @acronym{MIPS}
43464 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
43469 16-bit @acronym{MIPS16} mode breakpoint.
43472 16-bit @acronym{microMIPS} mode breakpoint.
43475 32-bit standard @acronym{MIPS} mode breakpoint.
43478 32-bit @acronym{microMIPS} mode breakpoint.
43482 @node Tracepoint Packets
43483 @section Tracepoint Packets
43484 @cindex tracepoint packets
43485 @cindex packets, tracepoint
43487 Here we describe the packets @value{GDBN} uses to implement
43488 tracepoints (@pxref{Tracepoints}).
43492 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
43493 @cindex @samp{QTDP} packet
43494 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
43495 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
43496 the tracepoint is disabled. The @var{step} gives the tracepoint's step
43497 count, and @var{pass} gives its pass count. If an @samp{F} is present,
43498 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
43499 the number of bytes that the target should copy elsewhere to make room
43500 for the tracepoint. If an @samp{X} is present, it introduces a
43501 tracepoint condition, which consists of a hexadecimal length, followed
43502 by a comma and hex-encoded bytes, in a manner similar to action
43503 encodings as described below. If the trailing @samp{-} is present,
43504 further @samp{QTDP} packets will follow to specify this tracepoint's
43510 The packet was understood and carried out.
43512 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
43514 The packet was not recognized.
43517 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
43518 Define actions to be taken when a tracepoint is hit. The @var{n} and
43519 @var{addr} must be the same as in the initial @samp{QTDP} packet for
43520 this tracepoint. This packet may only be sent immediately after
43521 another @samp{QTDP} packet that ended with a @samp{-}. If the
43522 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
43523 specifying more actions for this tracepoint.
43525 In the series of action packets for a given tracepoint, at most one
43526 can have an @samp{S} before its first @var{action}. If such a packet
43527 is sent, it and the following packets define ``while-stepping''
43528 actions. Any prior packets define ordinary actions --- that is, those
43529 taken when the tracepoint is first hit. If no action packet has an
43530 @samp{S}, then all the packets in the series specify ordinary
43531 tracepoint actions.
43533 The @samp{@var{action}@dots{}} portion of the packet is a series of
43534 actions, concatenated without separators. Each action has one of the
43540 Collect the registers whose bits are set in @var{mask},
43541 a hexadecimal number whose @var{i}'th bit is set if register number
43542 @var{i} should be collected. (The least significant bit is numbered
43543 zero.) Note that @var{mask} may be any number of digits long; it may
43544 not fit in a 32-bit word.
43546 @item M @var{basereg},@var{offset},@var{len}
43547 Collect @var{len} bytes of memory starting at the address in register
43548 number @var{basereg}, plus @var{offset}. If @var{basereg} is
43549 @samp{-1}, then the range has a fixed address: @var{offset} is the
43550 address of the lowest byte to collect. The @var{basereg},
43551 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
43552 values (the @samp{-1} value for @var{basereg} is a special case).
43554 @item X @var{len},@var{expr}
43555 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
43556 it directs. The agent expression @var{expr} is as described in
43557 @ref{Agent Expressions}. Each byte of the expression is encoded as a
43558 two-digit hex number in the packet; @var{len} is the number of bytes
43559 in the expression (and thus one-half the number of hex digits in the
43564 Any number of actions may be packed together in a single @samp{QTDP}
43565 packet, as long as the packet does not exceed the maximum packet
43566 length (400 bytes, for many stubs). There may be only one @samp{R}
43567 action per tracepoint, and it must precede any @samp{M} or @samp{X}
43568 actions. Any registers referred to by @samp{M} and @samp{X} actions
43569 must be collected by a preceding @samp{R} action. (The
43570 ``while-stepping'' actions are treated as if they were attached to a
43571 separate tracepoint, as far as these restrictions are concerned.)
43576 The packet was understood and carried out.
43578 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
43580 The packet was not recognized.
43583 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
43584 @cindex @samp{QTDPsrc} packet
43585 Specify a source string of tracepoint @var{n} at address @var{addr}.
43586 This is useful to get accurate reproduction of the tracepoints
43587 originally downloaded at the beginning of the trace run. The @var{type}
43588 is the name of the tracepoint part, such as @samp{cond} for the
43589 tracepoint's conditional expression (see below for a list of types), while
43590 @var{bytes} is the string, encoded in hexadecimal.
43592 @var{start} is the offset of the @var{bytes} within the overall source
43593 string, while @var{slen} is the total length of the source string.
43594 This is intended for handling source strings that are longer than will
43595 fit in a single packet.
43596 @c Add detailed example when this info is moved into a dedicated
43597 @c tracepoint descriptions section.
43599 The available string types are @samp{at} for the location,
43600 @samp{cond} for the conditional, and @samp{cmd} for an action command.
43601 @value{GDBN} sends a separate packet for each command in the action
43602 list, in the same order in which the commands are stored in the list.
43604 The target does not need to do anything with source strings except
43605 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
43608 Although this packet is optional, and @value{GDBN} will only send it
43609 if the target replies with @samp{TracepointSource} @xref{General
43610 Query Packets}, it makes both disconnected tracing and trace files
43611 much easier to use. Otherwise the user must be careful that the
43612 tracepoints in effect while looking at trace frames are identical to
43613 the ones in effect during the trace run; even a small discrepancy
43614 could cause @samp{tdump} not to work, or a particular trace frame not
43617 @item QTDV:@var{n}:@var{value}:@var{builtin}:@var{name}
43618 @cindex define trace state variable, remote request
43619 @cindex @samp{QTDV} packet
43620 Create a new trace state variable, number @var{n}, with an initial
43621 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
43622 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
43623 the option of not using this packet for initial values of zero; the
43624 target should simply create the trace state variables as they are
43625 mentioned in expressions. The value @var{builtin} should be 1 (one)
43626 if the trace state variable is builtin and 0 (zero) if it is not builtin.
43627 @value{GDBN} only sets @var{builtin} to 1 if a previous @samp{qTfV} or
43628 @samp{qTsV} packet had it set. The contents of @var{name} is the
43629 hex-encoded name (without the leading @samp{$}) of the trace state
43632 @item QTFrame:@var{n}
43633 @cindex @samp{QTFrame} packet
43634 Select the @var{n}'th tracepoint frame from the buffer, and use the
43635 register and memory contents recorded there to answer subsequent
43636 request packets from @value{GDBN}.
43638 A successful reply from the stub indicates that the stub has found the
43639 requested frame. The response is a series of parts, concatenated
43640 without separators, describing the frame we selected. Each part has
43641 one of the following forms:
43645 The selected frame is number @var{n} in the trace frame buffer;
43646 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
43647 was no frame matching the criteria in the request packet.
43650 The selected trace frame records a hit of tracepoint number @var{t};
43651 @var{t} is a hexadecimal number.
43655 @item QTFrame:pc:@var{addr}
43656 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
43657 currently selected frame whose PC is @var{addr};
43658 @var{addr} is a hexadecimal number.
43660 @item QTFrame:tdp:@var{t}
43661 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
43662 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
43663 is a hexadecimal number.
43665 @item QTFrame:range:@var{start}:@var{end}
43666 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
43667 currently selected frame whose PC is between @var{start} (inclusive)
43668 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
43671 @item QTFrame:outside:@var{start}:@var{end}
43672 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
43673 frame @emph{outside} the given range of addresses (exclusive).
43676 @cindex @samp{qTMinFTPILen} packet
43677 This packet requests the minimum length of instruction at which a fast
43678 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
43679 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
43680 it depends on the target system being able to create trampolines in
43681 the first 64K of memory, which might or might not be possible for that
43682 system. So the reply to this packet will be 4 if it is able to
43689 The minimum instruction length is currently unknown.
43691 The minimum instruction length is @var{length}, where @var{length}
43692 is a hexadecimal number greater or equal to 1. A reply
43693 of 1 means that a fast tracepoint may be placed on any instruction
43694 regardless of size.
43696 An error has occurred.
43698 An empty reply indicates that the request is not supported by the stub.
43702 @cindex @samp{QTStart} packet
43703 Begin the tracepoint experiment. Begin collecting data from
43704 tracepoint hits in the trace frame buffer. This packet supports the
43705 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
43706 instruction reply packet}).
43709 @cindex @samp{QTStop} packet
43710 End the tracepoint experiment. Stop collecting trace frames.
43712 @item QTEnable:@var{n}:@var{addr}
43714 @cindex @samp{QTEnable} packet
43715 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
43716 experiment. If the tracepoint was previously disabled, then collection
43717 of data from it will resume.
43719 @item QTDisable:@var{n}:@var{addr}
43721 @cindex @samp{QTDisable} packet
43722 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
43723 experiment. No more data will be collected from the tracepoint unless
43724 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
43727 @cindex @samp{QTinit} packet
43728 Clear the table of tracepoints, and empty the trace frame buffer.
43730 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
43731 @cindex @samp{QTro} packet
43732 Establish the given ranges of memory as ``transparent''. The stub
43733 will answer requests for these ranges from memory's current contents,
43734 if they were not collected as part of the tracepoint hit.
43736 @value{GDBN} uses this to mark read-only regions of memory, like those
43737 containing program code. Since these areas never change, they should
43738 still have the same contents they did when the tracepoint was hit, so
43739 there's no reason for the stub to refuse to provide their contents.
43741 @item QTDisconnected:@var{value}
43742 @cindex @samp{QTDisconnected} packet
43743 Set the choice to what to do with the tracing run when @value{GDBN}
43744 disconnects from the target. A @var{value} of 1 directs the target to
43745 continue the tracing run, while 0 tells the target to stop tracing if
43746 @value{GDBN} is no longer in the picture.
43749 @cindex @samp{qTStatus} packet
43750 Ask the stub if there is a trace experiment running right now.
43752 The reply has the form:
43756 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
43757 @var{running} is a single digit @code{1} if the trace is presently
43758 running, or @code{0} if not. It is followed by semicolon-separated
43759 optional fields that an agent may use to report additional status.
43763 If the trace is not running, the agent may report any of several
43764 explanations as one of the optional fields:
43769 No trace has been run yet.
43771 @item tstop[:@var{text}]:0
43772 The trace was stopped by a user-originated stop command. The optional
43773 @var{text} field is a user-supplied string supplied as part of the
43774 stop command (for instance, an explanation of why the trace was
43775 stopped manually). It is hex-encoded.
43778 The trace stopped because the trace buffer filled up.
43780 @item tdisconnected:0
43781 The trace stopped because @value{GDBN} disconnected from the target.
43783 @item tpasscount:@var{tpnum}
43784 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
43786 @item terror:@var{text}:@var{tpnum}
43787 The trace stopped because tracepoint @var{tpnum} had an error. The
43788 string @var{text} is available to describe the nature of the error
43789 (for instance, a divide by zero in the condition expression); it
43793 The trace stopped for some other reason.
43797 Additional optional fields supply statistical and other information.
43798 Although not required, they are extremely useful for users monitoring
43799 the progress of a trace run. If a trace has stopped, and these
43800 numbers are reported, they must reflect the state of the just-stopped
43805 @item tframes:@var{n}
43806 The number of trace frames in the buffer.
43808 @item tcreated:@var{n}
43809 The total number of trace frames created during the run. This may
43810 be larger than the trace frame count, if the buffer is circular.
43812 @item tsize:@var{n}
43813 The total size of the trace buffer, in bytes.
43815 @item tfree:@var{n}
43816 The number of bytes still unused in the buffer.
43818 @item circular:@var{n}
43819 The value of the circular trace buffer flag. @code{1} means that the
43820 trace buffer is circular and old trace frames will be discarded if
43821 necessary to make room, @code{0} means that the trace buffer is linear
43824 @item disconn:@var{n}
43825 The value of the disconnected tracing flag. @code{1} means that
43826 tracing will continue after @value{GDBN} disconnects, @code{0} means
43827 that the trace run will stop.
43831 @item qTP:@var{tp}:@var{addr}
43832 @cindex tracepoint status, remote request
43833 @cindex @samp{qTP} packet
43834 Ask the stub for the current state of tracepoint number @var{tp} at
43835 address @var{addr}.
43839 @item V@var{hits}:@var{usage}
43840 The tracepoint has been hit @var{hits} times so far during the trace
43841 run, and accounts for @var{usage} in the trace buffer. Note that
43842 @code{while-stepping} steps are not counted as separate hits, but the
43843 steps' space consumption is added into the usage number.
43847 @item qTV:@var{var}
43848 @cindex trace state variable value, remote request
43849 @cindex @samp{qTV} packet
43850 Ask the stub for the value of the trace state variable number @var{var}.
43855 The value of the variable is @var{value}. This will be the current
43856 value of the variable if the user is examining a running target, or a
43857 saved value if the variable was collected in the trace frame that the
43858 user is looking at. Note that multiple requests may result in
43859 different reply values, such as when requesting values while the
43860 program is running.
43863 The value of the variable is unknown. This would occur, for example,
43864 if the user is examining a trace frame in which the requested variable
43869 @cindex @samp{qTfP} packet
43871 @cindex @samp{qTsP} packet
43872 These packets request data about tracepoints that are being used by
43873 the target. @value{GDBN} sends @code{qTfP} to get the first piece
43874 of data, and multiple @code{qTsP} to get additional pieces. Replies
43875 to these packets generally take the form of the @code{QTDP} packets
43876 that define tracepoints. (FIXME add detailed syntax)
43879 @cindex @samp{qTfV} packet
43881 @cindex @samp{qTsV} packet
43882 These packets request data about trace state variables that are on the
43883 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
43884 and multiple @code{qTsV} to get additional variables. Replies to
43885 these packets follow the syntax of the @code{QTDV} packets that define
43886 trace state variables.
43892 @cindex @samp{qTfSTM} packet
43893 @cindex @samp{qTsSTM} packet
43894 These packets request data about static tracepoint markers that exist
43895 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
43896 first piece of data, and multiple @code{qTsSTM} to get additional
43897 pieces. Replies to these packets take the following form:
43901 @item m @var{address}:@var{id}:@var{extra}
43903 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
43904 a comma-separated list of markers
43906 (lower case letter @samp{L}) denotes end of list.
43908 An error occurred. The error number @var{nn} is given as hex digits.
43910 An empty reply indicates that the request is not supported by the
43914 The @var{address} is encoded in hex;
43915 @var{id} and @var{extra} are strings encoded in hex.
43917 In response to each query, the target will reply with a list of one or
43918 more markers, separated by commas. @value{GDBN} will respond to each
43919 reply with a request for more markers (using the @samp{qs} form of the
43920 query), until the target responds with @samp{l} (lower-case ell, for
43923 @item qTSTMat:@var{address}
43925 @cindex @samp{qTSTMat} packet
43926 This packets requests data about static tracepoint markers in the
43927 target program at @var{address}. Replies to this packet follow the
43928 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
43929 tracepoint markers.
43931 @item QTSave:@var{filename}
43932 @cindex @samp{QTSave} packet
43933 This packet directs the target to save trace data to the file name
43934 @var{filename} in the target's filesystem. The @var{filename} is encoded
43935 as a hex string; the interpretation of the file name (relative vs
43936 absolute, wild cards, etc) is up to the target.
43938 @item qTBuffer:@var{offset},@var{len}
43939 @cindex @samp{qTBuffer} packet
43940 Return up to @var{len} bytes of the current contents of trace buffer,
43941 starting at @var{offset}. The trace buffer is treated as if it were
43942 a contiguous collection of traceframes, as per the trace file format.
43943 The reply consists as many hex-encoded bytes as the target can deliver
43944 in a packet; it is not an error to return fewer than were asked for.
43945 A reply consisting of just @code{l} indicates that no bytes are
43948 @item QTBuffer:circular:@var{value}
43949 This packet directs the target to use a circular trace buffer if
43950 @var{value} is 1, or a linear buffer if the value is 0.
43952 @item QTBuffer:size:@var{size}
43953 @anchor{QTBuffer-size}
43954 @cindex @samp{QTBuffer size} packet
43955 This packet directs the target to make the trace buffer be of size
43956 @var{size} if possible. A value of @code{-1} tells the target to
43957 use whatever size it prefers.
43959 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
43960 @cindex @samp{QTNotes} packet
43961 This packet adds optional textual notes to the trace run. Allowable
43962 types include @code{user}, @code{notes}, and @code{tstop}, the
43963 @var{text} fields are arbitrary strings, hex-encoded.
43967 @subsection Relocate instruction reply packet
43968 When installing fast tracepoints in memory, the target may need to
43969 relocate the instruction currently at the tracepoint address to a
43970 different address in memory. For most instructions, a simple copy is
43971 enough, but, for example, call instructions that implicitly push the
43972 return address on the stack, and relative branches or other
43973 PC-relative instructions require offset adjustment, so that the effect
43974 of executing the instruction at a different address is the same as if
43975 it had executed in the original location.
43977 In response to several of the tracepoint packets, the target may also
43978 respond with a number of intermediate @samp{qRelocInsn} request
43979 packets before the final result packet, to have @value{GDBN} handle
43980 this relocation operation. If a packet supports this mechanism, its
43981 documentation will explicitly say so. See for example the above
43982 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
43983 format of the request is:
43986 @item qRelocInsn:@var{from};@var{to}
43988 This requests @value{GDBN} to copy instruction at address @var{from}
43989 to address @var{to}, possibly adjusted so that executing the
43990 instruction at @var{to} has the same effect as executing it at
43991 @var{from}. @value{GDBN} writes the adjusted instruction to target
43992 memory starting at @var{to}.
43997 @item qRelocInsn:@var{adjusted_size}
43998 Informs the stub the relocation is complete. The @var{adjusted_size} is
43999 the length in bytes of resulting relocated instruction sequence.
44001 A badly formed request was detected, or an error was encountered while
44002 relocating the instruction.
44005 @node Host I/O Packets
44006 @section Host I/O Packets
44007 @cindex Host I/O, remote protocol
44008 @cindex file transfer, remote protocol
44010 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
44011 operations on the far side of a remote link. For example, Host I/O is
44012 used to upload and download files to a remote target with its own
44013 filesystem. Host I/O uses the same constant values and data structure
44014 layout as the target-initiated File-I/O protocol. However, the
44015 Host I/O packets are structured differently. The target-initiated
44016 protocol relies on target memory to store parameters and buffers.
44017 Host I/O requests are initiated by @value{GDBN}, and the
44018 target's memory is not involved. @xref{File-I/O Remote Protocol
44019 Extension}, for more details on the target-initiated protocol.
44021 The Host I/O request packets all encode a single operation along with
44022 its arguments. They have this format:
44026 @item vFile:@var{operation}: @var{parameter}@dots{}
44027 @var{operation} is the name of the particular request; the target
44028 should compare the entire packet name up to the second colon when checking
44029 for a supported operation. The format of @var{parameter} depends on
44030 the operation. Numbers are always passed in hexadecimal. Negative
44031 numbers have an explicit minus sign (i.e.@: two's complement is not
44032 used). Strings (e.g.@: filenames) are encoded as a series of
44033 hexadecimal bytes. The last argument to a system call may be a
44034 buffer of escaped binary data (@pxref{Binary Data}).
44038 The valid responses to Host I/O packets are:
44042 @item F @var{result} [, @var{errno}] [; @var{attachment}]
44043 @var{result} is the integer value returned by this operation, usually
44044 non-negative for success and -1 for errors. If an error has occured,
44045 @var{errno} will be included in the result specifying a
44046 value defined by the File-I/O protocol (@pxref{Errno Values}). For
44047 operations which return data, @var{attachment} supplies the data as a
44048 binary buffer. Binary buffers in response packets are escaped in the
44049 normal way (@pxref{Binary Data}). See the individual packet
44050 documentation for the interpretation of @var{result} and
44054 An empty response indicates that this operation is not recognized.
44058 These are the supported Host I/O operations:
44061 @item vFile:open: @var{filename}, @var{flags}, @var{mode}
44062 Open a file at @var{filename} and return a file descriptor for it, or
44063 return -1 if an error occurs. The @var{filename} is a string,
44064 @var{flags} is an integer indicating a mask of open flags
44065 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
44066 of mode bits to use if the file is created (@pxref{mode_t Values}).
44067 @xref{open}, for details of the open flags and mode values.
44069 @item vFile:close: @var{fd}
44070 Close the open file corresponding to @var{fd} and return 0, or
44071 -1 if an error occurs.
44073 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
44074 Read data from the open file corresponding to @var{fd}. Up to
44075 @var{count} bytes will be read from the file, starting at @var{offset}
44076 relative to the start of the file. The target may read fewer bytes;
44077 common reasons include packet size limits and an end-of-file
44078 condition. The number of bytes read is returned. Zero should only be
44079 returned for a successful read at the end of the file, or if
44080 @var{count} was zero.
44082 The data read should be returned as a binary attachment on success.
44083 If zero bytes were read, the response should include an empty binary
44084 attachment (i.e.@: a trailing semicolon). The return value is the
44085 number of target bytes read; the binary attachment may be longer if
44086 some characters were escaped.
44088 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
44089 Write @var{data} (a binary buffer) to the open file corresponding
44090 to @var{fd}. Start the write at @var{offset} from the start of the
44091 file. Unlike many @code{write} system calls, there is no
44092 separate @var{count} argument; the length of @var{data} in the
44093 packet is used. @samp{vFile:pwrite} returns the number of bytes written,
44094 which may be shorter than the length of @var{data}, or -1 if an
44097 @item vFile:fstat: @var{fd}
44098 Get information about the open file corresponding to @var{fd}.
44099 On success the information is returned as a binary attachment
44100 and the return value is the size of this attachment in bytes.
44101 If an error occurs the return value is -1. The format of the
44102 returned binary attachment is as described in @ref{struct stat}.
44104 @item vFile:unlink: @var{filename}
44105 Delete the file at @var{filename} on the target. Return 0,
44106 or -1 if an error occurs. The @var{filename} is a string.
44108 @item vFile:readlink: @var{filename}
44109 Read value of symbolic link @var{filename} on the target. Return
44110 the number of bytes read, or -1 if an error occurs.
44112 The data read should be returned as a binary attachment on success.
44113 If zero bytes were read, the response should include an empty binary
44114 attachment (i.e.@: a trailing semicolon). The return value is the
44115 number of target bytes read; the binary attachment may be longer if
44116 some characters were escaped.
44118 @item vFile:setfs: @var{pid}
44119 Select the filesystem on which @code{vFile} operations with
44120 @var{filename} arguments will operate. This is required for
44121 @value{GDBN} to be able to access files on remote targets where
44122 the remote stub does not share a common filesystem with the
44125 If @var{pid} is nonzero, select the filesystem as seen by process
44126 @var{pid}. If @var{pid} is zero, select the filesystem as seen by
44127 the remote stub. Return 0 on success, or -1 if an error occurs.
44128 If @code{vFile:setfs:} indicates success, the selected filesystem
44129 remains selected until the next successful @code{vFile:setfs:}
44135 @section Interrupts
44136 @cindex interrupts (remote protocol)
44137 @anchor{interrupting remote targets}
44139 In all-stop mode, when a program on the remote target is running,
44140 @value{GDBN} may attempt to interrupt it by sending a @samp{Ctrl-C},
44141 @code{BREAK} or a @code{BREAK} followed by @code{g}, control of which
44142 is specified via @value{GDBN}'s @samp{interrupt-sequence}.
44144 The precise meaning of @code{BREAK} is defined by the transport
44145 mechanism and may, in fact, be undefined. @value{GDBN} does not
44146 currently define a @code{BREAK} mechanism for any of the network
44147 interfaces except for TCP, in which case @value{GDBN} sends the
44148 @code{telnet} BREAK sequence.
44150 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
44151 transport mechanisms. It is represented by sending the single byte
44152 @code{0x03} without any of the usual packet overhead described in
44153 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
44154 transmitted as part of a packet, it is considered to be packet data
44155 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
44156 (@pxref{X packet}), used for binary downloads, may include an unescaped
44157 @code{0x03} as part of its packet.
44159 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
44160 When Linux kernel receives this sequence from serial port,
44161 it stops execution and connects to gdb.
44163 In non-stop mode, because packet resumptions are asynchronous
44164 (@pxref{vCont packet}), @value{GDBN} is always free to send a remote
44165 command to the remote stub, even when the target is running. For that
44166 reason, @value{GDBN} instead sends a regular packet (@pxref{vCtrlC
44167 packet}) with the usual packet framing instead of the single byte
44170 Stubs are not required to recognize these interrupt mechanisms and the
44171 precise meaning associated with receipt of the interrupt is
44172 implementation defined. If the target supports debugging of multiple
44173 threads and/or processes, it should attempt to interrupt all
44174 currently-executing threads and processes.
44175 If the stub is successful at interrupting the
44176 running program, it should send one of the stop
44177 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
44178 of successfully stopping the program in all-stop mode, and a stop reply
44179 for each stopped thread in non-stop mode.
44180 Interrupts received while the
44181 program is stopped are queued and the program will be interrupted when
44182 it is resumed next time.
44184 @node Notification Packets
44185 @section Notification Packets
44186 @cindex notification packets
44187 @cindex packets, notification
44189 The @value{GDBN} remote serial protocol includes @dfn{notifications},
44190 packets that require no acknowledgment. Both the GDB and the stub
44191 may send notifications (although the only notifications defined at
44192 present are sent by the stub). Notifications carry information
44193 without incurring the round-trip latency of an acknowledgment, and so
44194 are useful for low-impact communications where occasional packet loss
44197 A notification packet has the form @samp{% @var{data} #
44198 @var{checksum}}, where @var{data} is the content of the notification,
44199 and @var{checksum} is a checksum of @var{data}, computed and formatted
44200 as for ordinary @value{GDBN} packets. A notification's @var{data}
44201 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
44202 receiving a notification, the recipient sends no @samp{+} or @samp{-}
44203 to acknowledge the notification's receipt or to report its corruption.
44205 Every notification's @var{data} begins with a name, which contains no
44206 colon characters, followed by a colon character.
44208 Recipients should silently ignore corrupted notifications and
44209 notifications they do not understand. Recipients should restart
44210 timeout periods on receipt of a well-formed notification, whether or
44211 not they understand it.
44213 Senders should only send the notifications described here when this
44214 protocol description specifies that they are permitted. In the
44215 future, we may extend the protocol to permit existing notifications in
44216 new contexts; this rule helps older senders avoid confusing newer
44219 (Older versions of @value{GDBN} ignore bytes received until they see
44220 the @samp{$} byte that begins an ordinary packet, so new stubs may
44221 transmit notifications without fear of confusing older clients. There
44222 are no notifications defined for @value{GDBN} to send at the moment, but we
44223 assume that most older stubs would ignore them, as well.)
44225 Each notification is comprised of three parts:
44227 @item @var{name}:@var{event}
44228 The notification packet is sent by the side that initiates the
44229 exchange (currently, only the stub does that), with @var{event}
44230 carrying the specific information about the notification, and
44231 @var{name} specifying the name of the notification.
44233 The acknowledge sent by the other side, usually @value{GDBN}, to
44234 acknowledge the exchange and request the event.
44237 The purpose of an asynchronous notification mechanism is to report to
44238 @value{GDBN} that something interesting happened in the remote stub.
44240 The remote stub may send notification @var{name}:@var{event}
44241 at any time, but @value{GDBN} acknowledges the notification when
44242 appropriate. The notification event is pending before @value{GDBN}
44243 acknowledges. Only one notification at a time may be pending; if
44244 additional events occur before @value{GDBN} has acknowledged the
44245 previous notification, they must be queued by the stub for later
44246 synchronous transmission in response to @var{ack} packets from
44247 @value{GDBN}. Because the notification mechanism is unreliable,
44248 the stub is permitted to resend a notification if it believes
44249 @value{GDBN} may not have received it.
44251 Specifically, notifications may appear when @value{GDBN} is not
44252 otherwise reading input from the stub, or when @value{GDBN} is
44253 expecting to read a normal synchronous response or a
44254 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
44255 Notification packets are distinct from any other communication from
44256 the stub so there is no ambiguity.
44258 After receiving a notification, @value{GDBN} shall acknowledge it by
44259 sending a @var{ack} packet as a regular, synchronous request to the
44260 stub. Such acknowledgment is not required to happen immediately, as
44261 @value{GDBN} is permitted to send other, unrelated packets to the
44262 stub first, which the stub should process normally.
44264 Upon receiving a @var{ack} packet, if the stub has other queued
44265 events to report to @value{GDBN}, it shall respond by sending a
44266 normal @var{event}. @value{GDBN} shall then send another @var{ack}
44267 packet to solicit further responses; again, it is permitted to send
44268 other, unrelated packets as well which the stub should process
44271 If the stub receives a @var{ack} packet and there are no additional
44272 @var{event} to report, the stub shall return an @samp{OK} response.
44273 At this point, @value{GDBN} has finished processing a notification
44274 and the stub has completed sending any queued events. @value{GDBN}
44275 won't accept any new notifications until the final @samp{OK} is
44276 received . If further notification events occur, the stub shall send
44277 a new notification, @value{GDBN} shall accept the notification, and
44278 the process shall be repeated.
44280 The process of asynchronous notification can be illustrated by the
44283 <- @code{%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
44286 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
44288 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
44293 The following notifications are defined:
44294 @multitable @columnfractions 0.12 0.12 0.38 0.38
44303 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
44304 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
44305 for information on how these notifications are acknowledged by
44307 @tab Report an asynchronous stop event in non-stop mode.
44311 @node Remote Non-Stop
44312 @section Remote Protocol Support for Non-Stop Mode
44314 @value{GDBN}'s remote protocol supports non-stop debugging of
44315 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
44316 supports non-stop mode, it should report that to @value{GDBN} by including
44317 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
44319 @value{GDBN} typically sends a @samp{QNonStop} packet only when
44320 establishing a new connection with the stub. Entering non-stop mode
44321 does not alter the state of any currently-running threads, but targets
44322 must stop all threads in any already-attached processes when entering
44323 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
44324 probe the target state after a mode change.
44326 In non-stop mode, when an attached process encounters an event that
44327 would otherwise be reported with a stop reply, it uses the
44328 asynchronous notification mechanism (@pxref{Notification Packets}) to
44329 inform @value{GDBN}. In contrast to all-stop mode, where all threads
44330 in all processes are stopped when a stop reply is sent, in non-stop
44331 mode only the thread reporting the stop event is stopped. That is,
44332 when reporting a @samp{S} or @samp{T} response to indicate completion
44333 of a step operation, hitting a breakpoint, or a fault, only the
44334 affected thread is stopped; any other still-running threads continue
44335 to run. When reporting a @samp{W} or @samp{X} response, all running
44336 threads belonging to other attached processes continue to run.
44338 In non-stop mode, the target shall respond to the @samp{?} packet as
44339 follows. First, any incomplete stop reply notification/@samp{vStopped}
44340 sequence in progress is abandoned. The target must begin a new
44341 sequence reporting stop events for all stopped threads, whether or not
44342 it has previously reported those events to @value{GDBN}. The first
44343 stop reply is sent as a synchronous reply to the @samp{?} packet, and
44344 subsequent stop replies are sent as responses to @samp{vStopped} packets
44345 using the mechanism described above. The target must not send
44346 asynchronous stop reply notifications until the sequence is complete.
44347 If all threads are running when the target receives the @samp{?} packet,
44348 or if the target is not attached to any process, it shall respond
44351 If the stub supports non-stop mode, it should also support the
44352 @samp{swbreak} stop reason if software breakpoints are supported, and
44353 the @samp{hwbreak} stop reason if hardware breakpoints are supported
44354 (@pxref{swbreak stop reason}). This is because given the asynchronous
44355 nature of non-stop mode, between the time a thread hits a breakpoint
44356 and the time the event is finally processed by @value{GDBN}, the
44357 breakpoint may have already been removed from the target. Due to
44358 this, @value{GDBN} needs to be able to tell whether a trap stop was
44359 caused by a delayed breakpoint event, which should be ignored, as
44360 opposed to a random trap signal, which should be reported to the user.
44361 Note the @samp{swbreak} feature implies that the target is responsible
44362 for adjusting the PC when a software breakpoint triggers, if
44363 necessary, such as on the x86 architecture.
44365 @node Packet Acknowledgment
44366 @section Packet Acknowledgment
44368 @cindex acknowledgment, for @value{GDBN} remote
44369 @cindex packet acknowledgment, for @value{GDBN} remote
44370 By default, when either the host or the target machine receives a packet,
44371 the first response expected is an acknowledgment: either @samp{+} (to indicate
44372 the package was received correctly) or @samp{-} (to request retransmission).
44373 This mechanism allows the @value{GDBN} remote protocol to operate over
44374 unreliable transport mechanisms, such as a serial line.
44376 In cases where the transport mechanism is itself reliable (such as a pipe or
44377 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
44378 It may be desirable to disable them in that case to reduce communication
44379 overhead, or for other reasons. This can be accomplished by means of the
44380 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
44382 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
44383 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
44384 and response format still includes the normal checksum, as described in
44385 @ref{Overview}, but the checksum may be ignored by the receiver.
44387 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
44388 no-acknowledgment mode, it should report that to @value{GDBN}
44389 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
44390 @pxref{qSupported}.
44391 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
44392 disabled via the @code{set remote noack-packet off} command
44393 (@pxref{Remote Configuration}),
44394 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
44395 Only then may the stub actually turn off packet acknowledgments.
44396 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
44397 response, which can be safely ignored by the stub.
44399 Note that @code{set remote noack-packet} command only affects negotiation
44400 between @value{GDBN} and the stub when subsequent connections are made;
44401 it does not affect the protocol acknowledgment state for any current
44403 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
44404 new connection is established,
44405 there is also no protocol request to re-enable the acknowledgments
44406 for the current connection, once disabled.
44411 Example sequence of a target being re-started. Notice how the restart
44412 does not get any direct output:
44417 @emph{target restarts}
44420 <- @code{T001:1234123412341234}
44424 Example sequence of a target being stepped by a single instruction:
44427 -> @code{G1445@dots{}}
44432 <- @code{T001:1234123412341234}
44436 <- @code{1455@dots{}}
44440 @node File-I/O Remote Protocol Extension
44441 @section File-I/O Remote Protocol Extension
44442 @cindex File-I/O remote protocol extension
44445 * File-I/O Overview::
44446 * Protocol Basics::
44447 * The F Request Packet::
44448 * The F Reply Packet::
44449 * The Ctrl-C Message::
44451 * List of Supported Calls::
44452 * Protocol-specific Representation of Datatypes::
44454 * File-I/O Examples::
44457 @node File-I/O Overview
44458 @subsection File-I/O Overview
44459 @cindex file-i/o overview
44461 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
44462 target to use the host's file system and console I/O to perform various
44463 system calls. System calls on the target system are translated into a
44464 remote protocol packet to the host system, which then performs the needed
44465 actions and returns a response packet to the target system.
44466 This simulates file system operations even on targets that lack file systems.
44468 The protocol is defined to be independent of both the host and target systems.
44469 It uses its own internal representation of datatypes and values. Both
44470 @value{GDBN} and the target's @value{GDBN} stub are responsible for
44471 translating the system-dependent value representations into the internal
44472 protocol representations when data is transmitted.
44474 The communication is synchronous. A system call is possible only when
44475 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
44476 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
44477 the target is stopped to allow deterministic access to the target's
44478 memory. Therefore File-I/O is not interruptible by target signals. On
44479 the other hand, it is possible to interrupt File-I/O by a user interrupt
44480 (@samp{Ctrl-C}) within @value{GDBN}.
44482 The target's request to perform a host system call does not finish
44483 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
44484 after finishing the system call, the target returns to continuing the
44485 previous activity (continue, step). No additional continue or step
44486 request from @value{GDBN} is required.
44489 (@value{GDBP}) continue
44490 <- target requests 'system call X'
44491 target is stopped, @value{GDBN} executes system call
44492 -> @value{GDBN} returns result
44493 ... target continues, @value{GDBN} returns to wait for the target
44494 <- target hits breakpoint and sends a Txx packet
44497 The protocol only supports I/O on the console and to regular files on
44498 the host file system. Character or block special devices, pipes,
44499 named pipes, sockets or any other communication method on the host
44500 system are not supported by this protocol.
44502 File I/O is not supported in non-stop mode.
44504 @node Protocol Basics
44505 @subsection Protocol Basics
44506 @cindex protocol basics, file-i/o
44508 The File-I/O protocol uses the @code{F} packet as the request as well
44509 as reply packet. Since a File-I/O system call can only occur when
44510 @value{GDBN} is waiting for a response from the continuing or stepping target,
44511 the File-I/O request is a reply that @value{GDBN} has to expect as a result
44512 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
44513 This @code{F} packet contains all information needed to allow @value{GDBN}
44514 to call the appropriate host system call:
44518 A unique identifier for the requested system call.
44521 All parameters to the system call. Pointers are given as addresses
44522 in the target memory address space. Pointers to strings are given as
44523 pointer/length pair. Numerical values are given as they are.
44524 Numerical control flags are given in a protocol-specific representation.
44528 At this point, @value{GDBN} has to perform the following actions.
44532 If the parameters include pointer values to data needed as input to a
44533 system call, @value{GDBN} requests this data from the target with a
44534 standard @code{m} packet request. This additional communication has to be
44535 expected by the target implementation and is handled as any other @code{m}
44539 @value{GDBN} translates all value from protocol representation to host
44540 representation as needed. Datatypes are coerced into the host types.
44543 @value{GDBN} calls the system call.
44546 It then coerces datatypes back to protocol representation.
44549 If the system call is expected to return data in buffer space specified
44550 by pointer parameters to the call, the data is transmitted to the
44551 target using a @code{M} or @code{X} packet. This packet has to be expected
44552 by the target implementation and is handled as any other @code{M} or @code{X}
44557 Eventually @value{GDBN} replies with another @code{F} packet which contains all
44558 necessary information for the target to continue. This at least contains
44565 @code{errno}, if has been changed by the system call.
44572 After having done the needed type and value coercion, the target continues
44573 the latest continue or step action.
44575 @node The F Request Packet
44576 @subsection The @code{F} Request Packet
44577 @cindex file-i/o request packet
44578 @cindex @code{F} request packet
44580 The @code{F} request packet has the following format:
44583 @item F@var{call-id},@var{parameter@dots{}}
44585 @var{call-id} is the identifier to indicate the host system call to be called.
44586 This is just the name of the function.
44588 @var{parameter@dots{}} are the parameters to the system call.
44589 Parameters are hexadecimal integer values, either the actual values in case
44590 of scalar datatypes, pointers to target buffer space in case of compound
44591 datatypes and unspecified memory areas, or pointer/length pairs in case
44592 of string parameters. These are appended to the @var{call-id} as a
44593 comma-delimited list. All values are transmitted in ASCII
44594 string representation, pointer/length pairs separated by a slash.
44600 @node The F Reply Packet
44601 @subsection The @code{F} Reply Packet
44602 @cindex file-i/o reply packet
44603 @cindex @code{F} reply packet
44605 The @code{F} reply packet has the following format:
44609 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
44611 @var{retcode} is the return code of the system call as hexadecimal value.
44613 @var{errno} is the @code{errno} set by the call, in protocol-specific
44615 This parameter can be omitted if the call was successful.
44617 @var{Ctrl-C flag} is only sent if the user requested a break. In this
44618 case, @var{errno} must be sent as well, even if the call was successful.
44619 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
44626 or, if the call was interrupted before the host call has been performed:
44633 assuming 4 is the protocol-specific representation of @code{EINTR}.
44638 @node The Ctrl-C Message
44639 @subsection The @samp{Ctrl-C} Message
44640 @cindex ctrl-c message, in file-i/o protocol
44642 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
44643 reply packet (@pxref{The F Reply Packet}),
44644 the target should behave as if it had
44645 gotten a break message. The meaning for the target is ``system call
44646 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
44647 (as with a break message) and return to @value{GDBN} with a @code{T02}
44650 It's important for the target to know in which
44651 state the system call was interrupted. There are two possible cases:
44655 The system call hasn't been performed on the host yet.
44658 The system call on the host has been finished.
44662 These two states can be distinguished by the target by the value of the
44663 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
44664 call hasn't been performed. This is equivalent to the @code{EINTR} handling
44665 on POSIX systems. In any other case, the target may presume that the
44666 system call has been finished --- successfully or not --- and should behave
44667 as if the break message arrived right after the system call.
44669 @value{GDBN} must behave reliably. If the system call has not been called
44670 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
44671 @code{errno} in the packet. If the system call on the host has been finished
44672 before the user requests a break, the full action must be finished by
44673 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
44674 The @code{F} packet may only be sent when either nothing has happened
44675 or the full action has been completed.
44678 @subsection Console I/O
44679 @cindex console i/o as part of file-i/o
44681 By default and if not explicitly closed by the target system, the file
44682 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
44683 on the @value{GDBN} console is handled as any other file output operation
44684 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
44685 by @value{GDBN} so that after the target read request from file descriptor
44686 0 all following typing is buffered until either one of the following
44691 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
44693 system call is treated as finished.
44696 The user presses @key{RET}. This is treated as end of input with a trailing
44700 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
44701 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
44705 If the user has typed more characters than fit in the buffer given to
44706 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
44707 either another @code{read(0, @dots{})} is requested by the target, or debugging
44708 is stopped at the user's request.
44711 @node List of Supported Calls
44712 @subsection List of Supported Calls
44713 @cindex list of supported file-i/o calls
44730 @unnumberedsubsubsec open
44731 @cindex open, file-i/o system call
44736 int open(const char *pathname, int flags);
44737 int open(const char *pathname, int flags, mode_t mode);
44741 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
44744 @var{flags} is the bitwise @code{OR} of the following values:
44748 If the file does not exist it will be created. The host
44749 rules apply as far as file ownership and time stamps
44753 When used with @code{O_CREAT}, if the file already exists it is
44754 an error and open() fails.
44757 If the file already exists and the open mode allows
44758 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
44759 truncated to zero length.
44762 The file is opened in append mode.
44765 The file is opened for reading only.
44768 The file is opened for writing only.
44771 The file is opened for reading and writing.
44775 Other bits are silently ignored.
44779 @var{mode} is the bitwise @code{OR} of the following values:
44783 User has read permission.
44786 User has write permission.
44789 Group has read permission.
44792 Group has write permission.
44795 Others have read permission.
44798 Others have write permission.
44802 Other bits are silently ignored.
44805 @item Return value:
44806 @code{open} returns the new file descriptor or -1 if an error
44813 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
44816 @var{pathname} refers to a directory.
44819 The requested access is not allowed.
44822 @var{pathname} was too long.
44825 A directory component in @var{pathname} does not exist.
44828 @var{pathname} refers to a device, pipe, named pipe or socket.
44831 @var{pathname} refers to a file on a read-only filesystem and
44832 write access was requested.
44835 @var{pathname} is an invalid pointer value.
44838 No space on device to create the file.
44841 The process already has the maximum number of files open.
44844 The limit on the total number of files open on the system
44848 The call was interrupted by the user.
44854 @unnumberedsubsubsec close
44855 @cindex close, file-i/o system call
44864 @samp{Fclose,@var{fd}}
44866 @item Return value:
44867 @code{close} returns zero on success, or -1 if an error occurred.
44873 @var{fd} isn't a valid open file descriptor.
44876 The call was interrupted by the user.
44882 @unnumberedsubsubsec read
44883 @cindex read, file-i/o system call
44888 int read(int fd, void *buf, unsigned int count);
44892 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
44894 @item Return value:
44895 On success, the number of bytes read is returned.
44896 Zero indicates end of file. If count is zero, read
44897 returns zero as well. On error, -1 is returned.
44903 @var{fd} is not a valid file descriptor or is not open for
44907 @var{bufptr} is an invalid pointer value.
44910 The call was interrupted by the user.
44916 @unnumberedsubsubsec write
44917 @cindex write, file-i/o system call
44922 int write(int fd, const void *buf, unsigned int count);
44926 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
44928 @item Return value:
44929 On success, the number of bytes written are returned.
44930 Zero indicates nothing was written. On error, -1
44937 @var{fd} is not a valid file descriptor or is not open for
44941 @var{bufptr} is an invalid pointer value.
44944 An attempt was made to write a file that exceeds the
44945 host-specific maximum file size allowed.
44948 No space on device to write the data.
44951 The call was interrupted by the user.
44957 @unnumberedsubsubsec lseek
44958 @cindex lseek, file-i/o system call
44963 long lseek (int fd, long offset, int flag);
44967 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
44969 @var{flag} is one of:
44973 The offset is set to @var{offset} bytes.
44976 The offset is set to its current location plus @var{offset}
44980 The offset is set to the size of the file plus @var{offset}
44984 @item Return value:
44985 On success, the resulting unsigned offset in bytes from
44986 the beginning of the file is returned. Otherwise, a
44987 value of -1 is returned.
44993 @var{fd} is not a valid open file descriptor.
44996 @var{fd} is associated with the @value{GDBN} console.
44999 @var{flag} is not a proper value.
45002 The call was interrupted by the user.
45008 @unnumberedsubsubsec rename
45009 @cindex rename, file-i/o system call
45014 int rename(const char *oldpath, const char *newpath);
45018 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
45020 @item Return value:
45021 On success, zero is returned. On error, -1 is returned.
45027 @var{newpath} is an existing directory, but @var{oldpath} is not a
45031 @var{newpath} is a non-empty directory.
45034 @var{oldpath} or @var{newpath} is a directory that is in use by some
45038 An attempt was made to make a directory a subdirectory
45042 A component used as a directory in @var{oldpath} or new
45043 path is not a directory. Or @var{oldpath} is a directory
45044 and @var{newpath} exists but is not a directory.
45047 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
45050 No access to the file or the path of the file.
45054 @var{oldpath} or @var{newpath} was too long.
45057 A directory component in @var{oldpath} or @var{newpath} does not exist.
45060 The file is on a read-only filesystem.
45063 The device containing the file has no room for the new
45067 The call was interrupted by the user.
45073 @unnumberedsubsubsec unlink
45074 @cindex unlink, file-i/o system call
45079 int unlink(const char *pathname);
45083 @samp{Funlink,@var{pathnameptr}/@var{len}}
45085 @item Return value:
45086 On success, zero is returned. On error, -1 is returned.
45092 No access to the file or the path of the file.
45095 The system does not allow unlinking of directories.
45098 The file @var{pathname} cannot be unlinked because it's
45099 being used by another process.
45102 @var{pathnameptr} is an invalid pointer value.
45105 @var{pathname} was too long.
45108 A directory component in @var{pathname} does not exist.
45111 A component of the path is not a directory.
45114 The file is on a read-only filesystem.
45117 The call was interrupted by the user.
45123 @unnumberedsubsubsec stat/fstat
45124 @cindex fstat, file-i/o system call
45125 @cindex stat, file-i/o system call
45130 int stat(const char *pathname, struct stat *buf);
45131 int fstat(int fd, struct stat *buf);
45135 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
45136 @samp{Ffstat,@var{fd},@var{bufptr}}
45138 @item Return value:
45139 On success, zero is returned. On error, -1 is returned.
45145 @var{fd} is not a valid open file.
45148 A directory component in @var{pathname} does not exist or the
45149 path is an empty string.
45152 A component of the path is not a directory.
45155 @var{pathnameptr} is an invalid pointer value.
45158 No access to the file or the path of the file.
45161 @var{pathname} was too long.
45164 The call was interrupted by the user.
45170 @unnumberedsubsubsec gettimeofday
45171 @cindex gettimeofday, file-i/o system call
45176 int gettimeofday(struct timeval *tv, void *tz);
45180 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
45182 @item Return value:
45183 On success, 0 is returned, -1 otherwise.
45189 @var{tz} is a non-NULL pointer.
45192 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
45198 @unnumberedsubsubsec isatty
45199 @cindex isatty, file-i/o system call
45204 int isatty(int fd);
45208 @samp{Fisatty,@var{fd}}
45210 @item Return value:
45211 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
45217 The call was interrupted by the user.
45222 Note that the @code{isatty} call is treated as a special case: it returns
45223 1 to the target if the file descriptor is attached
45224 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
45225 would require implementing @code{ioctl} and would be more complex than
45230 @unnumberedsubsubsec system
45231 @cindex system, file-i/o system call
45236 int system(const char *command);
45240 @samp{Fsystem,@var{commandptr}/@var{len}}
45242 @item Return value:
45243 If @var{len} is zero, the return value indicates whether a shell is
45244 available. A zero return value indicates a shell is not available.
45245 For non-zero @var{len}, the value returned is -1 on error and the
45246 return status of the command otherwise. Only the exit status of the
45247 command is returned, which is extracted from the host's @code{system}
45248 return value by calling @code{WEXITSTATUS(retval)}. In case
45249 @file{/bin/sh} could not be executed, 127 is returned.
45255 The call was interrupted by the user.
45260 @value{GDBN} takes over the full task of calling the necessary host calls
45261 to perform the @code{system} call. The return value of @code{system} on
45262 the host is simplified before it's returned
45263 to the target. Any termination signal information from the child process
45264 is discarded, and the return value consists
45265 entirely of the exit status of the called command.
45267 Due to security concerns, the @code{system} call is by default refused
45268 by @value{GDBN}. The user has to allow this call explicitly with the
45269 @code{set remote system-call-allowed 1} command.
45272 @item set remote system-call-allowed
45273 @kindex set remote system-call-allowed
45274 Control whether to allow the @code{system} calls in the File I/O
45275 protocol for the remote target. The default is zero (disabled).
45277 @item show remote system-call-allowed
45278 @kindex show remote system-call-allowed
45279 Show whether the @code{system} calls are allowed in the File I/O
45283 @node Protocol-specific Representation of Datatypes
45284 @subsection Protocol-specific Representation of Datatypes
45285 @cindex protocol-specific representation of datatypes, in file-i/o protocol
45288 * Integral Datatypes::
45290 * Memory Transfer::
45295 @node Integral Datatypes
45296 @unnumberedsubsubsec Integral Datatypes
45297 @cindex integral datatypes, in file-i/o protocol
45299 The integral datatypes used in the system calls are @code{int},
45300 @code{unsigned int}, @code{long}, @code{unsigned long},
45301 @code{mode_t}, and @code{time_t}.
45303 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
45304 implemented as 32 bit values in this protocol.
45306 @code{long} and @code{unsigned long} are implemented as 64 bit types.
45308 @xref{Limits}, for corresponding MIN and MAX values (similar to those
45309 in @file{limits.h}) to allow range checking on host and target.
45311 @code{time_t} datatypes are defined as seconds since the Epoch.
45313 All integral datatypes transferred as part of a memory read or write of a
45314 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
45317 @node Pointer Values
45318 @unnumberedsubsubsec Pointer Values
45319 @cindex pointer values, in file-i/o protocol
45321 Pointers to target data are transmitted as they are. An exception
45322 is made for pointers to buffers for which the length isn't
45323 transmitted as part of the function call, namely strings. Strings
45324 are transmitted as a pointer/length pair, both as hex values, e.g.@:
45331 which is a pointer to data of length 18 bytes at position 0x1aaf.
45332 The length is defined as the full string length in bytes, including
45333 the trailing null byte. For example, the string @code{"hello world"}
45334 at address 0x123456 is transmitted as
45340 @node Memory Transfer
45341 @unnumberedsubsubsec Memory Transfer
45342 @cindex memory transfer, in file-i/o protocol
45344 Structured data which is transferred using a memory read or write (for
45345 example, a @code{struct stat}) is expected to be in a protocol-specific format
45346 with all scalar multibyte datatypes being big endian. Translation to
45347 this representation needs to be done both by the target before the @code{F}
45348 packet is sent, and by @value{GDBN} before
45349 it transfers memory to the target. Transferred pointers to structured
45350 data should point to the already-coerced data at any time.
45354 @unnumberedsubsubsec struct stat
45355 @cindex struct stat, in file-i/o protocol
45357 The buffer of type @code{struct stat} used by the target and @value{GDBN}
45358 is defined as follows:
45362 unsigned int st_dev; /* device */
45363 unsigned int st_ino; /* inode */
45364 mode_t st_mode; /* protection */
45365 unsigned int st_nlink; /* number of hard links */
45366 unsigned int st_uid; /* user ID of owner */
45367 unsigned int st_gid; /* group ID of owner */
45368 unsigned int st_rdev; /* device type (if inode device) */
45369 unsigned long st_size; /* total size, in bytes */
45370 unsigned long st_blksize; /* blocksize for filesystem I/O */
45371 unsigned long st_blocks; /* number of blocks allocated */
45372 time_t st_atime; /* time of last access */
45373 time_t st_mtime; /* time of last modification */
45374 time_t st_ctime; /* time of last change */
45378 The integral datatypes conform to the definitions given in the
45379 appropriate section (see @ref{Integral Datatypes}, for details) so this
45380 structure is of size 64 bytes.
45382 The values of several fields have a restricted meaning and/or
45388 A value of 0 represents a file, 1 the console.
45391 No valid meaning for the target. Transmitted unchanged.
45394 Valid mode bits are described in @ref{Constants}. Any other
45395 bits have currently no meaning for the target.
45400 No valid meaning for the target. Transmitted unchanged.
45405 These values have a host and file system dependent
45406 accuracy. Especially on Windows hosts, the file system may not
45407 support exact timing values.
45410 The target gets a @code{struct stat} of the above representation and is
45411 responsible for coercing it to the target representation before
45414 Note that due to size differences between the host, target, and protocol
45415 representations of @code{struct stat} members, these members could eventually
45416 get truncated on the target.
45418 @node struct timeval
45419 @unnumberedsubsubsec struct timeval
45420 @cindex struct timeval, in file-i/o protocol
45422 The buffer of type @code{struct timeval} used by the File-I/O protocol
45423 is defined as follows:
45427 time_t tv_sec; /* second */
45428 long tv_usec; /* microsecond */
45432 The integral datatypes conform to the definitions given in the
45433 appropriate section (see @ref{Integral Datatypes}, for details) so this
45434 structure is of size 8 bytes.
45437 @subsection Constants
45438 @cindex constants, in file-i/o protocol
45440 The following values are used for the constants inside of the
45441 protocol. @value{GDBN} and target are responsible for translating these
45442 values before and after the call as needed.
45453 @unnumberedsubsubsec Open Flags
45454 @cindex open flags, in file-i/o protocol
45456 All values are given in hexadecimal representation.
45468 @node mode_t Values
45469 @unnumberedsubsubsec mode_t Values
45470 @cindex mode_t values, in file-i/o protocol
45472 All values are given in octal representation.
45489 @unnumberedsubsubsec Errno Values
45490 @cindex errno values, in file-i/o protocol
45492 All values are given in decimal representation.
45517 @code{EUNKNOWN} is used as a fallback error value if a host system returns
45518 any error value not in the list of supported error numbers.
45521 @unnumberedsubsubsec Lseek Flags
45522 @cindex lseek flags, in file-i/o protocol
45531 @unnumberedsubsubsec Limits
45532 @cindex limits, in file-i/o protocol
45534 All values are given in decimal representation.
45537 INT_MIN -2147483648
45539 UINT_MAX 4294967295
45540 LONG_MIN -9223372036854775808
45541 LONG_MAX 9223372036854775807
45542 ULONG_MAX 18446744073709551615
45545 @node File-I/O Examples
45546 @subsection File-I/O Examples
45547 @cindex file-i/o examples
45549 Example sequence of a write call, file descriptor 3, buffer is at target
45550 address 0x1234, 6 bytes should be written:
45553 <- @code{Fwrite,3,1234,6}
45554 @emph{request memory read from target}
45557 @emph{return "6 bytes written"}
45561 Example sequence of a read call, file descriptor 3, buffer is at target
45562 address 0x1234, 6 bytes should be read:
45565 <- @code{Fread,3,1234,6}
45566 @emph{request memory write to target}
45567 -> @code{X1234,6:XXXXXX}
45568 @emph{return "6 bytes read"}
45572 Example sequence of a read call, call fails on the host due to invalid
45573 file descriptor (@code{EBADF}):
45576 <- @code{Fread,3,1234,6}
45580 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
45584 <- @code{Fread,3,1234,6}
45589 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
45593 <- @code{Fread,3,1234,6}
45594 -> @code{X1234,6:XXXXXX}
45598 @node Library List Format
45599 @section Library List Format
45600 @cindex library list format, remote protocol
45602 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
45603 same process as your application to manage libraries. In this case,
45604 @value{GDBN} can use the loader's symbol table and normal memory
45605 operations to maintain a list of shared libraries. On other
45606 platforms, the operating system manages loaded libraries.
45607 @value{GDBN} can not retrieve the list of currently loaded libraries
45608 through memory operations, so it uses the @samp{qXfer:libraries:read}
45609 packet (@pxref{qXfer library list read}) instead. The remote stub
45610 queries the target's operating system and reports which libraries
45613 The @samp{qXfer:libraries:read} packet returns an XML document which
45614 lists loaded libraries and their offsets. Each library has an
45615 associated name and one or more segment or section base addresses,
45616 which report where the library was loaded in memory.
45618 For the common case of libraries that are fully linked binaries, the
45619 library should have a list of segments. If the target supports
45620 dynamic linking of a relocatable object file, its library XML element
45621 should instead include a list of allocated sections. The segment or
45622 section bases are start addresses, not relocation offsets; they do not
45623 depend on the library's link-time base addresses.
45625 @value{GDBN} must be linked with the Expat library to support XML
45626 library lists. @xref{Expat}.
45628 A simple memory map, with one loaded library relocated by a single
45629 offset, looks like this:
45633 <library name="/lib/libc.so.6">
45634 <segment address="0x10000000"/>
45639 Another simple memory map, with one loaded library with three
45640 allocated sections (.text, .data, .bss), looks like this:
45644 <library name="sharedlib.o">
45645 <section address="0x10000000"/>
45646 <section address="0x20000000"/>
45647 <section address="0x30000000"/>
45652 The format of a library list is described by this DTD:
45655 <!-- library-list: Root element with versioning -->
45656 <!ELEMENT library-list (library)*>
45657 <!ATTLIST library-list version CDATA #FIXED "1.0">
45658 <!ELEMENT library (segment*, section*)>
45659 <!ATTLIST library name CDATA #REQUIRED>
45660 <!ELEMENT segment EMPTY>
45661 <!ATTLIST segment address CDATA #REQUIRED>
45662 <!ELEMENT section EMPTY>
45663 <!ATTLIST section address CDATA #REQUIRED>
45666 In addition, segments and section descriptors cannot be mixed within a
45667 single library element, and you must supply at least one segment or
45668 section for each library.
45670 @node Library List Format for SVR4 Targets
45671 @section Library List Format for SVR4 Targets
45672 @cindex library list format, remote protocol
45674 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
45675 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
45676 shared libraries. Still a special library list provided by this packet is
45677 more efficient for the @value{GDBN} remote protocol.
45679 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
45680 loaded libraries and their SVR4 linker parameters. For each library on SVR4
45681 target, the following parameters are reported:
45685 @code{name}, the absolute file name from the @code{l_name} field of
45686 @code{struct link_map}.
45688 @code{lm} with address of @code{struct link_map} used for TLS
45689 (Thread Local Storage) access.
45691 @code{l_addr}, the displacement as read from the field @code{l_addr} of
45692 @code{struct link_map}. For prelinked libraries this is not an absolute
45693 memory address. It is a displacement of absolute memory address against
45694 address the file was prelinked to during the library load.
45696 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
45699 Additionally the single @code{main-lm} attribute specifies address of
45700 @code{struct link_map} used for the main executable. This parameter is used
45701 for TLS access and its presence is optional.
45703 @value{GDBN} must be linked with the Expat library to support XML
45704 SVR4 library lists. @xref{Expat}.
45706 A simple memory map, with two loaded libraries (which do not use prelink),
45710 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
45711 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
45713 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
45715 </library-list-svr>
45718 The format of an SVR4 library list is described by this DTD:
45721 <!-- library-list-svr4: Root element with versioning -->
45722 <!ELEMENT library-list-svr4 (library)*>
45723 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
45724 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
45725 <!ELEMENT library EMPTY>
45726 <!ATTLIST library name CDATA #REQUIRED>
45727 <!ATTLIST library lm CDATA #REQUIRED>
45728 <!ATTLIST library l_addr CDATA #REQUIRED>
45729 <!ATTLIST library l_ld CDATA #REQUIRED>
45732 @node Memory Map Format
45733 @section Memory Map Format
45734 @cindex memory map format
45736 To be able to write into flash memory, @value{GDBN} needs to obtain a
45737 memory map from the target. This section describes the format of the
45740 The memory map is obtained using the @samp{qXfer:memory-map:read}
45741 (@pxref{qXfer memory map read}) packet and is an XML document that
45742 lists memory regions.
45744 @value{GDBN} must be linked with the Expat library to support XML
45745 memory maps. @xref{Expat}.
45747 The top-level structure of the document is shown below:
45750 <?xml version="1.0"?>
45751 <!DOCTYPE memory-map
45752 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
45753 "http://sourceware.org/gdb/gdb-memory-map.dtd">
45759 Each region can be either:
45764 A region of RAM starting at @var{addr} and extending for @var{length}
45768 <memory type="ram" start="@var{addr}" length="@var{length}"/>
45773 A region of read-only memory:
45776 <memory type="rom" start="@var{addr}" length="@var{length}"/>
45781 A region of flash memory, with erasure blocks @var{blocksize}
45785 <memory type="flash" start="@var{addr}" length="@var{length}">
45786 <property name="blocksize">@var{blocksize}</property>
45792 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
45793 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
45794 packets to write to addresses in such ranges.
45796 The formal DTD for memory map format is given below:
45799 <!-- ................................................... -->
45800 <!-- Memory Map XML DTD ................................ -->
45801 <!-- File: memory-map.dtd .............................. -->
45802 <!-- .................................... .............. -->
45803 <!-- memory-map.dtd -->
45804 <!-- memory-map: Root element with versioning -->
45805 <!ELEMENT memory-map (memory)*>
45806 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
45807 <!ELEMENT memory (property)*>
45808 <!-- memory: Specifies a memory region,
45809 and its type, or device. -->
45810 <!ATTLIST memory type (ram|rom|flash) #REQUIRED
45811 start CDATA #REQUIRED
45812 length CDATA #REQUIRED>
45813 <!-- property: Generic attribute tag -->
45814 <!ELEMENT property (#PCDATA | property)*>
45815 <!ATTLIST property name (blocksize) #REQUIRED>
45818 @node Thread List Format
45819 @section Thread List Format
45820 @cindex thread list format
45822 To efficiently update the list of threads and their attributes,
45823 @value{GDBN} issues the @samp{qXfer:threads:read} packet
45824 (@pxref{qXfer threads read}) and obtains the XML document with
45825 the following structure:
45828 <?xml version="1.0"?>
45830 <thread id="id" core="0" name="name">
45831 ... description ...
45836 Each @samp{thread} element must have the @samp{id} attribute that
45837 identifies the thread (@pxref{thread-id syntax}). The
45838 @samp{core} attribute, if present, specifies which processor core
45839 the thread was last executing on. The @samp{name} attribute, if
45840 present, specifies the human-readable name of the thread. The content
45841 of the of @samp{thread} element is interpreted as human-readable
45842 auxiliary information. The @samp{handle} attribute, if present,
45843 is a hex encoded representation of the thread handle.
45846 @node Traceframe Info Format
45847 @section Traceframe Info Format
45848 @cindex traceframe info format
45850 To be able to know which objects in the inferior can be examined when
45851 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
45852 memory ranges, registers and trace state variables that have been
45853 collected in a traceframe.
45855 This list is obtained using the @samp{qXfer:traceframe-info:read}
45856 (@pxref{qXfer traceframe info read}) packet and is an XML document.
45858 @value{GDBN} must be linked with the Expat library to support XML
45859 traceframe info discovery. @xref{Expat}.
45861 The top-level structure of the document is shown below:
45864 <?xml version="1.0"?>
45865 <!DOCTYPE traceframe-info
45866 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
45867 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
45873 Each traceframe block can be either:
45878 A region of collected memory starting at @var{addr} and extending for
45879 @var{length} bytes from there:
45882 <memory start="@var{addr}" length="@var{length}"/>
45886 A block indicating trace state variable numbered @var{number} has been
45890 <tvar id="@var{number}"/>
45895 The formal DTD for the traceframe info format is given below:
45898 <!ELEMENT traceframe-info (memory | tvar)* >
45899 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
45901 <!ELEMENT memory EMPTY>
45902 <!ATTLIST memory start CDATA #REQUIRED
45903 length CDATA #REQUIRED>
45905 <!ATTLIST tvar id CDATA #REQUIRED>
45908 @node Branch Trace Format
45909 @section Branch Trace Format
45910 @cindex branch trace format
45912 In order to display the branch trace of an inferior thread,
45913 @value{GDBN} needs to obtain the list of branches. This list is
45914 represented as list of sequential code blocks that are connected via
45915 branches. The code in each block has been executed sequentially.
45917 This list is obtained using the @samp{qXfer:btrace:read}
45918 (@pxref{qXfer btrace read}) packet and is an XML document.
45920 @value{GDBN} must be linked with the Expat library to support XML
45921 traceframe info discovery. @xref{Expat}.
45923 The top-level structure of the document is shown below:
45926 <?xml version="1.0"?>
45928 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
45929 "http://sourceware.org/gdb/gdb-btrace.dtd">
45938 A block of sequentially executed instructions starting at @var{begin}
45939 and ending at @var{end}:
45942 <block begin="@var{begin}" end="@var{end}"/>
45947 The formal DTD for the branch trace format is given below:
45950 <!ELEMENT btrace (block* | pt) >
45951 <!ATTLIST btrace version CDATA #FIXED "1.0">
45953 <!ELEMENT block EMPTY>
45954 <!ATTLIST block begin CDATA #REQUIRED
45955 end CDATA #REQUIRED>
45957 <!ELEMENT pt (pt-config?, raw?)>
45959 <!ELEMENT pt-config (cpu?)>
45961 <!ELEMENT cpu EMPTY>
45962 <!ATTLIST cpu vendor CDATA #REQUIRED
45963 family CDATA #REQUIRED
45964 model CDATA #REQUIRED
45965 stepping CDATA #REQUIRED>
45967 <!ELEMENT raw (#PCDATA)>
45970 @node Branch Trace Configuration Format
45971 @section Branch Trace Configuration Format
45972 @cindex branch trace configuration format
45974 For each inferior thread, @value{GDBN} can obtain the branch trace
45975 configuration using the @samp{qXfer:btrace-conf:read}
45976 (@pxref{qXfer btrace-conf read}) packet.
45978 The configuration describes the branch trace format and configuration
45979 settings for that format. The following information is described:
45983 This thread uses the @dfn{Branch Trace Store} (@acronym{BTS}) format.
45986 The size of the @acronym{BTS} ring buffer in bytes.
45989 This thread uses the @dfn{Intel Processor Trace} (@acronym{Intel
45993 The size of the @acronym{Intel PT} ring buffer in bytes.
45997 @value{GDBN} must be linked with the Expat library to support XML
45998 branch trace configuration discovery. @xref{Expat}.
46000 The formal DTD for the branch trace configuration format is given below:
46003 <!ELEMENT btrace-conf (bts?, pt?)>
46004 <!ATTLIST btrace-conf version CDATA #FIXED "1.0">
46006 <!ELEMENT bts EMPTY>
46007 <!ATTLIST bts size CDATA #IMPLIED>
46009 <!ELEMENT pt EMPTY>
46010 <!ATTLIST pt size CDATA #IMPLIED>
46013 @include agentexpr.texi
46015 @node Target Descriptions
46016 @appendix Target Descriptions
46017 @cindex target descriptions
46019 One of the challenges of using @value{GDBN} to debug embedded systems
46020 is that there are so many minor variants of each processor
46021 architecture in use. It is common practice for vendors to start with
46022 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
46023 and then make changes to adapt it to a particular market niche. Some
46024 architectures have hundreds of variants, available from dozens of
46025 vendors. This leads to a number of problems:
46029 With so many different customized processors, it is difficult for
46030 the @value{GDBN} maintainers to keep up with the changes.
46032 Since individual variants may have short lifetimes or limited
46033 audiences, it may not be worthwhile to carry information about every
46034 variant in the @value{GDBN} source tree.
46036 When @value{GDBN} does support the architecture of the embedded system
46037 at hand, the task of finding the correct architecture name to give the
46038 @command{set architecture} command can be error-prone.
46041 To address these problems, the @value{GDBN} remote protocol allows a
46042 target system to not only identify itself to @value{GDBN}, but to
46043 actually describe its own features. This lets @value{GDBN} support
46044 processor variants it has never seen before --- to the extent that the
46045 descriptions are accurate, and that @value{GDBN} understands them.
46047 @value{GDBN} must be linked with the Expat library to support XML
46048 target descriptions. @xref{Expat}.
46051 * Retrieving Descriptions:: How descriptions are fetched from a target.
46052 * Target Description Format:: The contents of a target description.
46053 * Predefined Target Types:: Standard types available for target
46055 * Enum Target Types:: How to define enum target types.
46056 * Standard Target Features:: Features @value{GDBN} knows about.
46059 @node Retrieving Descriptions
46060 @section Retrieving Descriptions
46062 Target descriptions can be read from the target automatically, or
46063 specified by the user manually. The default behavior is to read the
46064 description from the target. @value{GDBN} retrieves it via the remote
46065 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
46066 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
46067 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
46068 XML document, of the form described in @ref{Target Description
46071 Alternatively, you can specify a file to read for the target description.
46072 If a file is set, the target will not be queried. The commands to
46073 specify a file are:
46076 @cindex set tdesc filename
46077 @item set tdesc filename @var{path}
46078 Read the target description from @var{path}.
46080 @cindex unset tdesc filename
46081 @item unset tdesc filename
46082 Do not read the XML target description from a file. @value{GDBN}
46083 will use the description supplied by the current target.
46085 @cindex show tdesc filename
46086 @item show tdesc filename
46087 Show the filename to read for a target description, if any.
46091 @node Target Description Format
46092 @section Target Description Format
46093 @cindex target descriptions, XML format
46095 A target description annex is an @uref{http://www.w3.org/XML/, XML}
46096 document which complies with the Document Type Definition provided in
46097 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
46098 means you can use generally available tools like @command{xmllint} to
46099 check that your feature descriptions are well-formed and valid.
46100 However, to help people unfamiliar with XML write descriptions for
46101 their targets, we also describe the grammar here.
46103 Target descriptions can identify the architecture of the remote target
46104 and (for some architectures) provide information about custom register
46105 sets. They can also identify the OS ABI of the remote target.
46106 @value{GDBN} can use this information to autoconfigure for your
46107 target, or to warn you if you connect to an unsupported target.
46109 Here is a simple target description:
46112 <target version="1.0">
46113 <architecture>i386:x86-64</architecture>
46118 This minimal description only says that the target uses
46119 the x86-64 architecture.
46121 A target description has the following overall form, with [ ] marking
46122 optional elements and @dots{} marking repeatable elements. The elements
46123 are explained further below.
46126 <?xml version="1.0"?>
46127 <!DOCTYPE target SYSTEM "gdb-target.dtd">
46128 <target version="1.0">
46129 @r{[}@var{architecture}@r{]}
46130 @r{[}@var{osabi}@r{]}
46131 @r{[}@var{compatible}@r{]}
46132 @r{[}@var{feature}@dots{}@r{]}
46137 The description is generally insensitive to whitespace and line
46138 breaks, under the usual common-sense rules. The XML version
46139 declaration and document type declaration can generally be omitted
46140 (@value{GDBN} does not require them), but specifying them may be
46141 useful for XML validation tools. The @samp{version} attribute for
46142 @samp{<target>} may also be omitted, but we recommend
46143 including it; if future versions of @value{GDBN} use an incompatible
46144 revision of @file{gdb-target.dtd}, they will detect and report
46145 the version mismatch.
46147 @subsection Inclusion
46148 @cindex target descriptions, inclusion
46151 @cindex <xi:include>
46154 It can sometimes be valuable to split a target description up into
46155 several different annexes, either for organizational purposes, or to
46156 share files between different possible target descriptions. You can
46157 divide a description into multiple files by replacing any element of
46158 the target description with an inclusion directive of the form:
46161 <xi:include href="@var{document}"/>
46165 When @value{GDBN} encounters an element of this form, it will retrieve
46166 the named XML @var{document}, and replace the inclusion directive with
46167 the contents of that document. If the current description was read
46168 using @samp{qXfer}, then so will be the included document;
46169 @var{document} will be interpreted as the name of an annex. If the
46170 current description was read from a file, @value{GDBN} will look for
46171 @var{document} as a file in the same directory where it found the
46172 original description.
46174 @subsection Architecture
46175 @cindex <architecture>
46177 An @samp{<architecture>} element has this form:
46180 <architecture>@var{arch}</architecture>
46183 @var{arch} is one of the architectures from the set accepted by
46184 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
46187 @cindex @code{<osabi>}
46189 This optional field was introduced in @value{GDBN} version 7.0.
46190 Previous versions of @value{GDBN} ignore it.
46192 An @samp{<osabi>} element has this form:
46195 <osabi>@var{abi-name}</osabi>
46198 @var{abi-name} is an OS ABI name from the same selection accepted by
46199 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
46201 @subsection Compatible Architecture
46202 @cindex @code{<compatible>}
46204 This optional field was introduced in @value{GDBN} version 7.0.
46205 Previous versions of @value{GDBN} ignore it.
46207 A @samp{<compatible>} element has this form:
46210 <compatible>@var{arch}</compatible>
46213 @var{arch} is one of the architectures from the set accepted by
46214 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
46216 A @samp{<compatible>} element is used to specify that the target
46217 is able to run binaries in some other than the main target architecture
46218 given by the @samp{<architecture>} element. For example, on the
46219 Cell Broadband Engine, the main architecture is @code{powerpc:common}
46220 or @code{powerpc:common64}, but the system is able to run binaries
46221 in the @code{spu} architecture as well. The way to describe this
46222 capability with @samp{<compatible>} is as follows:
46225 <architecture>powerpc:common</architecture>
46226 <compatible>spu</compatible>
46229 @subsection Features
46232 Each @samp{<feature>} describes some logical portion of the target
46233 system. Features are currently used to describe available CPU
46234 registers and the types of their contents. A @samp{<feature>} element
46238 <feature name="@var{name}">
46239 @r{[}@var{type}@dots{}@r{]}
46245 Each feature's name should be unique within the description. The name
46246 of a feature does not matter unless @value{GDBN} has some special
46247 knowledge of the contents of that feature; if it does, the feature
46248 should have its standard name. @xref{Standard Target Features}.
46252 Any register's value is a collection of bits which @value{GDBN} must
46253 interpret. The default interpretation is a two's complement integer,
46254 but other types can be requested by name in the register description.
46255 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
46256 Target Types}), and the description can define additional composite
46259 Each type element must have an @samp{id} attribute, which gives
46260 a unique (within the containing @samp{<feature>}) name to the type.
46261 Types must be defined before they are used.
46264 Some targets offer vector registers, which can be treated as arrays
46265 of scalar elements. These types are written as @samp{<vector>} elements,
46266 specifying the array element type, @var{type}, and the number of elements,
46270 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
46274 If a register's value is usefully viewed in multiple ways, define it
46275 with a union type containing the useful representations. The
46276 @samp{<union>} element contains one or more @samp{<field>} elements,
46277 each of which has a @var{name} and a @var{type}:
46280 <union id="@var{id}">
46281 <field name="@var{name}" type="@var{type}"/>
46288 If a register's value is composed from several separate values, define
46289 it with either a structure type or a flags type.
46290 A flags type may only contain bitfields.
46291 A structure type may either contain only bitfields or contain no bitfields.
46292 If the value contains only bitfields, its total size in bytes must be
46295 Non-bitfield values have a @var{name} and @var{type}.
46298 <struct id="@var{id}">
46299 <field name="@var{name}" type="@var{type}"/>
46304 Both @var{name} and @var{type} values are required.
46305 No implicit padding is added.
46307 Bitfield values have a @var{name}, @var{start}, @var{end} and @var{type}.
46310 <struct id="@var{id}" size="@var{size}">
46311 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
46317 <flags id="@var{id}" size="@var{size}">
46318 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
46323 The @var{name} value is required.
46324 Bitfield values may be named with the empty string, @samp{""},
46325 in which case the field is ``filler'' and its value is not printed.
46326 Not all bits need to be specified, so ``filler'' fields are optional.
46328 The @var{start} and @var{end} values are required, and @var{type}
46330 The field's @var{start} must be less than or equal to its @var{end},
46331 and zero represents the least significant bit.
46333 The default value of @var{type} is @code{bool} for single bit fields,
46334 and an unsigned integer otherwise.
46336 Which to choose? Structures or flags?
46338 Registers defined with @samp{flags} have these advantages over
46339 defining them with @samp{struct}:
46343 Arithmetic may be performed on them as if they were integers.
46345 They are printed in a more readable fashion.
46348 Registers defined with @samp{struct} have one advantage over
46349 defining them with @samp{flags}:
46353 One can fetch individual fields like in @samp{C}.
46356 (@value{GDBP}) print $my_struct_reg.field3
46362 @subsection Registers
46365 Each register is represented as an element with this form:
46368 <reg name="@var{name}"
46369 bitsize="@var{size}"
46370 @r{[}regnum="@var{num}"@r{]}
46371 @r{[}save-restore="@var{save-restore}"@r{]}
46372 @r{[}type="@var{type}"@r{]}
46373 @r{[}group="@var{group}"@r{]}/>
46377 The components are as follows:
46382 The register's name; it must be unique within the target description.
46385 The register's size, in bits.
46388 The register's number. If omitted, a register's number is one greater
46389 than that of the previous register (either in the current feature or in
46390 a preceding feature); the first register in the target description
46391 defaults to zero. This register number is used to read or write
46392 the register; e.g.@: it is used in the remote @code{p} and @code{P}
46393 packets, and registers appear in the @code{g} and @code{G} packets
46394 in order of increasing register number.
46397 Whether the register should be preserved across inferior function
46398 calls; this must be either @code{yes} or @code{no}. The default is
46399 @code{yes}, which is appropriate for most registers except for
46400 some system control registers; this is not related to the target's
46404 The type of the register. It may be a predefined type, a type
46405 defined in the current feature, or one of the special types @code{int}
46406 and @code{float}. @code{int} is an integer type of the correct size
46407 for @var{bitsize}, and @code{float} is a floating point type (in the
46408 architecture's normal floating point format) of the correct size for
46409 @var{bitsize}. The default is @code{int}.
46412 The register group to which this register belongs. It can be one of the
46413 standard register groups @code{general}, @code{float}, @code{vector} or an
46414 arbitrary string. Group names should be limited to alphanumeric characters.
46415 If a group name is made up of multiple words the words may be separated by
46416 hyphens; e.g.@: @code{special-group} or @code{ultra-special-group}. If no
46417 @var{group} is specified, @value{GDBN} will not display the register in
46418 @code{info registers}.
46422 @node Predefined Target Types
46423 @section Predefined Target Types
46424 @cindex target descriptions, predefined types
46426 Type definitions in the self-description can build up composite types
46427 from basic building blocks, but can not define fundamental types. Instead,
46428 standard identifiers are provided by @value{GDBN} for the fundamental
46429 types. The currently supported types are:
46434 Boolean type, occupying a single bit.
46442 Signed integer types holding the specified number of bits.
46450 Unsigned integer types holding the specified number of bits.
46454 Pointers to unspecified code and data. The program counter and
46455 any dedicated return address register may be marked as code
46456 pointers; printing a code pointer converts it into a symbolic
46457 address. The stack pointer and any dedicated address registers
46458 may be marked as data pointers.
46461 Single precision IEEE floating point.
46464 Double precision IEEE floating point.
46467 The 12-byte extended precision format used by ARM FPA registers.
46470 The 10-byte extended precision format used by x87 registers.
46473 32bit @sc{eflags} register used by x86.
46476 32bit @sc{mxcsr} register used by x86.
46480 @node Enum Target Types
46481 @section Enum Target Types
46482 @cindex target descriptions, enum types
46484 Enum target types are useful in @samp{struct} and @samp{flags}
46485 register descriptions. @xref{Target Description Format}.
46487 Enum types have a name, size and a list of name/value pairs.
46490 <enum id="@var{id}" size="@var{size}">
46491 <evalue name="@var{name}" value="@var{value}"/>
46496 Enums must be defined before they are used.
46499 <enum id="levels_type" size="4">
46500 <evalue name="low" value="0"/>
46501 <evalue name="high" value="1"/>
46503 <flags id="flags_type" size="4">
46504 <field name="X" start="0"/>
46505 <field name="LEVEL" start="1" end="1" type="levels_type"/>
46507 <reg name="flags" bitsize="32" type="flags_type"/>
46510 Given that description, a value of 3 for the @samp{flags} register
46511 would be printed as:
46514 (@value{GDBP}) info register flags
46515 flags 0x3 [ X LEVEL=high ]
46518 @node Standard Target Features
46519 @section Standard Target Features
46520 @cindex target descriptions, standard features
46522 A target description must contain either no registers or all the
46523 target's registers. If the description contains no registers, then
46524 @value{GDBN} will assume a default register layout, selected based on
46525 the architecture. If the description contains any registers, the
46526 default layout will not be used; the standard registers must be
46527 described in the target description, in such a way that @value{GDBN}
46528 can recognize them.
46530 This is accomplished by giving specific names to feature elements
46531 which contain standard registers. @value{GDBN} will look for features
46532 with those names and verify that they contain the expected registers;
46533 if any known feature is missing required registers, or if any required
46534 feature is missing, @value{GDBN} will reject the target
46535 description. You can add additional registers to any of the
46536 standard features --- @value{GDBN} will display them just as if
46537 they were added to an unrecognized feature.
46539 This section lists the known features and their expected contents.
46540 Sample XML documents for these features are included in the
46541 @value{GDBN} source tree, in the directory @file{gdb/features}.
46543 Names recognized by @value{GDBN} should include the name of the
46544 company or organization which selected the name, and the overall
46545 architecture to which the feature applies; so e.g.@: the feature
46546 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
46548 The names of registers are not case sensitive for the purpose
46549 of recognizing standard features, but @value{GDBN} will only display
46550 registers using the capitalization used in the description.
46553 * AArch64 Features::
46557 * MicroBlaze Features::
46561 * Nios II Features::
46562 * OpenRISC 1000 Features::
46563 * PowerPC Features::
46564 * RISC-V Features::
46566 * S/390 and System z Features::
46572 @node AArch64 Features
46573 @subsection AArch64 Features
46574 @cindex target descriptions, AArch64 features
46576 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
46577 targets. It should contain registers @samp{x0} through @samp{x30},
46578 @samp{sp}, @samp{pc}, and @samp{cpsr}.
46580 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
46581 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
46584 The @samp{org.gnu.gdb.aarch64.sve} feature is optional. If present,
46585 it should contain registers @samp{z0} through @samp{z31}, @samp{p0}
46586 through @samp{p15}, @samp{ffr} and @samp{vg}.
46588 The @samp{org.gnu.gdb.aarch64.pauth} feature is optional. If present,
46589 it should contain registers @samp{pauth_dmask} and @samp{pauth_cmask}.
46592 @subsection ARC Features
46593 @cindex target descriptions, ARC Features
46595 ARC processors are highly configurable, so even core registers and their number
46596 are not completely predetermined. In addition flags and PC registers which are
46597 important to @value{GDBN} are not ``core'' registers in ARC. It is required
46598 that one of the core registers features is present.
46599 @samp{org.gnu.gdb.arc.aux-minimal} feature is mandatory.
46601 The @samp{org.gnu.gdb.arc.core.v2} feature is required for ARC EM and ARC HS
46602 targets with a normal register file. It should contain registers @samp{r0}
46603 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
46604 @samp{lp_count} and @samp{pcl}. This feature may contain register @samp{ilink}
46605 and any of extension core registers @samp{r32} through @samp{r59/acch}.
46606 @samp{ilink} and extension core registers are not available to read/write, when
46607 debugging GNU/Linux applications, thus @samp{ilink} is made optional.
46609 The @samp{org.gnu.gdb.arc.core-reduced.v2} feature is required for ARC EM and
46610 ARC HS targets with a reduced register file. It should contain registers
46611 @samp{r0} through @samp{r3}, @samp{r10} through @samp{r15}, @samp{gp},
46612 @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink}, @samp{lp_count} and @samp{pcl}.
46613 This feature may contain register @samp{ilink} and any of extension core
46614 registers @samp{r32} through @samp{r59/acch}.
46616 The @samp{org.gnu.gdb.arc.core.arcompact} feature is required for ARCompact
46617 targets with a normal register file. It should contain registers @samp{r0}
46618 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
46619 @samp{lp_count} and @samp{pcl}. This feature may contain registers
46620 @samp{ilink1}, @samp{ilink2} and any of extension core registers @samp{r32}
46621 through @samp{r59/acch}. @samp{ilink1} and @samp{ilink2} and extension core
46622 registers are not available when debugging GNU/Linux applications. The only
46623 difference with @samp{org.gnu.gdb.arc.core.v2} feature is in the names of
46624 @samp{ilink1} and @samp{ilink2} registers and that @samp{r30} is mandatory in
46625 ARC v2, but @samp{ilink2} is optional on ARCompact.
46627 The @samp{org.gnu.gdb.arc.aux-minimal} feature is required for all ARC
46628 targets. It should contain registers @samp{pc} and @samp{status32}.
46631 @subsection ARM Features
46632 @cindex target descriptions, ARM features
46634 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
46636 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
46637 @samp{lr}, @samp{pc}, and @samp{cpsr}.
46639 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
46640 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
46641 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
46644 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
46645 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
46647 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
46648 it should contain at least registers @samp{wR0} through @samp{wR15} and
46649 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
46650 @samp{wCSSF}, and @samp{wCASF} registers are optional.
46652 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
46653 should contain at least registers @samp{d0} through @samp{d15}. If
46654 they are present, @samp{d16} through @samp{d31} should also be included.
46655 @value{GDBN} will synthesize the single-precision registers from
46656 halves of the double-precision registers.
46658 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
46659 need to contain registers; it instructs @value{GDBN} to display the
46660 VFP double-precision registers as vectors and to synthesize the
46661 quad-precision registers from pairs of double-precision registers.
46662 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
46663 be present and include 32 double-precision registers.
46665 @node i386 Features
46666 @subsection i386 Features
46667 @cindex target descriptions, i386 features
46669 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
46670 targets. It should describe the following registers:
46674 @samp{eax} through @samp{edi} plus @samp{eip} for i386
46676 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
46678 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
46679 @samp{fs}, @samp{gs}
46681 @samp{st0} through @samp{st7}
46683 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
46684 @samp{foseg}, @samp{fooff} and @samp{fop}
46687 The register sets may be different, depending on the target.
46689 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
46690 describe registers:
46694 @samp{xmm0} through @samp{xmm7} for i386
46696 @samp{xmm0} through @samp{xmm15} for amd64
46701 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
46702 @samp{org.gnu.gdb.i386.sse} feature. It should
46703 describe the upper 128 bits of @sc{ymm} registers:
46707 @samp{ymm0h} through @samp{ymm7h} for i386
46709 @samp{ymm0h} through @samp{ymm15h} for amd64
46712 The @samp{org.gnu.gdb.i386.mpx} is an optional feature representing Intel
46713 Memory Protection Extension (MPX). It should describe the following registers:
46717 @samp{bnd0raw} through @samp{bnd3raw} for i386 and amd64.
46719 @samp{bndcfgu} and @samp{bndstatus} for i386 and amd64.
46722 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
46723 describe a single register, @samp{orig_eax}.
46725 The @samp{org.gnu.gdb.i386.segments} feature is optional. It should
46726 describe two system registers: @samp{fs_base} and @samp{gs_base}.
46728 The @samp{org.gnu.gdb.i386.avx512} feature is optional and requires the
46729 @samp{org.gnu.gdb.i386.avx} feature. It should
46730 describe additional @sc{xmm} registers:
46734 @samp{xmm16h} through @samp{xmm31h}, only valid for amd64.
46737 It should describe the upper 128 bits of additional @sc{ymm} registers:
46741 @samp{ymm16h} through @samp{ymm31h}, only valid for amd64.
46745 describe the upper 256 bits of @sc{zmm} registers:
46749 @samp{zmm0h} through @samp{zmm7h} for i386.
46751 @samp{zmm0h} through @samp{zmm15h} for amd64.
46755 describe the additional @sc{zmm} registers:
46759 @samp{zmm16h} through @samp{zmm31h}, only valid for amd64.
46762 The @samp{org.gnu.gdb.i386.pkeys} feature is optional. It should
46763 describe a single register, @samp{pkru}. It is a 32-bit register
46764 valid for i386 and amd64.
46766 @node MicroBlaze Features
46767 @subsection MicroBlaze Features
46768 @cindex target descriptions, MicroBlaze features
46770 The @samp{org.gnu.gdb.microblaze.core} feature is required for MicroBlaze
46771 targets. It should contain registers @samp{r0} through @samp{r31},
46772 @samp{rpc}, @samp{rmsr}, @samp{rear}, @samp{resr}, @samp{rfsr}, @samp{rbtr},
46773 @samp{rpvr}, @samp{rpvr1} through @samp{rpvr11}, @samp{redr}, @samp{rpid},
46774 @samp{rzpr}, @samp{rtlbx}, @samp{rtlbsx}, @samp{rtlblo}, and @samp{rtlbhi}.
46776 The @samp{org.gnu.gdb.microblaze.stack-protect} feature is optional.
46777 If present, it should contain registers @samp{rshr} and @samp{rslr}
46779 @node MIPS Features
46780 @subsection @acronym{MIPS} Features
46781 @cindex target descriptions, @acronym{MIPS} features
46783 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
46784 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
46785 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
46788 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
46789 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
46790 registers. They may be 32-bit or 64-bit depending on the target.
46792 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
46793 it may be optional in a future version of @value{GDBN}. It should
46794 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
46795 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
46797 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
46798 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
46799 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
46800 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
46802 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
46803 contain a single register, @samp{restart}, which is used by the
46804 Linux kernel to control restartable syscalls.
46806 @node M68K Features
46807 @subsection M68K Features
46808 @cindex target descriptions, M68K features
46811 @item @samp{org.gnu.gdb.m68k.core}
46812 @itemx @samp{org.gnu.gdb.coldfire.core}
46813 @itemx @samp{org.gnu.gdb.fido.core}
46814 One of those features must be always present.
46815 The feature that is present determines which flavor of m68k is
46816 used. The feature that is present should contain registers
46817 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
46818 @samp{sp}, @samp{ps} and @samp{pc}.
46820 @item @samp{org.gnu.gdb.coldfire.fp}
46821 This feature is optional. If present, it should contain registers
46822 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
46826 @node NDS32 Features
46827 @subsection NDS32 Features
46828 @cindex target descriptions, NDS32 features
46830 The @samp{org.gnu.gdb.nds32.core} feature is required for NDS32
46831 targets. It should contain at least registers @samp{r0} through
46832 @samp{r10}, @samp{r15}, @samp{fp}, @samp{gp}, @samp{lp}, @samp{sp},
46835 The @samp{org.gnu.gdb.nds32.fpu} feature is optional. If present,
46836 it should contain 64-bit double-precision floating-point registers
46837 @samp{fd0} through @emph{fdN}, which should be @samp{fd3}, @samp{fd7},
46838 @samp{fd15}, or @samp{fd31} based on the FPU configuration implemented.
46840 @emph{Note:} The first sixteen 64-bit double-precision floating-point
46841 registers are overlapped with the thirty-two 32-bit single-precision
46842 floating-point registers. The 32-bit single-precision registers, if
46843 not being listed explicitly, will be synthesized from halves of the
46844 overlapping 64-bit double-precision registers. Listing 32-bit
46845 single-precision registers explicitly is deprecated, and the
46846 support to it could be totally removed some day.
46848 @node Nios II Features
46849 @subsection Nios II Features
46850 @cindex target descriptions, Nios II features
46852 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
46853 targets. It should contain the 32 core registers (@samp{zero},
46854 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
46855 @samp{pc}, and the 16 control registers (@samp{status} through
46858 @node OpenRISC 1000 Features
46859 @subsection Openrisc 1000 Features
46860 @cindex target descriptions, OpenRISC 1000 features
46862 The @samp{org.gnu.gdb.or1k.group0} feature is required for OpenRISC 1000
46863 targets. It should contain the 32 general purpose registers (@samp{r0}
46864 through @samp{r31}), @samp{ppc}, @samp{npc} and @samp{sr}.
46866 @node PowerPC Features
46867 @subsection PowerPC Features
46868 @cindex target descriptions, PowerPC features
46870 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
46871 targets. It should contain registers @samp{r0} through @samp{r31},
46872 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
46873 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
46875 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
46876 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
46878 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
46879 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr}, and
46880 @samp{vrsave}. @value{GDBN} will define pseudo-registers @samp{v0}
46881 through @samp{v31} as aliases for the corresponding @samp{vrX}
46884 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
46885 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN} will
46886 combine these registers with the floating point registers (@samp{f0}
46887 through @samp{f31}) and the altivec registers (@samp{vr0} through
46888 @samp{vr31}) to present the 128-bit wide registers @samp{vs0} through
46889 @samp{vs63}, the set of vector-scalar registers for POWER7.
46890 Therefore, this feature requires both @samp{org.gnu.gdb.power.fpu} and
46891 @samp{org.gnu.gdb.power.altivec}.
46893 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
46894 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
46895 @samp{spefscr}. SPE targets should provide 32-bit registers in
46896 @samp{org.gnu.gdb.power.core} and provide the upper halves in
46897 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
46898 these to present registers @samp{ev0} through @samp{ev31} to the
46901 The @samp{org.gnu.gdb.power.ppr} feature is optional. It should
46902 contain the 64-bit register @samp{ppr}.
46904 The @samp{org.gnu.gdb.power.dscr} feature is optional. It should
46905 contain the 64-bit register @samp{dscr}.
46907 The @samp{org.gnu.gdb.power.tar} feature is optional. It should
46908 contain the 64-bit register @samp{tar}.
46910 The @samp{org.gnu.gdb.power.ebb} feature is optional. It should
46911 contain registers @samp{bescr}, @samp{ebbhr} and @samp{ebbrr}, all
46914 The @samp{org.gnu.gdb.power.linux.pmu} feature is optional. It should
46915 contain registers @samp{mmcr0}, @samp{mmcr2}, @samp{siar}, @samp{sdar}
46916 and @samp{sier}, all 64-bit wide. This is the subset of the isa 2.07
46917 server PMU registers provided by @sc{gnu}/Linux.
46919 The @samp{org.gnu.gdb.power.htm.spr} feature is optional. It should
46920 contain registers @samp{tfhar}, @samp{texasr} and @samp{tfiar}, all
46923 The @samp{org.gnu.gdb.power.htm.core} feature is optional. It should
46924 contain the checkpointed general-purpose registers @samp{cr0} through
46925 @samp{cr31}, as well as the checkpointed registers @samp{clr} and
46926 @samp{cctr}. These registers may all be either 32-bit or 64-bit
46927 depending on the target. It should also contain the checkpointed
46928 registers @samp{ccr} and @samp{cxer}, which should both be 32-bit
46931 The @samp{org.gnu.gdb.power.htm.fpu} feature is optional. It should
46932 contain the checkpointed 64-bit floating-point registers @samp{cf0}
46933 through @samp{cf31}, as well as the checkpointed 64-bit register
46936 The @samp{org.gnu.gdb.power.htm.altivec} feature is optional. It
46937 should contain the checkpointed altivec registers @samp{cvr0} through
46938 @samp{cvr31}, all 128-bit wide. It should also contain the
46939 checkpointed registers @samp{cvscr} and @samp{cvrsave}, both 32-bit
46942 The @samp{org.gnu.gdb.power.htm.vsx} feature is optional. It should
46943 contain registers @samp{cvs0h} through @samp{cvs31h}. @value{GDBN}
46944 will combine these registers with the checkpointed floating point
46945 registers (@samp{cf0} through @samp{cf31}) and the checkpointed
46946 altivec registers (@samp{cvr0} through @samp{cvr31}) to present the
46947 128-bit wide checkpointed vector-scalar registers @samp{cvs0} through
46948 @samp{cvs63}. Therefore, this feature requires both
46949 @samp{org.gnu.gdb.power.htm.altivec} and
46950 @samp{org.gnu.gdb.power.htm.fpu}.
46952 The @samp{org.gnu.gdb.power.htm.ppr} feature is optional. It should
46953 contain the 64-bit checkpointed register @samp{cppr}.
46955 The @samp{org.gnu.gdb.power.htm.dscr} feature is optional. It should
46956 contain the 64-bit checkpointed register @samp{cdscr}.
46958 The @samp{org.gnu.gdb.power.htm.tar} feature is optional. It should
46959 contain the 64-bit checkpointed register @samp{ctar}.
46962 @node RISC-V Features
46963 @subsection RISC-V Features
46964 @cindex target descriptions, RISC-V Features
46966 The @samp{org.gnu.gdb.riscv.cpu} feature is required for RISC-V
46967 targets. It should contain the registers @samp{x0} through
46968 @samp{x31}, and @samp{pc}. Either the architectural names (@samp{x0},
46969 @samp{x1}, etc) can be used, or the ABI names (@samp{zero}, @samp{ra},
46972 The @samp{org.gnu.gdb.riscv.fpu} feature is optional. If present, it
46973 should contain registers @samp{f0} through @samp{f31}, @samp{fflags},
46974 @samp{frm}, and @samp{fcsr}. As with the cpu feature, either the
46975 architectural register names, or the ABI names can be used.
46977 The @samp{org.gnu.gdb.riscv.virtual} feature is optional. If present,
46978 it should contain registers that are not backed by real registers on
46979 the target, but are instead virtual, where the register value is
46980 derived from other target state. In many ways these are like
46981 @value{GDBN}s pseudo-registers, except implemented by the target.
46982 Currently the only register expected in this set is the one byte
46983 @samp{priv} register that contains the target's privilege level in the
46984 least significant two bits.
46986 The @samp{org.gnu.gdb.riscv.csr} feature is optional. If present, it
46987 should contain all of the target's standard CSRs. Standard CSRs are
46988 those defined in the RISC-V specification documents. There is some
46989 overlap between this feature and the fpu feature; the @samp{fflags},
46990 @samp{frm}, and @samp{fcsr} registers could be in either feature. The
46991 expectation is that these registers will be in the fpu feature if the
46992 target has floating point hardware, but can be moved into the csr
46993 feature if the target has the floating point control registers, but no
46994 other floating point hardware.
46997 @subsection RX Features
46998 @cindex target descriptions, RX Features
47000 The @samp{org.gnu.gdb.rx.core} feature is required for RX
47001 targets. It should contain the registers @samp{r0} through
47002 @samp{r15}, @samp{usp}, @samp{isp}, @samp{psw}, @samp{pc}, @samp{intb},
47003 @samp{bpsw}, @samp{bpc}, @samp{fintv}, @samp{fpsw}, and @samp{acc}.
47005 @node S/390 and System z Features
47006 @subsection S/390 and System z Features
47007 @cindex target descriptions, S/390 features
47008 @cindex target descriptions, System z features
47010 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
47011 System z targets. It should contain the PSW and the 16 general
47012 registers. In particular, System z targets should provide the 64-bit
47013 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
47014 S/390 targets should provide the 32-bit versions of these registers.
47015 A System z target that runs in 31-bit addressing mode should provide
47016 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
47017 register's upper halves @samp{r0h} through @samp{r15h}, and their
47018 lower halves @samp{r0l} through @samp{r15l}.
47020 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
47021 contain the 64-bit registers @samp{f0} through @samp{f15}, and
47024 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
47025 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
47027 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
47028 contain the register @samp{orig_r2}, which is 64-bit wide on System z
47029 targets and 32-bit otherwise. In addition, the feature may contain
47030 the @samp{last_break} register, whose width depends on the addressing
47031 mode, as well as the @samp{system_call} register, which is always
47034 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
47035 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
47036 @samp{atia}, and @samp{tr0} through @samp{tr15}.
47038 The @samp{org.gnu.gdb.s390.vx} feature is optional. It should contain
47039 64-bit wide registers @samp{v0l} through @samp{v15l}, which will be
47040 combined by @value{GDBN} with the floating point registers @samp{f0}
47041 through @samp{f15} to present the 128-bit wide vector registers
47042 @samp{v0} through @samp{v15}. In addition, this feature should
47043 contain the 128-bit wide vector registers @samp{v16} through
47046 The @samp{org.gnu.gdb.s390.gs} feature is optional. It should contain
47047 the 64-bit wide guarded-storage-control registers @samp{gsd},
47048 @samp{gssm}, and @samp{gsepla}.
47050 The @samp{org.gnu.gdb.s390.gsbc} feature is optional. It should contain
47051 the 64-bit wide guarded-storage broadcast control registers
47052 @samp{bc_gsd}, @samp{bc_gssm}, and @samp{bc_gsepla}.
47054 @node Sparc Features
47055 @subsection Sparc Features
47056 @cindex target descriptions, sparc32 features
47057 @cindex target descriptions, sparc64 features
47058 The @samp{org.gnu.gdb.sparc.cpu} feature is required for sparc32/sparc64
47059 targets. It should describe the following registers:
47063 @samp{g0} through @samp{g7}
47065 @samp{o0} through @samp{o7}
47067 @samp{l0} through @samp{l7}
47069 @samp{i0} through @samp{i7}
47072 They may be 32-bit or 64-bit depending on the target.
47074 Also the @samp{org.gnu.gdb.sparc.fpu} feature is required for sparc32/sparc64
47075 targets. It should describe the following registers:
47079 @samp{f0} through @samp{f31}
47081 @samp{f32} through @samp{f62} for sparc64
47084 The @samp{org.gnu.gdb.sparc.cp0} feature is required for sparc32/sparc64
47085 targets. It should describe the following registers:
47089 @samp{y}, @samp{psr}, @samp{wim}, @samp{tbr}, @samp{pc}, @samp{npc},
47090 @samp{fsr}, and @samp{csr} for sparc32
47092 @samp{pc}, @samp{npc}, @samp{state}, @samp{fsr}, @samp{fprs}, and @samp{y}
47096 @node TIC6x Features
47097 @subsection TMS320C6x Features
47098 @cindex target descriptions, TIC6x features
47099 @cindex target descriptions, TMS320C6x features
47100 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
47101 targets. It should contain registers @samp{A0} through @samp{A15},
47102 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
47104 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
47105 contain registers @samp{A16} through @samp{A31} and @samp{B16}
47106 through @samp{B31}.
47108 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
47109 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
47111 @node Operating System Information
47112 @appendix Operating System Information
47113 @cindex operating system information
47119 Users of @value{GDBN} often wish to obtain information about the state of
47120 the operating system running on the target---for example the list of
47121 processes, or the list of open files. This section describes the
47122 mechanism that makes it possible. This mechanism is similar to the
47123 target features mechanism (@pxref{Target Descriptions}), but focuses
47124 on a different aspect of target.
47126 Operating system information is retrieved from the target via the
47127 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
47128 read}). The object name in the request should be @samp{osdata}, and
47129 the @var{annex} identifies the data to be fetched.
47132 @appendixsection Process list
47133 @cindex operating system information, process list
47135 When requesting the process list, the @var{annex} field in the
47136 @samp{qXfer} request should be @samp{processes}. The returned data is
47137 an XML document. The formal syntax of this document is defined in
47138 @file{gdb/features/osdata.dtd}.
47140 An example document is:
47143 <?xml version="1.0"?>
47144 <!DOCTYPE target SYSTEM "osdata.dtd">
47145 <osdata type="processes">
47147 <column name="pid">1</column>
47148 <column name="user">root</column>
47149 <column name="command">/sbin/init</column>
47150 <column name="cores">1,2,3</column>
47155 Each item should include a column whose name is @samp{pid}. The value
47156 of that column should identify the process on the target. The
47157 @samp{user} and @samp{command} columns are optional, and will be
47158 displayed by @value{GDBN}. The @samp{cores} column, if present,
47159 should contain a comma-separated list of cores that this process
47160 is running on. Target may provide additional columns,
47161 which @value{GDBN} currently ignores.
47163 @node Trace File Format
47164 @appendix Trace File Format
47165 @cindex trace file format
47167 The trace file comes in three parts: a header, a textual description
47168 section, and a trace frame section with binary data.
47170 The header has the form @code{\x7fTRACE0\n}. The first byte is
47171 @code{0x7f} so as to indicate that the file contains binary data,
47172 while the @code{0} is a version number that may have different values
47175 The description section consists of multiple lines of @sc{ascii} text
47176 separated by newline characters (@code{0xa}). The lines may include a
47177 variety of optional descriptive or context-setting information, such
47178 as tracepoint definitions or register set size. @value{GDBN} will
47179 ignore any line that it does not recognize. An empty line marks the end
47184 Specifies the size of a register block in bytes. This is equal to the
47185 size of a @code{g} packet payload in the remote protocol. @var{size}
47186 is an ascii decimal number. There should be only one such line in
47187 a single trace file.
47189 @item status @var{status}
47190 Trace status. @var{status} has the same format as a @code{qTStatus}
47191 remote packet reply. There should be only one such line in a single trace
47194 @item tp @var{payload}
47195 Tracepoint definition. The @var{payload} has the same format as
47196 @code{qTfP}/@code{qTsP} remote packet reply payload. A single tracepoint
47197 may take multiple lines of definition, corresponding to the multiple
47200 @item tsv @var{payload}
47201 Trace state variable definition. The @var{payload} has the same format as
47202 @code{qTfV}/@code{qTsV} remote packet reply payload. A single variable
47203 may take multiple lines of definition, corresponding to the multiple
47206 @item tdesc @var{payload}
47207 Target description in XML format. The @var{payload} is a single line of
47208 the XML file. All such lines should be concatenated together to get
47209 the original XML file. This file is in the same format as @code{qXfer}
47210 @code{features} payload, and corresponds to the main @code{target.xml}
47211 file. Includes are not allowed.
47215 The trace frame section consists of a number of consecutive frames.
47216 Each frame begins with a two-byte tracepoint number, followed by a
47217 four-byte size giving the amount of data in the frame. The data in
47218 the frame consists of a number of blocks, each introduced by a
47219 character indicating its type (at least register, memory, and trace
47220 state variable). The data in this section is raw binary, not a
47221 hexadecimal or other encoding; its endianness matches the target's
47224 @c FIXME bi-arch may require endianness/arch info in description section
47227 @item R @var{bytes}
47228 Register block. The number and ordering of bytes matches that of a
47229 @code{g} packet in the remote protocol. Note that these are the
47230 actual bytes, in target order, not a hexadecimal encoding.
47232 @item M @var{address} @var{length} @var{bytes}...
47233 Memory block. This is a contiguous block of memory, at the 8-byte
47234 address @var{address}, with a 2-byte length @var{length}, followed by
47235 @var{length} bytes.
47237 @item V @var{number} @var{value}
47238 Trace state variable block. This records the 8-byte signed value
47239 @var{value} of trace state variable numbered @var{number}.
47243 Future enhancements of the trace file format may include additional types
47246 @node Index Section Format
47247 @appendix @code{.gdb_index} section format
47248 @cindex .gdb_index section format
47249 @cindex index section format
47251 This section documents the index section that is created by @code{save
47252 gdb-index} (@pxref{Index Files}). The index section is
47253 DWARF-specific; some knowledge of DWARF is assumed in this
47256 The mapped index file format is designed to be directly
47257 @code{mmap}able on any architecture. In most cases, a datum is
47258 represented using a little-endian 32-bit integer value, called an
47259 @code{offset_type}. Big endian machines must byte-swap the values
47260 before using them. Exceptions to this rule are noted. The data is
47261 laid out such that alignment is always respected.
47263 A mapped index consists of several areas, laid out in order.
47267 The file header. This is a sequence of values, of @code{offset_type}
47268 unless otherwise noted:
47272 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
47273 Version 4 uses a different hashing function from versions 5 and 6.
47274 Version 6 includes symbols for inlined functions, whereas versions 4
47275 and 5 do not. Version 7 adds attributes to the CU indices in the
47276 symbol table. Version 8 specifies that symbols from DWARF type units
47277 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
47278 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
47280 @value{GDBN} will only read version 4, 5, or 6 indices
47281 by specifying @code{set use-deprecated-index-sections on}.
47282 GDB has a workaround for potentially broken version 7 indices so it is
47283 currently not flagged as deprecated.
47286 The offset, from the start of the file, of the CU list.
47289 The offset, from the start of the file, of the types CU list. Note
47290 that this area can be empty, in which case this offset will be equal
47291 to the next offset.
47294 The offset, from the start of the file, of the address area.
47297 The offset, from the start of the file, of the symbol table.
47300 The offset, from the start of the file, of the constant pool.
47304 The CU list. This is a sequence of pairs of 64-bit little-endian
47305 values, sorted by the CU offset. The first element in each pair is
47306 the offset of a CU in the @code{.debug_info} section. The second
47307 element in each pair is the length of that CU. References to a CU
47308 elsewhere in the map are done using a CU index, which is just the
47309 0-based index into this table. Note that if there are type CUs, then
47310 conceptually CUs and type CUs form a single list for the purposes of
47314 The types CU list. This is a sequence of triplets of 64-bit
47315 little-endian values. In a triplet, the first value is the CU offset,
47316 the second value is the type offset in the CU, and the third value is
47317 the type signature. The types CU list is not sorted.
47320 The address area. The address area consists of a sequence of address
47321 entries. Each address entry has three elements:
47325 The low address. This is a 64-bit little-endian value.
47328 The high address. This is a 64-bit little-endian value. Like
47329 @code{DW_AT_high_pc}, the value is one byte beyond the end.
47332 The CU index. This is an @code{offset_type} value.
47336 The symbol table. This is an open-addressed hash table. The size of
47337 the hash table is always a power of 2.
47339 Each slot in the hash table consists of a pair of @code{offset_type}
47340 values. The first value is the offset of the symbol's name in the
47341 constant pool. The second value is the offset of the CU vector in the
47344 If both values are 0, then this slot in the hash table is empty. This
47345 is ok because while 0 is a valid constant pool index, it cannot be a
47346 valid index for both a string and a CU vector.
47348 The hash value for a table entry is computed by applying an
47349 iterative hash function to the symbol's name. Starting with an
47350 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
47351 the string is incorporated into the hash using the formula depending on the
47356 The formula is @code{r = r * 67 + c - 113}.
47358 @item Versions 5 to 7
47359 The formula is @code{r = r * 67 + tolower (c) - 113}.
47362 The terminating @samp{\0} is not incorporated into the hash.
47364 The step size used in the hash table is computed via
47365 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
47366 value, and @samp{size} is the size of the hash table. The step size
47367 is used to find the next candidate slot when handling a hash
47370 The names of C@t{++} symbols in the hash table are canonicalized. We
47371 don't currently have a simple description of the canonicalization
47372 algorithm; if you intend to create new index sections, you must read
47376 The constant pool. This is simply a bunch of bytes. It is organized
47377 so that alignment is correct: CU vectors are stored first, followed by
47380 A CU vector in the constant pool is a sequence of @code{offset_type}
47381 values. The first value is the number of CU indices in the vector.
47382 Each subsequent value is the index and symbol attributes of a CU in
47383 the CU list. This element in the hash table is used to indicate which
47384 CUs define the symbol and how the symbol is used.
47385 See below for the format of each CU index+attributes entry.
47387 A string in the constant pool is zero-terminated.
47390 Attributes were added to CU index values in @code{.gdb_index} version 7.
47391 If a symbol has multiple uses within a CU then there is one
47392 CU index+attributes value for each use.
47394 The format of each CU index+attributes entry is as follows
47400 This is the index of the CU in the CU list.
47402 These bits are reserved for future purposes and must be zero.
47404 The kind of the symbol in the CU.
47408 This value is reserved and should not be used.
47409 By reserving zero the full @code{offset_type} value is backwards compatible
47410 with previous versions of the index.
47412 The symbol is a type.
47414 The symbol is a variable or an enum value.
47416 The symbol is a function.
47418 Any other kind of symbol.
47420 These values are reserved.
47424 This bit is zero if the value is global and one if it is static.
47426 The determination of whether a symbol is global or static is complicated.
47427 The authorative reference is the file @file{dwarf2read.c} in
47428 @value{GDBN} sources.
47432 This pseudo-code describes the computation of a symbol's kind and
47433 global/static attributes in the index.
47436 is_external = get_attribute (die, DW_AT_external);
47437 language = get_attribute (cu_die, DW_AT_language);
47440 case DW_TAG_typedef:
47441 case DW_TAG_base_type:
47442 case DW_TAG_subrange_type:
47446 case DW_TAG_enumerator:
47448 is_static = language != CPLUS;
47450 case DW_TAG_subprogram:
47452 is_static = ! (is_external || language == ADA);
47454 case DW_TAG_constant:
47456 is_static = ! is_external;
47458 case DW_TAG_variable:
47460 is_static = ! is_external;
47462 case DW_TAG_namespace:
47466 case DW_TAG_class_type:
47467 case DW_TAG_interface_type:
47468 case DW_TAG_structure_type:
47469 case DW_TAG_union_type:
47470 case DW_TAG_enumeration_type:
47472 is_static = language != CPLUS;
47480 @appendix Manual pages
47484 * gdb man:: The GNU Debugger man page
47485 * gdbserver man:: Remote Server for the GNU Debugger man page
47486 * gcore man:: Generate a core file of a running program
47487 * gdbinit man:: gdbinit scripts
47488 * gdb-add-index man:: Add index files to speed up GDB
47494 @c man title gdb The GNU Debugger
47496 @c man begin SYNOPSIS gdb
47497 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
47498 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
47499 [@option{-b}@w{ }@var{bps}]
47500 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
47501 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
47502 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
47503 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
47504 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
47507 @c man begin DESCRIPTION gdb
47508 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
47509 going on ``inside'' another program while it executes -- or what another
47510 program was doing at the moment it crashed.
47512 @value{GDBN} can do four main kinds of things (plus other things in support of
47513 these) to help you catch bugs in the act:
47517 Start your program, specifying anything that might affect its behavior.
47520 Make your program stop on specified conditions.
47523 Examine what has happened, when your program has stopped.
47526 Change things in your program, so you can experiment with correcting the
47527 effects of one bug and go on to learn about another.
47530 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
47533 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
47534 commands from the terminal until you tell it to exit with the @value{GDBN}
47535 command @code{quit}. You can get online help from @value{GDBN} itself
47536 by using the command @code{help}.
47538 You can run @code{gdb} with no arguments or options; but the most
47539 usual way to start @value{GDBN} is with one argument or two, specifying an
47540 executable program as the argument:
47546 You can also start with both an executable program and a core file specified:
47552 You can, instead, specify a process ID as a second argument or use option
47553 @code{-p}, if you want to debug a running process:
47561 would attach @value{GDBN} to process @code{1234}. With option @option{-p} you
47562 can omit the @var{program} filename.
47564 Here are some of the most frequently needed @value{GDBN} commands:
47566 @c pod2man highlights the right hand side of the @item lines.
47568 @item break [@var{file}:]@var{function}
47569 Set a breakpoint at @var{function} (in @var{file}).
47571 @item run [@var{arglist}]
47572 Start your program (with @var{arglist}, if specified).
47575 Backtrace: display the program stack.
47577 @item print @var{expr}
47578 Display the value of an expression.
47581 Continue running your program (after stopping, e.g. at a breakpoint).
47584 Execute next program line (after stopping); step @emph{over} any
47585 function calls in the line.
47587 @item edit [@var{file}:]@var{function}
47588 look at the program line where it is presently stopped.
47590 @item list [@var{file}:]@var{function}
47591 type the text of the program in the vicinity of where it is presently stopped.
47594 Execute next program line (after stopping); step @emph{into} any
47595 function calls in the line.
47597 @item help [@var{name}]
47598 Show information about @value{GDBN} command @var{name}, or general information
47599 about using @value{GDBN}.
47602 Exit from @value{GDBN}.
47606 For full details on @value{GDBN},
47607 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
47608 by Richard M. Stallman and Roland H. Pesch. The same text is available online
47609 as the @code{gdb} entry in the @code{info} program.
47613 @c man begin OPTIONS gdb
47614 Any arguments other than options specify an executable
47615 file and core file (or process ID); that is, the first argument
47616 encountered with no
47617 associated option flag is equivalent to a @option{-se} option, and the second,
47618 if any, is equivalent to a @option{-c} option if it's the name of a file.
47620 both long and short forms; both are shown here. The long forms are also
47621 recognized if you truncate them, so long as enough of the option is
47622 present to be unambiguous. (If you prefer, you can flag option
47623 arguments with @option{+} rather than @option{-}, though we illustrate the
47624 more usual convention.)
47626 All the options and command line arguments you give are processed
47627 in sequential order. The order makes a difference when the @option{-x}
47633 List all options, with brief explanations.
47635 @item -symbols=@var{file}
47636 @itemx -s @var{file}
47637 Read symbol table from file @var{file}.
47640 Enable writing into executable and core files.
47642 @item -exec=@var{file}
47643 @itemx -e @var{file}
47644 Use file @var{file} as the executable file to execute when
47645 appropriate, and for examining pure data in conjunction with a core
47648 @item -se=@var{file}
47649 Read symbol table from file @var{file} and use it as the executable
47652 @item -core=@var{file}
47653 @itemx -c @var{file}
47654 Use file @var{file} as a core dump to examine.
47656 @item -command=@var{file}
47657 @itemx -x @var{file}
47658 Execute @value{GDBN} commands from file @var{file}.
47660 @item -ex @var{command}
47661 Execute given @value{GDBN} @var{command}.
47663 @item -directory=@var{directory}
47664 @itemx -d @var{directory}
47665 Add @var{directory} to the path to search for source files.
47668 Do not execute commands from @file{~/.gdbinit}.
47672 Do not execute commands from any @file{.gdbinit} initialization files.
47676 ``Quiet''. Do not print the introductory and copyright messages. These
47677 messages are also suppressed in batch mode.
47680 Run in batch mode. Exit with status @code{0} after processing all the command
47681 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
47682 Exit with nonzero status if an error occurs in executing the @value{GDBN}
47683 commands in the command files.
47685 Batch mode may be useful for running @value{GDBN} as a filter, for example to
47686 download and run a program on another computer; in order to make this
47687 more useful, the message
47690 Program exited normally.
47694 (which is ordinarily issued whenever a program running under @value{GDBN} control
47695 terminates) is not issued when running in batch mode.
47697 @item -cd=@var{directory}
47698 Run @value{GDBN} using @var{directory} as its working directory,
47699 instead of the current directory.
47703 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
47704 @value{GDBN} to output the full file name and line number in a standard,
47705 recognizable fashion each time a stack frame is displayed (which
47706 includes each time the program stops). This recognizable format looks
47707 like two @samp{\032} characters, followed by the file name, line number
47708 and character position separated by colons, and a newline. The
47709 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
47710 characters as a signal to display the source code for the frame.
47713 Set the line speed (baud rate or bits per second) of any serial
47714 interface used by @value{GDBN} for remote debugging.
47716 @item -tty=@var{device}
47717 Run using @var{device} for your program's standard input and output.
47721 @c man begin SEEALSO gdb
47723 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
47724 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
47725 documentation are properly installed at your site, the command
47732 should give you access to the complete manual.
47734 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
47735 Richard M. Stallman and Roland H. Pesch, July 1991.
47739 @node gdbserver man
47740 @heading gdbserver man
47742 @c man title gdbserver Remote Server for the GNU Debugger
47744 @c man begin SYNOPSIS gdbserver
47745 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
47747 gdbserver --attach @var{comm} @var{pid}
47749 gdbserver --multi @var{comm}
47753 @c man begin DESCRIPTION gdbserver
47754 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
47755 than the one which is running the program being debugged.
47758 @subheading Usage (server (target) side)
47761 Usage (server (target) side):
47764 First, you need to have a copy of the program you want to debug put onto
47765 the target system. The program can be stripped to save space if needed, as
47766 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
47767 the @value{GDBN} running on the host system.
47769 To use the server, you log on to the target system, and run the @command{gdbserver}
47770 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
47771 your program, and (c) its arguments. The general syntax is:
47774 target> gdbserver @var{comm} @var{program} [@var{args} ...]
47777 For example, using a serial port, you might say:
47781 @c @file would wrap it as F</dev/com1>.
47782 target> gdbserver /dev/com1 emacs foo.txt
47785 target> gdbserver @file{/dev/com1} emacs foo.txt
47789 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
47790 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
47791 waits patiently for the host @value{GDBN} to communicate with it.
47793 To use a TCP connection, you could say:
47796 target> gdbserver host:2345 emacs foo.txt
47799 This says pretty much the same thing as the last example, except that we are
47800 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
47801 that we are expecting to see a TCP connection from @code{host} to local TCP port
47802 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
47803 want for the port number as long as it does not conflict with any existing TCP
47804 ports on the target system. This same port number must be used in the host
47805 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
47806 you chose a port number that conflicts with another service, @command{gdbserver} will
47807 print an error message and exit.
47809 @command{gdbserver} can also attach to running programs.
47810 This is accomplished via the @option{--attach} argument. The syntax is:
47813 target> gdbserver --attach @var{comm} @var{pid}
47816 @var{pid} is the process ID of a currently running process. It isn't
47817 necessary to point @command{gdbserver} at a binary for the running process.
47819 To start @code{gdbserver} without supplying an initial command to run
47820 or process ID to attach, use the @option{--multi} command line option.
47821 In such case you should connect using @kbd{target extended-remote} to start
47822 the program you want to debug.
47825 target> gdbserver --multi @var{comm}
47829 @subheading Usage (host side)
47835 You need an unstripped copy of the target program on your host system, since
47836 @value{GDBN} needs to examine its symbol tables and such. Start up @value{GDBN} as you normally
47837 would, with the target program as the first argument. (You may need to use the
47838 @option{--baud} option if the serial line is running at anything except 9600 baud.)
47839 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
47840 new command you need to know about is @code{target remote}
47841 (or @code{target extended-remote}). Its argument is either
47842 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
47843 descriptor. For example:
47847 @c @file would wrap it as F</dev/ttyb>.
47848 (@value{GDBP}) target remote /dev/ttyb
47851 (@value{GDBP}) target remote @file{/dev/ttyb}
47856 communicates with the server via serial line @file{/dev/ttyb}, and:
47859 (@value{GDBP}) target remote the-target:2345
47863 communicates via a TCP connection to port 2345 on host `the-target', where
47864 you previously started up @command{gdbserver} with the same port number. Note that for
47865 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
47866 command, otherwise you may get an error that looks something like
47867 `Connection refused'.
47869 @command{gdbserver} can also debug multiple inferiors at once,
47872 the @value{GDBN} manual in node @code{Inferiors and Programs}
47873 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
47876 @ref{Inferiors and Programs}.
47878 In such case use the @code{extended-remote} @value{GDBN} command variant:
47881 (@value{GDBP}) target extended-remote the-target:2345
47884 The @command{gdbserver} option @option{--multi} may or may not be used in such
47888 @c man begin OPTIONS gdbserver
47889 There are three different modes for invoking @command{gdbserver}:
47894 Debug a specific program specified by its program name:
47897 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
47900 The @var{comm} parameter specifies how should the server communicate
47901 with @value{GDBN}; it is either a device name (to use a serial line),
47902 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
47903 stdin/stdout of @code{gdbserver}. Specify the name of the program to
47904 debug in @var{prog}. Any remaining arguments will be passed to the
47905 program verbatim. When the program exits, @value{GDBN} will close the
47906 connection, and @code{gdbserver} will exit.
47909 Debug a specific program by specifying the process ID of a running
47913 gdbserver --attach @var{comm} @var{pid}
47916 The @var{comm} parameter is as described above. Supply the process ID
47917 of a running program in @var{pid}; @value{GDBN} will do everything
47918 else. Like with the previous mode, when the process @var{pid} exits,
47919 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
47922 Multi-process mode -- debug more than one program/process:
47925 gdbserver --multi @var{comm}
47928 In this mode, @value{GDBN} can instruct @command{gdbserver} which
47929 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
47930 close the connection when a process being debugged exits, so you can
47931 debug several processes in the same session.
47934 In each of the modes you may specify these options:
47939 List all options, with brief explanations.
47942 This option causes @command{gdbserver} to print its version number and exit.
47945 @command{gdbserver} will attach to a running program. The syntax is:
47948 target> gdbserver --attach @var{comm} @var{pid}
47951 @var{pid} is the process ID of a currently running process. It isn't
47952 necessary to point @command{gdbserver} at a binary for the running process.
47955 To start @code{gdbserver} without supplying an initial command to run
47956 or process ID to attach, use this command line option.
47957 Then you can connect using @kbd{target extended-remote} and start
47958 the program you want to debug. The syntax is:
47961 target> gdbserver --multi @var{comm}
47965 Instruct @code{gdbserver} to display extra status information about the debugging
47967 This option is intended for @code{gdbserver} development and for bug reports to
47970 @item --remote-debug
47971 Instruct @code{gdbserver} to display remote protocol debug output.
47972 This option is intended for @code{gdbserver} development and for bug reports to
47975 @item --debug-file=@var{filename}
47976 Instruct @code{gdbserver} to send any debug output to the given @var{filename}.
47977 This option is intended for @code{gdbserver} development and for bug reports to
47980 @item --debug-format=option1@r{[},option2,...@r{]}
47981 Instruct @code{gdbserver} to include extra information in each line
47982 of debugging output.
47983 @xref{Other Command-Line Arguments for gdbserver}.
47986 Specify a wrapper to launch programs
47987 for debugging. The option should be followed by the name of the
47988 wrapper, then any command-line arguments to pass to the wrapper, then
47989 @kbd{--} indicating the end of the wrapper arguments.
47992 By default, @command{gdbserver} keeps the listening TCP port open, so that
47993 additional connections are possible. However, if you start @code{gdbserver}
47994 with the @option{--once} option, it will stop listening for any further
47995 connection attempts after connecting to the first @value{GDBN} session.
47997 @c --disable-packet is not documented for users.
47999 @c --disable-randomization and --no-disable-randomization are superseded by
48000 @c QDisableRandomization.
48005 @c man begin SEEALSO gdbserver
48007 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
48008 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
48009 documentation are properly installed at your site, the command
48015 should give you access to the complete manual.
48017 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
48018 Richard M. Stallman and Roland H. Pesch, July 1991.
48025 @c man title gcore Generate a core file of a running program
48028 @c man begin SYNOPSIS gcore
48029 gcore [-a] [-o @var{prefix}] @var{pid1} [@var{pid2}...@var{pidN}]
48033 @c man begin DESCRIPTION gcore
48034 Generate core dumps of one or more running programs with process IDs
48035 @var{pid1}, @var{pid2}, etc. A core file produced by @command{gcore}
48036 is equivalent to one produced by the kernel when the process crashes
48037 (and when @kbd{ulimit -c} was used to set up an appropriate core dump
48038 limit). However, unlike after a crash, after @command{gcore} finishes
48039 its job the program remains running without any change.
48042 @c man begin OPTIONS gcore
48045 Dump all memory mappings. The actual effect of this option depends on
48046 the Operating System. On @sc{gnu}/Linux, it will disable
48047 @code{use-coredump-filter} (@pxref{set use-coredump-filter}) and
48048 enable @code{dump-excluded-mappings} (@pxref{set
48049 dump-excluded-mappings}).
48051 @item -o @var{prefix}
48052 The optional argument @var{prefix} specifies the prefix to be used
48053 when composing the file names of the core dumps. The file name is
48054 composed as @file{@var{prefix}.@var{pid}}, where @var{pid} is the
48055 process ID of the running program being analyzed by @command{gcore}.
48056 If not specified, @var{prefix} defaults to @var{gcore}.
48060 @c man begin SEEALSO gcore
48062 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
48063 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
48064 documentation are properly installed at your site, the command
48071 should give you access to the complete manual.
48073 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
48074 Richard M. Stallman and Roland H. Pesch, July 1991.
48081 @c man title gdbinit GDB initialization scripts
48084 @c man begin SYNOPSIS gdbinit
48085 @ifset SYSTEM_GDBINIT
48086 @value{SYSTEM_GDBINIT}
48089 @ifset SYSTEM_GDBINIT_DIR
48090 @value{SYSTEM_GDBINIT_DIR}/*
48099 @c man begin DESCRIPTION gdbinit
48100 These files contain @value{GDBN} commands to automatically execute during
48101 @value{GDBN} startup. The lines of contents are canned sequences of commands,
48104 the @value{GDBN} manual in node @code{Sequences}
48105 -- shell command @code{info -f gdb -n Sequences}.
48111 Please read more in
48113 the @value{GDBN} manual in node @code{Startup}
48114 -- shell command @code{info -f gdb -n Startup}.
48121 @ifset SYSTEM_GDBINIT
48122 @item @value{SYSTEM_GDBINIT}
48124 @ifclear SYSTEM_GDBINIT
48125 @item (not enabled with @code{--with-system-gdbinit} during compilation)
48127 System-wide initialization file. It is executed unless user specified
48128 @value{GDBN} option @code{-nx} or @code{-n}.
48131 the @value{GDBN} manual in node @code{System-wide configuration}
48132 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
48134 @ifset SYSTEM_GDBINIT_DIR
48135 @item @value{SYSTEM_GDBINIT_DIR}
48137 @ifclear SYSTEM_GDBINIT_DIR
48138 @item (not enabled with @code{--with-system-gdbinit-dir} during compilation)
48140 System-wide initialization directory. All files in this directory are
48141 executed on startup unless user specified @value{GDBN} option @code{-nx} or
48142 @code{-n}, as long as they have a recognized file extension.
48145 the @value{GDBN} manual in node @code{System-wide configuration}
48146 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
48149 @ref{System-wide configuration}.
48153 User initialization file. It is executed unless user specified
48154 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
48157 Initialization file for current directory. It may need to be enabled with
48158 @value{GDBN} security command @code{set auto-load local-gdbinit}.
48161 the @value{GDBN} manual in node @code{Init File in the Current Directory}
48162 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
48165 @ref{Init File in the Current Directory}.
48170 @c man begin SEEALSO gdbinit
48172 gdb(1), @code{info -f gdb -n Startup}
48174 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
48175 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
48176 documentation are properly installed at your site, the command
48182 should give you access to the complete manual.
48184 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
48185 Richard M. Stallman and Roland H. Pesch, July 1991.
48189 @node gdb-add-index man
48190 @heading gdb-add-index
48191 @pindex gdb-add-index
48192 @anchor{gdb-add-index}
48194 @c man title gdb-add-index Add index files to speed up GDB
48196 @c man begin SYNOPSIS gdb-add-index
48197 gdb-add-index @var{filename}
48200 @c man begin DESCRIPTION gdb-add-index
48201 When @value{GDBN} finds a symbol file, it scans the symbols in the
48202 file in order to construct an internal symbol table. This lets most
48203 @value{GDBN} operations work quickly--at the cost of a delay early on.
48204 For large programs, this delay can be quite lengthy, so @value{GDBN}
48205 provides a way to build an index, which speeds up startup.
48207 To determine whether a file contains such an index, use the command
48208 @kbd{readelf -S filename}: the index is stored in a section named
48209 @code{.gdb_index}. The index file can only be produced on systems
48210 which use ELF binaries and DWARF debug information (i.e., sections
48211 named @code{.debug_*}).
48213 @command{gdb-add-index} uses @value{GDBN} and @command{objdump} found
48214 in the @env{PATH} environment variable. If you want to use different
48215 versions of these programs, you can specify them through the
48216 @env{GDB} and @env{OBJDUMP} environment variables.
48220 the @value{GDBN} manual in node @code{Index Files}
48221 -- shell command @kbd{info -f gdb -n "Index Files"}.
48228 @c man begin SEEALSO gdb-add-index
48230 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
48231 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
48232 documentation are properly installed at your site, the command
48238 should give you access to the complete manual.
48240 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
48241 Richard M. Stallman and Roland H. Pesch, July 1991.
48247 @node GNU Free Documentation License
48248 @appendix GNU Free Documentation License
48251 @node Concept Index
48252 @unnumbered Concept Index
48256 @node Command and Variable Index
48257 @unnumbered Command, Variable, and Function Index
48262 % I think something like @@colophon should be in texinfo. In the
48264 \long\def\colophon{\hbox to0pt{}\vfill
48265 \centerline{The body of this manual is set in}
48266 \centerline{\fontname\tenrm,}
48267 \centerline{with headings in {\bf\fontname\tenbf}}
48268 \centerline{and examples in {\tt\fontname\tentt}.}
48269 \centerline{{\it\fontname\tenit\/},}
48270 \centerline{{\bf\fontname\tenbf}, and}
48271 \centerline{{\sl\fontname\tensl\/}}
48272 \centerline{are used for emphasis.}\vfill}
48274 % Blame: doc@@cygnus.com, 1991.