1 \input texinfo @c -*-texinfo-*-
2 @c Copyright (C) 1988-2019 Free Software Foundation, Inc.
5 @c makeinfo ignores cmds prev to setfilename, so its arg cannot make use
6 @c of @set vars. However, you can override filename with makeinfo -o.
13 @settitle Debugging with @value{GDBN}
14 @setchapternewpage odd
23 @c To avoid file-name clashes between index.html and Index.html, when
24 @c the manual is produced on a Posix host and then moved to a
25 @c case-insensitive filesystem (e.g., MS-Windows), we separate the
26 @c indices into two: Concept Index and all the rest.
30 @c readline appendices use @vindex, @findex and @ftable,
31 @c annotate.texi and gdbmi use @findex.
34 @c !!set GDB manual's edition---not the same as GDB version!
35 @c This is updated by GNU Press.
38 @c !!set GDB edit command default editor
41 @c THIS MANUAL REQUIRES TEXINFO 4.0 OR LATER.
43 @c This is a dir.info fragment to support semi-automated addition of
44 @c manuals to an info tree.
45 @dircategory Software development
47 * Gdb: (gdb). The GNU debugger.
48 * gdbserver: (gdb) Server. The GNU debugging server.
52 @c man begin COPYRIGHT
53 Copyright @copyright{} 1988-2019 Free Software Foundation, Inc.
55 Permission is granted to copy, distribute and/or modify this document
56 under the terms of the GNU Free Documentation License, Version 1.3 or
57 any later version published by the Free Software Foundation; with the
58 Invariant Sections being ``Free Software'' and ``Free Software Needs
59 Free Documentation'', with the Front-Cover Texts being ``A GNU Manual,''
60 and with the Back-Cover Texts as in (a) below.
62 (a) The FSF's Back-Cover Text is: ``You are free to copy and modify
63 this GNU Manual. Buying copies from GNU Press supports the FSF in
64 developing GNU and promoting software freedom.''
69 This file documents the @sc{gnu} debugger @value{GDBN}.
71 This is the @value{EDITION} Edition, of @cite{Debugging with
72 @value{GDBN}: the @sc{gnu} Source-Level Debugger} for @value{GDBN}
73 @ifset VERSION_PACKAGE
74 @value{VERSION_PACKAGE}
76 Version @value{GDBVN}.
82 @title Debugging with @value{GDBN}
83 @subtitle The @sc{gnu} Source-Level Debugger
85 @subtitle @value{EDITION} Edition, for @value{GDBN} version @value{GDBVN}
86 @ifset VERSION_PACKAGE
88 @subtitle @value{VERSION_PACKAGE}
90 @author Richard Stallman, Roland Pesch, Stan Shebs, et al.
94 \hfill (Send bugs and comments on @value{GDBN} to @value{BUGURL}.)\par
95 \hfill {\it Debugging with @value{GDBN}}\par
96 \hfill \TeX{}info \texinfoversion\par
100 @vskip 0pt plus 1filll
101 Published by the Free Software Foundation @*
102 51 Franklin Street, Fifth Floor,
103 Boston, MA 02110-1301, USA@*
104 ISBN 978-0-9831592-3-0 @*
111 @node Top, Summary, (dir), (dir)
113 @top Debugging with @value{GDBN}
115 This file describes @value{GDBN}, the @sc{gnu} symbolic debugger.
117 This is the @value{EDITION} Edition, for @value{GDBN}
118 @ifset VERSION_PACKAGE
119 @value{VERSION_PACKAGE}
121 Version @value{GDBVN}.
123 Copyright (C) 1988-2019 Free Software Foundation, Inc.
125 This edition of the GDB manual is dedicated to the memory of Fred
126 Fish. Fred was a long-standing contributor to GDB and to Free
127 software in general. We will miss him.
130 * Summary:: Summary of @value{GDBN}
131 * Sample Session:: A sample @value{GDBN} session
133 * Invocation:: Getting in and out of @value{GDBN}
134 * Commands:: @value{GDBN} commands
135 * Running:: Running programs under @value{GDBN}
136 * Stopping:: Stopping and continuing
137 * Reverse Execution:: Running programs backward
138 * Process Record and Replay:: Recording inferior's execution and replaying it
139 * Stack:: Examining the stack
140 * Source:: Examining source files
141 * Data:: Examining data
142 * Optimized Code:: Debugging optimized code
143 * Macros:: Preprocessor Macros
144 * Tracepoints:: Debugging remote targets non-intrusively
145 * Overlays:: Debugging programs that use overlays
147 * Languages:: Using @value{GDBN} with different languages
149 * Symbols:: Examining the symbol table
150 * Altering:: Altering execution
151 * GDB Files:: @value{GDBN} files
152 * Targets:: Specifying a debugging target
153 * Remote Debugging:: Debugging remote programs
154 * Configurations:: Configuration-specific information
155 * Controlling GDB:: Controlling @value{GDBN}
156 * Extending GDB:: Extending @value{GDBN}
157 * Interpreters:: Command Interpreters
158 * TUI:: @value{GDBN} Text User Interface
159 * Emacs:: Using @value{GDBN} under @sc{gnu} Emacs
160 * GDB/MI:: @value{GDBN}'s Machine Interface.
161 * Annotations:: @value{GDBN}'s annotation interface.
162 * JIT Interface:: Using the JIT debugging interface.
163 * In-Process Agent:: In-Process Agent
165 * GDB Bugs:: Reporting bugs in @value{GDBN}
167 @ifset SYSTEM_READLINE
168 * Command Line Editing: (rluserman). Command Line Editing
169 * Using History Interactively: (history). Using History Interactively
171 @ifclear SYSTEM_READLINE
172 * Command Line Editing:: Command Line Editing
173 * Using History Interactively:: Using History Interactively
175 * In Memoriam:: In Memoriam
176 * Formatting Documentation:: How to format and print @value{GDBN} documentation
177 * Installing GDB:: Installing GDB
178 * Maintenance Commands:: Maintenance Commands
179 * Remote Protocol:: GDB Remote Serial Protocol
180 * Agent Expressions:: The GDB Agent Expression Mechanism
181 * Target Descriptions:: How targets can describe themselves to
183 * Operating System Information:: Getting additional information from
185 * Trace File Format:: GDB trace file format
186 * Index Section Format:: .gdb_index section format
187 * Man Pages:: Manual pages
188 * Copying:: GNU General Public License says
189 how you can copy and share GDB
190 * GNU Free Documentation License:: The license for this documentation
191 * Concept Index:: Index of @value{GDBN} concepts
192 * Command and Variable Index:: Index of @value{GDBN} commands, variables,
193 functions, and Python data types
201 @unnumbered Summary of @value{GDBN}
203 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
204 going on ``inside'' another program while it executes---or what another
205 program was doing at the moment it crashed.
207 @value{GDBN} can do four main kinds of things (plus other things in support of
208 these) to help you catch bugs in the act:
212 Start your program, specifying anything that might affect its behavior.
215 Make your program stop on specified conditions.
218 Examine what has happened, when your program has stopped.
221 Change things in your program, so you can experiment with correcting the
222 effects of one bug and go on to learn about another.
225 You can use @value{GDBN} to debug programs written in C and C@t{++}.
226 For more information, see @ref{Supported Languages,,Supported Languages}.
227 For more information, see @ref{C,,C and C++}.
229 Support for D is partial. For information on D, see
233 Support for Modula-2 is partial. For information on Modula-2, see
234 @ref{Modula-2,,Modula-2}.
236 Support for OpenCL C is partial. For information on OpenCL C, see
237 @ref{OpenCL C,,OpenCL C}.
240 Debugging Pascal programs which use sets, subranges, file variables, or
241 nested functions does not currently work. @value{GDBN} does not support
242 entering expressions, printing values, or similar features using Pascal
246 @value{GDBN} can be used to debug programs written in Fortran, although
247 it may be necessary to refer to some variables with a trailing
250 @value{GDBN} can be used to debug programs written in Objective-C,
251 using either the Apple/NeXT or the GNU Objective-C runtime.
254 * Free Software:: Freely redistributable software
255 * Free Documentation:: Free Software Needs Free Documentation
256 * Contributors:: Contributors to GDB
260 @unnumberedsec Free Software
262 @value{GDBN} is @dfn{free software}, protected by the @sc{gnu}
263 General Public License
264 (GPL). The GPL gives you the freedom to copy or adapt a licensed
265 program---but every person getting a copy also gets with it the
266 freedom to modify that copy (which means that they must get access to
267 the source code), and the freedom to distribute further copies.
268 Typical software companies use copyrights to limit your freedoms; the
269 Free Software Foundation uses the GPL to preserve these freedoms.
271 Fundamentally, the General Public License is a license which says that
272 you have these freedoms and that you cannot take these freedoms away
275 @node Free Documentation
276 @unnumberedsec Free Software Needs Free Documentation
278 The biggest deficiency in the free software community today is not in
279 the software---it is the lack of good free documentation that we can
280 include with the free software. Many of our most important
281 programs do not come with free reference manuals and free introductory
282 texts. Documentation is an essential part of any software package;
283 when an important free software package does not come with a free
284 manual and a free tutorial, that is a major gap. We have many such
287 Consider Perl, for instance. The tutorial manuals that people
288 normally use are non-free. How did this come about? Because the
289 authors of those manuals published them with restrictive terms---no
290 copying, no modification, source files not available---which exclude
291 them from the free software world.
293 That wasn't the first time this sort of thing happened, and it was far
294 from the last. Many times we have heard a GNU user eagerly describe a
295 manual that he is writing, his intended contribution to the community,
296 only to learn that he had ruined everything by signing a publication
297 contract to make it non-free.
299 Free documentation, like free software, is a matter of freedom, not
300 price. The problem with the non-free manual is not that publishers
301 charge a price for printed copies---that in itself is fine. (The Free
302 Software Foundation sells printed copies of manuals, too.) The
303 problem is the restrictions on the use of the manual. Free manuals
304 are available in source code form, and give you permission to copy and
305 modify. Non-free manuals do not allow this.
307 The criteria of freedom for a free manual are roughly the same as for
308 free software. Redistribution (including the normal kinds of
309 commercial redistribution) must be permitted, so that the manual can
310 accompany every copy of the program, both on-line and on paper.
312 Permission for modification of the technical content is crucial too.
313 When people modify the software, adding or changing features, if they
314 are conscientious they will change the manual too---so they can
315 provide accurate and clear documentation for the modified program. A
316 manual that leaves you no choice but to write a new manual to document
317 a changed version of the program is not really available to our
320 Some kinds of limits on the way modification is handled are
321 acceptable. For example, requirements to preserve the original
322 author's copyright notice, the distribution terms, or the list of
323 authors, are ok. It is also no problem to require modified versions
324 to include notice that they were modified. Even entire sections that
325 may not be deleted or changed are acceptable, as long as they deal
326 with nontechnical topics (like this one). These kinds of restrictions
327 are acceptable because they don't obstruct the community's normal use
330 However, it must be possible to modify all the @emph{technical}
331 content of the manual, and then distribute the result in all the usual
332 media, through all the usual channels. Otherwise, the restrictions
333 obstruct the use of the manual, it is not free, and we need another
334 manual to replace it.
336 Please spread the word about this issue. Our community continues to
337 lose manuals to proprietary publishing. If we spread the word that
338 free software needs free reference manuals and free tutorials, perhaps
339 the next person who wants to contribute by writing documentation will
340 realize, before it is too late, that only free manuals contribute to
341 the free software community.
343 If you are writing documentation, please insist on publishing it under
344 the GNU Free Documentation License or another free documentation
345 license. Remember that this decision requires your approval---you
346 don't have to let the publisher decide. Some commercial publishers
347 will use a free license if you insist, but they will not propose the
348 option; it is up to you to raise the issue and say firmly that this is
349 what you want. If the publisher you are dealing with refuses, please
350 try other publishers. If you're not sure whether a proposed license
351 is free, write to @email{licensing@@gnu.org}.
353 You can encourage commercial publishers to sell more free, copylefted
354 manuals and tutorials by buying them, and particularly by buying
355 copies from the publishers that paid for their writing or for major
356 improvements. Meanwhile, try to avoid buying non-free documentation
357 at all. Check the distribution terms of a manual before you buy it,
358 and insist that whoever seeks your business must respect your freedom.
359 Check the history of the book, and try to reward the publishers that
360 have paid or pay the authors to work on it.
362 The Free Software Foundation maintains a list of free documentation
363 published by other publishers, at
364 @url{http://www.fsf.org/doc/other-free-books.html}.
367 @unnumberedsec Contributors to @value{GDBN}
369 Richard Stallman was the original author of @value{GDBN}, and of many
370 other @sc{gnu} programs. Many others have contributed to its
371 development. This section attempts to credit major contributors. One
372 of the virtues of free software is that everyone is free to contribute
373 to it; with regret, we cannot actually acknowledge everyone here. The
374 file @file{ChangeLog} in the @value{GDBN} distribution approximates a
375 blow-by-blow account.
377 Changes much prior to version 2.0 are lost in the mists of time.
380 @emph{Plea:} Additions to this section are particularly welcome. If you
381 or your friends (or enemies, to be evenhanded) have been unfairly
382 omitted from this list, we would like to add your names!
385 So that they may not regard their many labors as thankless, we
386 particularly thank those who shepherded @value{GDBN} through major
388 Andrew Cagney (releases 6.3, 6.2, 6.1, 6.0, 5.3, 5.2, 5.1 and 5.0);
389 Jim Blandy (release 4.18);
390 Jason Molenda (release 4.17);
391 Stan Shebs (release 4.14);
392 Fred Fish (releases 4.16, 4.15, 4.13, 4.12, 4.11, 4.10, and 4.9);
393 Stu Grossman and John Gilmore (releases 4.8, 4.7, 4.6, 4.5, and 4.4);
394 John Gilmore (releases 4.3, 4.2, 4.1, 4.0, and 3.9);
395 Jim Kingdon (releases 3.5, 3.4, and 3.3);
396 and Randy Smith (releases 3.2, 3.1, and 3.0).
398 Richard Stallman, assisted at various times by Peter TerMaat, Chris
399 Hanson, and Richard Mlynarik, handled releases through 2.8.
401 Michael Tiemann is the author of most of the @sc{gnu} C@t{++} support
402 in @value{GDBN}, with significant additional contributions from Per
403 Bothner and Daniel Berlin. James Clark wrote the @sc{gnu} C@t{++}
404 demangler. Early work on C@t{++} was by Peter TerMaat (who also did
405 much general update work leading to release 3.0).
407 @value{GDBN} uses the BFD subroutine library to examine multiple
408 object-file formats; BFD was a joint project of David V.
409 Henkel-Wallace, Rich Pixley, Steve Chamberlain, and John Gilmore.
411 David Johnson wrote the original COFF support; Pace Willison did
412 the original support for encapsulated COFF.
414 Brent Benson of Harris Computer Systems contributed DWARF 2 support.
416 Adam de Boor and Bradley Davis contributed the ISI Optimum V support.
417 Per Bothner, Noboyuki Hikichi, and Alessandro Forin contributed MIPS
419 Jean-Daniel Fekete contributed Sun 386i support.
420 Chris Hanson improved the HP9000 support.
421 Noboyuki Hikichi and Tomoyuki Hasei contributed Sony/News OS 3 support.
422 David Johnson contributed Encore Umax support.
423 Jyrki Kuoppala contributed Altos 3068 support.
424 Jeff Law contributed HP PA and SOM support.
425 Keith Packard contributed NS32K support.
426 Doug Rabson contributed Acorn Risc Machine support.
427 Bob Rusk contributed Harris Nighthawk CX-UX support.
428 Chris Smith contributed Convex support (and Fortran debugging).
429 Jonathan Stone contributed Pyramid support.
430 Michael Tiemann contributed SPARC support.
431 Tim Tucker contributed support for the Gould NP1 and Gould Powernode.
432 Pace Willison contributed Intel 386 support.
433 Jay Vosburgh contributed Symmetry support.
434 Marko Mlinar contributed OpenRISC 1000 support.
436 Andreas Schwab contributed M68K @sc{gnu}/Linux support.
438 Rich Schaefer and Peter Schauer helped with support of SunOS shared
441 Jay Fenlason and Roland McGrath ensured that @value{GDBN} and GAS agree
442 about several machine instruction sets.
444 Patrick Duval, Ted Goldstein, Vikram Koka and Glenn Engel helped develop
445 remote debugging. Intel Corporation, Wind River Systems, AMD, and ARM
446 contributed remote debugging modules for the i960, VxWorks, A29K UDI,
447 and RDI targets, respectively.
449 Brian Fox is the author of the readline libraries providing
450 command-line editing and command history.
452 Andrew Beers of SUNY Buffalo wrote the language-switching code, the
453 Modula-2 support, and contributed the Languages chapter of this manual.
455 Fred Fish wrote most of the support for Unix System Vr4.
456 He also enhanced the command-completion support to cover C@t{++} overloaded
459 Hitachi America (now Renesas America), Ltd. sponsored the support for
460 H8/300, H8/500, and Super-H processors.
462 NEC sponsored the support for the v850, Vr4xxx, and Vr5xxx processors.
464 Mitsubishi (now Renesas) sponsored the support for D10V, D30V, and M32R/D
467 Toshiba sponsored the support for the TX39 Mips processor.
469 Matsushita sponsored the support for the MN10200 and MN10300 processors.
471 Fujitsu sponsored the support for SPARClite and FR30 processors.
473 Kung Hsu, Jeff Law, and Rick Sladkey added support for hardware
476 Michael Snyder added support for tracepoints.
478 Stu Grossman wrote gdbserver.
480 Jim Kingdon, Peter Schauer, Ian Taylor, and Stu Grossman made
481 nearly innumerable bug fixes and cleanups throughout @value{GDBN}.
483 The following people at the Hewlett-Packard Company contributed
484 support for the PA-RISC 2.0 architecture, HP-UX 10.20, 10.30, and 11.0
485 (narrow mode), HP's implementation of kernel threads, HP's aC@t{++}
486 compiler, and the Text User Interface (nee Terminal User Interface):
487 Ben Krepp, Richard Title, John Bishop, Susan Macchia, Kathy Mann,
488 Satish Pai, India Paul, Steve Rehrauer, and Elena Zannoni. Kim Haase
489 provided HP-specific information in this manual.
491 DJ Delorie ported @value{GDBN} to MS-DOS, for the DJGPP project.
492 Robert Hoehne made significant contributions to the DJGPP port.
494 Cygnus Solutions has sponsored @value{GDBN} maintenance and much of its
495 development since 1991. Cygnus engineers who have worked on @value{GDBN}
496 fulltime include Mark Alexander, Jim Blandy, Per Bothner, Kevin
497 Buettner, Edith Epstein, Chris Faylor, Fred Fish, Martin Hunt, Jim
498 Ingham, John Gilmore, Stu Grossman, Kung Hsu, Jim Kingdon, John Metzler,
499 Fernando Nasser, Geoffrey Noer, Dawn Perchik, Rich Pixley, Zdenek
500 Radouch, Keith Seitz, Stan Shebs, David Taylor, and Elena Zannoni. In
501 addition, Dave Brolley, Ian Carmichael, Steve Chamberlain, Nick Clifton,
502 JT Conklin, Stan Cox, DJ Delorie, Ulrich Drepper, Frank Eigler, Doug
503 Evans, Sean Fagan, David Henkel-Wallace, Richard Henderson, Jeff
504 Holcomb, Jeff Law, Jim Lemke, Tom Lord, Bob Manson, Michael Meissner,
505 Jason Merrill, Catherine Moore, Drew Moseley, Ken Raeburn, Gavin
506 Romig-Koch, Rob Savoye, Jamie Smith, Mike Stump, Ian Taylor, Angela
507 Thomas, Michael Tiemann, Tom Tromey, Ron Unrau, Jim Wilson, and David
508 Zuhn have made contributions both large and small.
510 Andrew Cagney, Fernando Nasser, and Elena Zannoni, while working for
511 Cygnus Solutions, implemented the original @sc{gdb/mi} interface.
513 Jim Blandy added support for preprocessor macros, while working for Red
516 Andrew Cagney designed @value{GDBN}'s architecture vector. Many
517 people including Andrew Cagney, Stephane Carrez, Randolph Chung, Nick
518 Duffek, Richard Henderson, Mark Kettenis, Grace Sainsbury, Kei
519 Sakamoto, Yoshinori Sato, Michael Snyder, Andreas Schwab, Jason
520 Thorpe, Corinna Vinschen, Ulrich Weigand, and Elena Zannoni, helped
521 with the migration of old architectures to this new framework.
523 Andrew Cagney completely re-designed and re-implemented @value{GDBN}'s
524 unwinder framework, this consisting of a fresh new design featuring
525 frame IDs, independent frame sniffers, and the sentinel frame. Mark
526 Kettenis implemented the @sc{dwarf 2} unwinder, Jeff Johnston the
527 libunwind unwinder, and Andrew Cagney the dummy, sentinel, tramp, and
528 trad unwinders. The architecture-specific changes, each involving a
529 complete rewrite of the architecture's frame code, were carried out by
530 Jim Blandy, Joel Brobecker, Kevin Buettner, Andrew Cagney, Stephane
531 Carrez, Randolph Chung, Orjan Friberg, Richard Henderson, Daniel
532 Jacobowitz, Jeff Johnston, Mark Kettenis, Theodore A. Roth, Kei
533 Sakamoto, Yoshinori Sato, Michael Snyder, Corinna Vinschen, and Ulrich
536 Christian Zankel, Ross Morley, Bob Wilson, and Maxim Grigoriev from
537 Tensilica, Inc.@: contributed support for Xtensa processors. Others
538 who have worked on the Xtensa port of @value{GDBN} in the past include
539 Steve Tjiang, John Newlin, and Scott Foehner.
541 Michael Eager and staff of Xilinx, Inc., contributed support for the
542 Xilinx MicroBlaze architecture.
544 Initial support for the FreeBSD/mips target and native configuration
545 was developed by SRI International and the University of Cambridge
546 Computer Laboratory under DARPA/AFRL contract FA8750-10-C-0237
547 ("CTSRD"), as part of the DARPA CRASH research programme.
549 Initial support for the FreeBSD/riscv target and native configuration
550 was developed by SRI International and the University of Cambridge
551 Computer Laboratory (Department of Computer Science and Technology)
552 under DARPA contract HR0011-18-C-0016 ("ECATS"), as part of the DARPA
553 SSITH research programme.
555 The original port to the OpenRISC 1000 is believed to be due to
556 Alessandro Forin and Per Bothner. More recent ports have been the work
557 of Jeremy Bennett, Franck Jullien, Stefan Wallentowitz and
561 @chapter A Sample @value{GDBN} Session
563 You can use this manual at your leisure to read all about @value{GDBN}.
564 However, a handful of commands are enough to get started using the
565 debugger. This chapter illustrates those commands.
568 In this sample session, we emphasize user input like this: @b{input},
569 to make it easier to pick out from the surrounding output.
572 @c FIXME: this example may not be appropriate for some configs, where
573 @c FIXME...primary interest is in remote use.
575 One of the preliminary versions of @sc{gnu} @code{m4} (a generic macro
576 processor) exhibits the following bug: sometimes, when we change its
577 quote strings from the default, the commands used to capture one macro
578 definition within another stop working. In the following short @code{m4}
579 session, we define a macro @code{foo} which expands to @code{0000}; we
580 then use the @code{m4} built-in @code{defn} to define @code{bar} as the
581 same thing. However, when we change the open quote string to
582 @code{<QUOTE>} and the close quote string to @code{<UNQUOTE>}, the same
583 procedure fails to define a new synonym @code{baz}:
592 @b{define(bar,defn(`foo'))}
596 @b{changequote(<QUOTE>,<UNQUOTE>)}
598 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
601 m4: End of input: 0: fatal error: EOF in string
605 Let us use @value{GDBN} to try to see what is going on.
608 $ @b{@value{GDBP} m4}
609 @c FIXME: this falsifies the exact text played out, to permit smallbook
610 @c FIXME... format to come out better.
611 @value{GDBN} is free software and you are welcome to distribute copies
612 of it under certain conditions; type "show copying" to see
614 There is absolutely no warranty for @value{GDBN}; type "show warranty"
617 @value{GDBN} @value{GDBVN}, Copyright 1999 Free Software Foundation, Inc...
622 @value{GDBN} reads only enough symbol data to know where to find the
623 rest when needed; as a result, the first prompt comes up very quickly.
624 We now tell @value{GDBN} to use a narrower display width than usual, so
625 that examples fit in this manual.
628 (@value{GDBP}) @b{set width 70}
632 We need to see how the @code{m4} built-in @code{changequote} works.
633 Having looked at the source, we know the relevant subroutine is
634 @code{m4_changequote}, so we set a breakpoint there with the @value{GDBN}
635 @code{break} command.
638 (@value{GDBP}) @b{break m4_changequote}
639 Breakpoint 1 at 0x62f4: file builtin.c, line 879.
643 Using the @code{run} command, we start @code{m4} running under @value{GDBN}
644 control; as long as control does not reach the @code{m4_changequote}
645 subroutine, the program runs as usual:
648 (@value{GDBP}) @b{run}
649 Starting program: /work/Editorial/gdb/gnu/m4/m4
657 To trigger the breakpoint, we call @code{changequote}. @value{GDBN}
658 suspends execution of @code{m4}, displaying information about the
659 context where it stops.
662 @b{changequote(<QUOTE>,<UNQUOTE>)}
664 Breakpoint 1, m4_changequote (argc=3, argv=0x33c70)
666 879 if (bad_argc(TOKEN_DATA_TEXT(argv[0]),argc,1,3))
670 Now we use the command @code{n} (@code{next}) to advance execution to
671 the next line of the current function.
675 882 set_quotes((argc >= 2) ? TOKEN_DATA_TEXT(argv[1])\
680 @code{set_quotes} looks like a promising subroutine. We can go into it
681 by using the command @code{s} (@code{step}) instead of @code{next}.
682 @code{step} goes to the next line to be executed in @emph{any}
683 subroutine, so it steps into @code{set_quotes}.
687 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
689 530 if (lquote != def_lquote)
693 The display that shows the subroutine where @code{m4} is now
694 suspended (and its arguments) is called a stack frame display. It
695 shows a summary of the stack. We can use the @code{backtrace}
696 command (which can also be spelled @code{bt}), to see where we are
697 in the stack as a whole: the @code{backtrace} command displays a
698 stack frame for each active subroutine.
701 (@value{GDBP}) @b{bt}
702 #0 set_quotes (lq=0x34c78 "<QUOTE>", rq=0x34c88 "<UNQUOTE>")
704 #1 0x6344 in m4_changequote (argc=3, argv=0x33c70)
706 #2 0x8174 in expand_macro (sym=0x33320) at macro.c:242
707 #3 0x7a88 in expand_token (obs=0x0, t=209696, td=0xf7fffa30)
709 #4 0x79dc in expand_input () at macro.c:40
710 #5 0x2930 in main (argc=0, argv=0xf7fffb20) at m4.c:195
714 We step through a few more lines to see what happens. The first two
715 times, we can use @samp{s}; the next two times we use @code{n} to avoid
716 falling into the @code{xstrdup} subroutine.
720 0x3b5c 532 if (rquote != def_rquote)
722 0x3b80 535 lquote = (lq == nil || *lq == '\0') ? \
723 def_lquote : xstrdup(lq);
725 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
728 538 len_lquote = strlen(rquote);
732 The last line displayed looks a little odd; we can examine the variables
733 @code{lquote} and @code{rquote} to see if they are in fact the new left
734 and right quotes we specified. We use the command @code{p}
735 (@code{print}) to see their values.
738 (@value{GDBP}) @b{p lquote}
739 $1 = 0x35d40 "<QUOTE>"
740 (@value{GDBP}) @b{p rquote}
741 $2 = 0x35d50 "<UNQUOTE>"
745 @code{lquote} and @code{rquote} are indeed the new left and right quotes.
746 To look at some context, we can display ten lines of source
747 surrounding the current line with the @code{l} (@code{list}) command.
753 535 lquote = (lq == nil || *lq == '\0') ? def_lquote\
755 536 rquote = (rq == nil || *rq == '\0') ? def_rquote\
758 538 len_lquote = strlen(rquote);
759 539 len_rquote = strlen(lquote);
766 Let us step past the two lines that set @code{len_lquote} and
767 @code{len_rquote}, and then examine the values of those variables.
771 539 len_rquote = strlen(lquote);
774 (@value{GDBP}) @b{p len_lquote}
776 (@value{GDBP}) @b{p len_rquote}
781 That certainly looks wrong, assuming @code{len_lquote} and
782 @code{len_rquote} are meant to be the lengths of @code{lquote} and
783 @code{rquote} respectively. We can set them to better values using
784 the @code{p} command, since it can print the value of
785 any expression---and that expression can include subroutine calls and
789 (@value{GDBP}) @b{p len_lquote=strlen(lquote)}
791 (@value{GDBP}) @b{p len_rquote=strlen(rquote)}
796 Is that enough to fix the problem of using the new quotes with the
797 @code{m4} built-in @code{defn}? We can allow @code{m4} to continue
798 executing with the @code{c} (@code{continue}) command, and then try the
799 example that caused trouble initially:
805 @b{define(baz,defn(<QUOTE>foo<UNQUOTE>))}
812 Success! The new quotes now work just as well as the default ones. The
813 problem seems to have been just the two typos defining the wrong
814 lengths. We allow @code{m4} exit by giving it an EOF as input:
818 Program exited normally.
822 The message @samp{Program exited normally.} is from @value{GDBN}; it
823 indicates @code{m4} has finished executing. We can end our @value{GDBN}
824 session with the @value{GDBN} @code{quit} command.
827 (@value{GDBP}) @b{quit}
831 @chapter Getting In and Out of @value{GDBN}
833 This chapter discusses how to start @value{GDBN}, and how to get out of it.
837 type @samp{@value{GDBP}} to start @value{GDBN}.
839 type @kbd{quit} or @kbd{Ctrl-d} to exit.
843 * Invoking GDB:: How to start @value{GDBN}
844 * Quitting GDB:: How to quit @value{GDBN}
845 * Shell Commands:: How to use shell commands inside @value{GDBN}
846 * Logging Output:: How to log @value{GDBN}'s output to a file
850 @section Invoking @value{GDBN}
852 Invoke @value{GDBN} by running the program @code{@value{GDBP}}. Once started,
853 @value{GDBN} reads commands from the terminal until you tell it to exit.
855 You can also run @code{@value{GDBP}} with a variety of arguments and options,
856 to specify more of your debugging environment at the outset.
858 The command-line options described here are designed
859 to cover a variety of situations; in some environments, some of these
860 options may effectively be unavailable.
862 The most usual way to start @value{GDBN} is with one argument,
863 specifying an executable program:
866 @value{GDBP} @var{program}
870 You can also start with both an executable program and a core file
874 @value{GDBP} @var{program} @var{core}
877 You can, instead, specify a process ID as a second argument or use option
878 @code{-p}, if you want to debug a running process:
881 @value{GDBP} @var{program} 1234
886 would attach @value{GDBN} to process @code{1234}. With option @option{-p} you
887 can omit the @var{program} filename.
889 Taking advantage of the second command-line argument requires a fairly
890 complete operating system; when you use @value{GDBN} as a remote
891 debugger attached to a bare board, there may not be any notion of
892 ``process'', and there is often no way to get a core dump. @value{GDBN}
893 will warn you if it is unable to attach or to read core dumps.
895 You can optionally have @code{@value{GDBP}} pass any arguments after the
896 executable file to the inferior using @code{--args}. This option stops
899 @value{GDBP} --args gcc -O2 -c foo.c
901 This will cause @code{@value{GDBP}} to debug @code{gcc}, and to set
902 @code{gcc}'s command-line arguments (@pxref{Arguments}) to @samp{-O2 -c foo.c}.
904 You can run @code{@value{GDBP}} without printing the front material, which describes
905 @value{GDBN}'s non-warranty, by specifying @code{--silent}
906 (or @code{-q}/@code{--quiet}):
909 @value{GDBP} --silent
913 You can further control how @value{GDBN} starts up by using command-line
914 options. @value{GDBN} itself can remind you of the options available.
924 to display all available options and briefly describe their use
925 (@samp{@value{GDBP} -h} is a shorter equivalent).
927 All options and command line arguments you give are processed
928 in sequential order. The order makes a difference when the
929 @samp{-x} option is used.
933 * File Options:: Choosing files
934 * Mode Options:: Choosing modes
935 * Startup:: What @value{GDBN} does during startup
939 @subsection Choosing Files
941 When @value{GDBN} starts, it reads any arguments other than options as
942 specifying an executable file and core file (or process ID). This is
943 the same as if the arguments were specified by the @samp{-se} and
944 @samp{-c} (or @samp{-p}) options respectively. (@value{GDBN} reads the
945 first argument that does not have an associated option flag as
946 equivalent to the @samp{-se} option followed by that argument; and the
947 second argument that does not have an associated option flag, if any, as
948 equivalent to the @samp{-c}/@samp{-p} option followed by that argument.)
949 If the second argument begins with a decimal digit, @value{GDBN} will
950 first attempt to attach to it as a process, and if that fails, attempt
951 to open it as a corefile. If you have a corefile whose name begins with
952 a digit, you can prevent @value{GDBN} from treating it as a pid by
953 prefixing it with @file{./}, e.g.@: @file{./12345}.
955 If @value{GDBN} has not been configured to included core file support,
956 such as for most embedded targets, then it will complain about a second
957 argument and ignore it.
959 Many options have both long and short forms; both are shown in the
960 following list. @value{GDBN} also recognizes the long forms if you truncate
961 them, so long as enough of the option is present to be unambiguous.
962 (If you prefer, you can flag option arguments with @samp{--} rather
963 than @samp{-}, though we illustrate the more usual convention.)
965 @c NOTE: the @cindex entries here use double dashes ON PURPOSE. This
966 @c way, both those who look for -foo and --foo in the index, will find
970 @item -symbols @var{file}
972 @cindex @code{--symbols}
974 Read symbol table from file @var{file}.
976 @item -exec @var{file}
978 @cindex @code{--exec}
980 Use file @var{file} as the executable file to execute when appropriate,
981 and for examining pure data in conjunction with a core dump.
985 Read symbol table from file @var{file} and use it as the executable
988 @item -core @var{file}
990 @cindex @code{--core}
992 Use file @var{file} as a core dump to examine.
994 @item -pid @var{number}
995 @itemx -p @var{number}
998 Connect to process ID @var{number}, as with the @code{attach} command.
1000 @item -command @var{file}
1001 @itemx -x @var{file}
1002 @cindex @code{--command}
1004 Execute commands from file @var{file}. The contents of this file is
1005 evaluated exactly as the @code{source} command would.
1006 @xref{Command Files,, Command files}.
1008 @item -eval-command @var{command}
1009 @itemx -ex @var{command}
1010 @cindex @code{--eval-command}
1012 Execute a single @value{GDBN} command.
1014 This option may be used multiple times to call multiple commands. It may
1015 also be interleaved with @samp{-command} as required.
1018 @value{GDBP} -ex 'target sim' -ex 'load' \
1019 -x setbreakpoints -ex 'run' a.out
1022 @item -init-command @var{file}
1023 @itemx -ix @var{file}
1024 @cindex @code{--init-command}
1026 Execute commands from file @var{file} before loading the inferior (but
1027 after loading gdbinit files).
1030 @item -init-eval-command @var{command}
1031 @itemx -iex @var{command}
1032 @cindex @code{--init-eval-command}
1034 Execute a single @value{GDBN} command before loading the inferior (but
1035 after loading gdbinit files).
1038 @item -directory @var{directory}
1039 @itemx -d @var{directory}
1040 @cindex @code{--directory}
1042 Add @var{directory} to the path to search for source and script files.
1046 @cindex @code{--readnow}
1048 Read each symbol file's entire symbol table immediately, rather than
1049 the default, which is to read it incrementally as it is needed.
1050 This makes startup slower, but makes future operations faster.
1053 @anchor{--readnever}
1054 @cindex @code{--readnever}, command-line option
1055 Do not read each symbol file's symbolic debug information. This makes
1056 startup faster but at the expense of not being able to perform
1057 symbolic debugging. DWARF unwind information is also not read,
1058 meaning backtraces may become incomplete or inaccurate. One use of
1059 this is when a user simply wants to do the following sequence: attach,
1060 dump core, detach. Loading the debugging information in this case is
1061 an unnecessary cause of delay.
1065 @subsection Choosing Modes
1067 You can run @value{GDBN} in various alternative modes---for example, in
1068 batch mode or quiet mode.
1076 Do not execute commands found in any initialization file.
1077 There are three init files, loaded in the following order:
1080 @item @file{system.gdbinit}
1081 This is the system-wide init file.
1082 Its location is specified with the @code{--with-system-gdbinit}
1083 configure option (@pxref{System-wide configuration}).
1084 It is loaded first when @value{GDBN} starts, before command line options
1085 have been processed.
1086 @item @file{~/.gdbinit}
1087 This is the init file in your home directory.
1088 It is loaded next, after @file{system.gdbinit}, and before
1089 command options have been processed.
1090 @item @file{./.gdbinit}
1091 This is the init file in the current directory.
1092 It is loaded last, after command line options other than @code{-x} and
1093 @code{-ex} have been processed. Command line options @code{-x} and
1094 @code{-ex} are processed last, after @file{./.gdbinit} has been loaded.
1097 For further documentation on startup processing, @xref{Startup}.
1098 For documentation on how to write command files,
1099 @xref{Command Files,,Command Files}.
1104 Do not execute commands found in @file{~/.gdbinit}, the init file
1105 in your home directory.
1111 @cindex @code{--quiet}
1112 @cindex @code{--silent}
1114 ``Quiet''. Do not print the introductory and copyright messages. These
1115 messages are also suppressed in batch mode.
1118 @cindex @code{--batch}
1119 Run in batch mode. Exit with status @code{0} after processing all the
1120 command files specified with @samp{-x} (and all commands from
1121 initialization files, if not inhibited with @samp{-n}). Exit with
1122 nonzero status if an error occurs in executing the @value{GDBN} commands
1123 in the command files. Batch mode also disables pagination, sets unlimited
1124 terminal width and height @pxref{Screen Size}, and acts as if @kbd{set confirm
1125 off} were in effect (@pxref{Messages/Warnings}).
1127 Batch mode may be useful for running @value{GDBN} as a filter, for
1128 example to download and run a program on another computer; in order to
1129 make this more useful, the message
1132 Program exited normally.
1136 (which is ordinarily issued whenever a program running under
1137 @value{GDBN} control terminates) is not issued when running in batch
1141 @cindex @code{--batch-silent}
1142 Run in batch mode exactly like @samp{-batch}, but totally silently. All
1143 @value{GDBN} output to @code{stdout} is prevented (@code{stderr} is
1144 unaffected). This is much quieter than @samp{-silent} and would be useless
1145 for an interactive session.
1147 This is particularly useful when using targets that give @samp{Loading section}
1148 messages, for example.
1150 Note that targets that give their output via @value{GDBN}, as opposed to
1151 writing directly to @code{stdout}, will also be made silent.
1153 @item -return-child-result
1154 @cindex @code{--return-child-result}
1155 The return code from @value{GDBN} will be the return code from the child
1156 process (the process being debugged), with the following exceptions:
1160 @value{GDBN} exits abnormally. E.g., due to an incorrect argument or an
1161 internal error. In this case the exit code is the same as it would have been
1162 without @samp{-return-child-result}.
1164 The user quits with an explicit value. E.g., @samp{quit 1}.
1166 The child process never runs, or is not allowed to terminate, in which case
1167 the exit code will be -1.
1170 This option is useful in conjunction with @samp{-batch} or @samp{-batch-silent},
1171 when @value{GDBN} is being used as a remote program loader or simulator
1176 @cindex @code{--nowindows}
1178 ``No windows''. If @value{GDBN} comes with a graphical user interface
1179 (GUI) built in, then this option tells @value{GDBN} to only use the command-line
1180 interface. If no GUI is available, this option has no effect.
1184 @cindex @code{--windows}
1186 If @value{GDBN} includes a GUI, then this option requires it to be
1189 @item -cd @var{directory}
1191 Run @value{GDBN} using @var{directory} as its working directory,
1192 instead of the current directory.
1194 @item -data-directory @var{directory}
1195 @itemx -D @var{directory}
1196 @cindex @code{--data-directory}
1198 Run @value{GDBN} using @var{directory} as its data directory.
1199 The data directory is where @value{GDBN} searches for its
1200 auxiliary files. @xref{Data Files}.
1204 @cindex @code{--fullname}
1206 @sc{gnu} Emacs sets this option when it runs @value{GDBN} as a
1207 subprocess. It tells @value{GDBN} to output the full file name and line
1208 number in a standard, recognizable fashion each time a stack frame is
1209 displayed (which includes each time your program stops). This
1210 recognizable format looks like two @samp{\032} characters, followed by
1211 the file name, line number and character position separated by colons,
1212 and a newline. The Emacs-to-@value{GDBN} interface program uses the two
1213 @samp{\032} characters as a signal to display the source code for the
1216 @item -annotate @var{level}
1217 @cindex @code{--annotate}
1218 This option sets the @dfn{annotation level} inside @value{GDBN}. Its
1219 effect is identical to using @samp{set annotate @var{level}}
1220 (@pxref{Annotations}). The annotation @var{level} controls how much
1221 information @value{GDBN} prints together with its prompt, values of
1222 expressions, source lines, and other types of output. Level 0 is the
1223 normal, level 1 is for use when @value{GDBN} is run as a subprocess of
1224 @sc{gnu} Emacs, level 3 is the maximum annotation suitable for programs
1225 that control @value{GDBN}, and level 2 has been deprecated.
1227 The annotation mechanism has largely been superseded by @sc{gdb/mi}
1231 @cindex @code{--args}
1232 Change interpretation of command line so that arguments following the
1233 executable file are passed as command line arguments to the inferior.
1234 This option stops option processing.
1236 @item -baud @var{bps}
1238 @cindex @code{--baud}
1240 Set the line speed (baud rate or bits per second) of any serial
1241 interface used by @value{GDBN} for remote debugging.
1243 @item -l @var{timeout}
1245 Set the timeout (in seconds) of any communication used by @value{GDBN}
1246 for remote debugging.
1248 @item -tty @var{device}
1249 @itemx -t @var{device}
1250 @cindex @code{--tty}
1252 Run using @var{device} for your program's standard input and output.
1253 @c FIXME: kingdon thinks there is more to -tty. Investigate.
1255 @c resolve the situation of these eventually
1257 @cindex @code{--tui}
1258 Activate the @dfn{Text User Interface} when starting. The Text User
1259 Interface manages several text windows on the terminal, showing
1260 source, assembly, registers and @value{GDBN} command outputs
1261 (@pxref{TUI, ,@value{GDBN} Text User Interface}). Do not use this
1262 option if you run @value{GDBN} from Emacs (@pxref{Emacs, ,
1263 Using @value{GDBN} under @sc{gnu} Emacs}).
1265 @item -interpreter @var{interp}
1266 @cindex @code{--interpreter}
1267 Use the interpreter @var{interp} for interface with the controlling
1268 program or device. This option is meant to be set by programs which
1269 communicate with @value{GDBN} using it as a back end.
1270 @xref{Interpreters, , Command Interpreters}.
1272 @samp{--interpreter=mi} (or @samp{--interpreter=mi3}) causes
1273 @value{GDBN} to use the @dfn{@sc{gdb/mi} interface} version 3 (@pxref{GDB/MI, ,
1274 The @sc{gdb/mi} Interface}) included since @value{GDBN} version 9.1. @sc{gdb/mi}
1275 version 2 (@code{mi2}), included in @value{GDBN} 6.0 and version 1 (@code{mi1}),
1276 included in @value{GDBN} 5.3, are also available. Earlier @sc{gdb/mi}
1277 interfaces are no longer supported.
1280 @cindex @code{--write}
1281 Open the executable and core files for both reading and writing. This
1282 is equivalent to the @samp{set write on} command inside @value{GDBN}
1286 @cindex @code{--statistics}
1287 This option causes @value{GDBN} to print statistics about time and
1288 memory usage after it completes each command and returns to the prompt.
1291 @cindex @code{--version}
1292 This option causes @value{GDBN} to print its version number and
1293 no-warranty blurb, and exit.
1295 @item -configuration
1296 @cindex @code{--configuration}
1297 This option causes @value{GDBN} to print details about its build-time
1298 configuration parameters, and then exit. These details can be
1299 important when reporting @value{GDBN} bugs (@pxref{GDB Bugs}).
1304 @subsection What @value{GDBN} Does During Startup
1305 @cindex @value{GDBN} startup
1307 Here's the description of what @value{GDBN} does during session startup:
1311 Sets up the command interpreter as specified by the command line
1312 (@pxref{Mode Options, interpreter}).
1316 Reads the system-wide @dfn{init file} (if @option{--with-system-gdbinit} was
1317 used when building @value{GDBN}; @pxref{System-wide configuration,
1318 ,System-wide configuration and settings}) and executes all the commands in
1321 @anchor{Home Directory Init File}
1323 Reads the init file (if any) in your home directory@footnote{On
1324 DOS/Windows systems, the home directory is the one pointed to by the
1325 @code{HOME} environment variable.} and executes all the commands in
1328 @anchor{Option -init-eval-command}
1330 Executes commands and command files specified by the @samp{-iex} and
1331 @samp{-ix} options in their specified order. Usually you should use the
1332 @samp{-ex} and @samp{-x} options instead, but this way you can apply
1333 settings before @value{GDBN} init files get executed and before inferior
1337 Processes command line options and operands.
1339 @anchor{Init File in the Current Directory during Startup}
1341 Reads and executes the commands from init file (if any) in the current
1342 working directory as long as @samp{set auto-load local-gdbinit} is set to
1343 @samp{on} (@pxref{Init File in the Current Directory}).
1344 This is only done if the current directory is
1345 different from your home directory. Thus, you can have more than one
1346 init file, one generic in your home directory, and another, specific
1347 to the program you are debugging, in the directory where you invoke
1351 If the command line specified a program to debug, or a process to
1352 attach to, or a core file, @value{GDBN} loads any auto-loaded
1353 scripts provided for the program or for its loaded shared libraries.
1354 @xref{Auto-loading}.
1356 If you wish to disable the auto-loading during startup,
1357 you must do something like the following:
1360 $ gdb -iex "set auto-load python-scripts off" myprogram
1363 Option @samp{-ex} does not work because the auto-loading is then turned
1367 Executes commands and command files specified by the @samp{-ex} and
1368 @samp{-x} options in their specified order. @xref{Command Files}, for
1369 more details about @value{GDBN} command files.
1372 Reads the command history recorded in the @dfn{history file}.
1373 @xref{Command History}, for more details about the command history and the
1374 files where @value{GDBN} records it.
1377 Init files use the same syntax as @dfn{command files} (@pxref{Command
1378 Files}) and are processed by @value{GDBN} in the same way. The init
1379 file in your home directory can set options (such as @samp{set
1380 complaints}) that affect subsequent processing of command line options
1381 and operands. Init files are not executed if you use the @samp{-nx}
1382 option (@pxref{Mode Options, ,Choosing Modes}).
1384 To display the list of init files loaded by gdb at startup, you
1385 can use @kbd{gdb --help}.
1387 @cindex init file name
1388 @cindex @file{.gdbinit}
1389 @cindex @file{gdb.ini}
1390 The @value{GDBN} init files are normally called @file{.gdbinit}.
1391 The DJGPP port of @value{GDBN} uses the name @file{gdb.ini}, due to
1392 the limitations of file names imposed by DOS filesystems. The Windows
1393 port of @value{GDBN} uses the standard name, but if it finds a
1394 @file{gdb.ini} file in your home directory, it warns you about that
1395 and suggests to rename the file to the standard name.
1399 @section Quitting @value{GDBN}
1400 @cindex exiting @value{GDBN}
1401 @cindex leaving @value{GDBN}
1404 @kindex quit @r{[}@var{expression}@r{]}
1405 @kindex q @r{(@code{quit})}
1406 @item quit @r{[}@var{expression}@r{]}
1408 To exit @value{GDBN}, use the @code{quit} command (abbreviated
1409 @code{q}), or type an end-of-file character (usually @kbd{Ctrl-d}). If you
1410 do not supply @var{expression}, @value{GDBN} will terminate normally;
1411 otherwise it will terminate using the result of @var{expression} as the
1416 An interrupt (often @kbd{Ctrl-c}) does not exit from @value{GDBN}, but rather
1417 terminates the action of any @value{GDBN} command that is in progress and
1418 returns to @value{GDBN} command level. It is safe to type the interrupt
1419 character at any time because @value{GDBN} does not allow it to take effect
1420 until a time when it is safe.
1422 If you have been using @value{GDBN} to control an attached process or
1423 device, you can release it with the @code{detach} command
1424 (@pxref{Attach, ,Debugging an Already-running Process}).
1426 @node Shell Commands
1427 @section Shell Commands
1429 If you need to execute occasional shell commands during your
1430 debugging session, there is no need to leave or suspend @value{GDBN}; you can
1431 just use the @code{shell} command.
1436 @cindex shell escape
1437 @item shell @var{command-string}
1438 @itemx !@var{command-string}
1439 Invoke a standard shell to execute @var{command-string}.
1440 Note that no space is needed between @code{!} and @var{command-string}.
1441 If it exists, the environment variable @code{SHELL} determines which
1442 shell to run. Otherwise @value{GDBN} uses the default shell
1443 (@file{/bin/sh} on Unix systems, @file{COMMAND.COM} on MS-DOS, etc.).
1446 The utility @code{make} is often needed in development environments.
1447 You do not have to use the @code{shell} command for this purpose in
1452 @cindex calling make
1453 @item make @var{make-args}
1454 Execute the @code{make} program with the specified
1455 arguments. This is equivalent to @samp{shell make @var{make-args}}.
1461 @cindex send the output of a gdb command to a shell command
1463 @item pipe [@var{command}] | @var{shell_command}
1464 @itemx | [@var{command}] | @var{shell_command}
1465 @itemx pipe -d @var{delim} @var{command} @var{delim} @var{shell_command}
1466 @itemx | -d @var{delim} @var{command} @var{delim} @var{shell_command}
1467 Executes @var{command} and sends its output to @var{shell_command}.
1468 Note that no space is needed around @code{|}.
1469 If no @var{command} is provided, the last command executed is repeated.
1471 In case the @var{command} contains a @code{|}, the option @code{-d @var{delim}}
1472 can be used to specify an alternate delimiter string @var{delim} that separates
1473 the @var{command} from the @var{shell_command}.
1506 (gdb) | -d ! echo this contains a | char\n ! sed -e 's/|/PIPE/'
1507 this contains a PIPE char
1508 (gdb) | -d xxx echo this contains a | char!\n xxx sed -e 's/|/PIPE/'
1509 this contains a PIPE char!
1515 The convenience variables @code{$_shell_exitcode} and @code{$_shell_exitsignal}
1516 can be used to examine the exit status of the last shell command launched
1517 by @code{shell}, @code{make}, @code{pipe} and @code{|}.
1518 @xref{Convenience Vars,, Convenience Variables}.
1520 @node Logging Output
1521 @section Logging Output
1522 @cindex logging @value{GDBN} output
1523 @cindex save @value{GDBN} output to a file
1525 You may want to save the output of @value{GDBN} commands to a file.
1526 There are several commands to control @value{GDBN}'s logging.
1530 @item set logging on
1532 @item set logging off
1534 @cindex logging file name
1535 @item set logging file @var{file}
1536 Change the name of the current logfile. The default logfile is @file{gdb.txt}.
1537 @item set logging overwrite [on|off]
1538 By default, @value{GDBN} will append to the logfile. Set @code{overwrite} if
1539 you want @code{set logging on} to overwrite the logfile instead.
1540 @item set logging redirect [on|off]
1541 By default, @value{GDBN} output will go to both the terminal and the logfile.
1542 Set @code{redirect} if you want output to go only to the log file.
1543 @item set logging debugredirect [on|off]
1544 By default, @value{GDBN} debug output will go to both the terminal and the logfile.
1545 Set @code{debugredirect} if you want debug output to go only to the log file.
1546 @kindex show logging
1548 Show the current values of the logging settings.
1551 You can also redirect the output of a @value{GDBN} command to a
1552 shell command. @xref{pipe}.
1554 @chapter @value{GDBN} Commands
1556 You can abbreviate a @value{GDBN} command to the first few letters of the command
1557 name, if that abbreviation is unambiguous; and you can repeat certain
1558 @value{GDBN} commands by typing just @key{RET}. You can also use the @key{TAB}
1559 key to get @value{GDBN} to fill out the rest of a word in a command (or to
1560 show you the alternatives available, if there is more than one possibility).
1563 * Command Syntax:: How to give commands to @value{GDBN}
1564 * Completion:: Command completion
1565 * Command Options:: Command options
1566 * Help:: How to ask @value{GDBN} for help
1569 @node Command Syntax
1570 @section Command Syntax
1572 A @value{GDBN} command is a single line of input. There is no limit on
1573 how long it can be. It starts with a command name, which is followed by
1574 arguments whose meaning depends on the command name. For example, the
1575 command @code{step} accepts an argument which is the number of times to
1576 step, as in @samp{step 5}. You can also use the @code{step} command
1577 with no arguments. Some commands do not allow any arguments.
1579 @cindex abbreviation
1580 @value{GDBN} command names may always be truncated if that abbreviation is
1581 unambiguous. Other possible command abbreviations are listed in the
1582 documentation for individual commands. In some cases, even ambiguous
1583 abbreviations are allowed; for example, @code{s} is specially defined as
1584 equivalent to @code{step} even though there are other commands whose
1585 names start with @code{s}. You can test abbreviations by using them as
1586 arguments to the @code{help} command.
1588 @cindex repeating commands
1589 @kindex RET @r{(repeat last command)}
1590 A blank line as input to @value{GDBN} (typing just @key{RET}) means to
1591 repeat the previous command. Certain commands (for example, @code{run})
1592 will not repeat this way; these are commands whose unintentional
1593 repetition might cause trouble and which you are unlikely to want to
1594 repeat. User-defined commands can disable this feature; see
1595 @ref{Define, dont-repeat}.
1597 The @code{list} and @code{x} commands, when you repeat them with
1598 @key{RET}, construct new arguments rather than repeating
1599 exactly as typed. This permits easy scanning of source or memory.
1601 @value{GDBN} can also use @key{RET} in another way: to partition lengthy
1602 output, in a way similar to the common utility @code{more}
1603 (@pxref{Screen Size,,Screen Size}). Since it is easy to press one
1604 @key{RET} too many in this situation, @value{GDBN} disables command
1605 repetition after any command that generates this sort of display.
1607 @kindex # @r{(a comment)}
1609 Any text from a @kbd{#} to the end of the line is a comment; it does
1610 nothing. This is useful mainly in command files (@pxref{Command
1611 Files,,Command Files}).
1613 @cindex repeating command sequences
1614 @kindex Ctrl-o @r{(operate-and-get-next)}
1615 The @kbd{Ctrl-o} binding is useful for repeating a complex sequence of
1616 commands. This command accepts the current line, like @key{RET}, and
1617 then fetches the next line relative to the current line from the history
1621 @section Command Completion
1624 @cindex word completion
1625 @value{GDBN} can fill in the rest of a word in a command for you, if there is
1626 only one possibility; it can also show you what the valid possibilities
1627 are for the next word in a command, at any time. This works for @value{GDBN}
1628 commands, @value{GDBN} subcommands, command options, and the names of symbols
1631 Press the @key{TAB} key whenever you want @value{GDBN} to fill out the rest
1632 of a word. If there is only one possibility, @value{GDBN} fills in the
1633 word, and waits for you to finish the command (or press @key{RET} to
1634 enter it). For example, if you type
1636 @c FIXME "@key" does not distinguish its argument sufficiently to permit
1637 @c complete accuracy in these examples; space introduced for clarity.
1638 @c If texinfo enhancements make it unnecessary, it would be nice to
1639 @c replace " @key" by "@key" in the following...
1641 (@value{GDBP}) info bre @key{TAB}
1645 @value{GDBN} fills in the rest of the word @samp{breakpoints}, since that is
1646 the only @code{info} subcommand beginning with @samp{bre}:
1649 (@value{GDBP}) info breakpoints
1653 You can either press @key{RET} at this point, to run the @code{info
1654 breakpoints} command, or backspace and enter something else, if
1655 @samp{breakpoints} does not look like the command you expected. (If you
1656 were sure you wanted @code{info breakpoints} in the first place, you
1657 might as well just type @key{RET} immediately after @samp{info bre},
1658 to exploit command abbreviations rather than command completion).
1660 If there is more than one possibility for the next word when you press
1661 @key{TAB}, @value{GDBN} sounds a bell. You can either supply more
1662 characters and try again, or just press @key{TAB} a second time;
1663 @value{GDBN} displays all the possible completions for that word. For
1664 example, you might want to set a breakpoint on a subroutine whose name
1665 begins with @samp{make_}, but when you type @kbd{b make_@key{TAB}} @value{GDBN}
1666 just sounds the bell. Typing @key{TAB} again displays all the
1667 function names in your program that begin with those characters, for
1671 (@value{GDBP}) b make_ @key{TAB}
1672 @exdent @value{GDBN} sounds bell; press @key{TAB} again, to see:
1673 make_a_section_from_file make_environ
1674 make_abs_section make_function_type
1675 make_blockvector make_pointer_type
1676 make_cleanup make_reference_type
1677 make_command make_symbol_completion_list
1678 (@value{GDBP}) b make_
1682 After displaying the available possibilities, @value{GDBN} copies your
1683 partial input (@samp{b make_} in the example) so you can finish the
1686 If you just want to see the list of alternatives in the first place, you
1687 can press @kbd{M-?} rather than pressing @key{TAB} twice. @kbd{M-?}
1688 means @kbd{@key{META} ?}. You can type this either by holding down a
1689 key designated as the @key{META} shift on your keyboard (if there is
1690 one) while typing @kbd{?}, or as @key{ESC} followed by @kbd{?}.
1692 If the number of possible completions is large, @value{GDBN} will
1693 print as much of the list as it has collected, as well as a message
1694 indicating that the list may be truncated.
1697 (@value{GDBP}) b m@key{TAB}@key{TAB}
1699 <... the rest of the possible completions ...>
1700 *** List may be truncated, max-completions reached. ***
1705 This behavior can be controlled with the following commands:
1708 @kindex set max-completions
1709 @item set max-completions @var{limit}
1710 @itemx set max-completions unlimited
1711 Set the maximum number of completion candidates. @value{GDBN} will
1712 stop looking for more completions once it collects this many candidates.
1713 This is useful when completing on things like function names as collecting
1714 all the possible candidates can be time consuming.
1715 The default value is 200. A value of zero disables tab-completion.
1716 Note that setting either no limit or a very large limit can make
1718 @kindex show max-completions
1719 @item show max-completions
1720 Show the maximum number of candidates that @value{GDBN} will collect and show
1724 @cindex quotes in commands
1725 @cindex completion of quoted strings
1726 Sometimes the string you need, while logically a ``word'', may contain
1727 parentheses or other characters that @value{GDBN} normally excludes from
1728 its notion of a word. To permit word completion to work in this
1729 situation, you may enclose words in @code{'} (single quote marks) in
1730 @value{GDBN} commands.
1732 A likely situation where you might need this is in typing an
1733 expression that involves a C@t{++} symbol name with template
1734 parameters. This is because when completing expressions, GDB treats
1735 the @samp{<} character as word delimiter, assuming that it's the
1736 less-than comparison operator (@pxref{C Operators, , C and C@t{++}
1739 For example, when you want to call a C@t{++} template function
1740 interactively using the @code{print} or @code{call} commands, you may
1741 need to distinguish whether you mean the version of @code{name} that
1742 was specialized for @code{int}, @code{name<int>()}, or the version
1743 that was specialized for @code{float}, @code{name<float>()}. To use
1744 the word-completion facilities in this situation, type a single quote
1745 @code{'} at the beginning of the function name. This alerts
1746 @value{GDBN} that it may need to consider more information than usual
1747 when you press @key{TAB} or @kbd{M-?} to request word completion:
1750 (@value{GDBP}) p 'func< @kbd{M-?}
1751 func<int>() func<float>()
1752 (@value{GDBP}) p 'func<
1755 When setting breakpoints however (@pxref{Specify Location}), you don't
1756 usually need to type a quote before the function name, because
1757 @value{GDBN} understands that you want to set a breakpoint on a
1761 (@value{GDBP}) b func< @kbd{M-?}
1762 func<int>() func<float>()
1763 (@value{GDBP}) b func<
1766 This is true even in the case of typing the name of C@t{++} overloaded
1767 functions (multiple definitions of the same function, distinguished by
1768 argument type). For example, when you want to set a breakpoint you
1769 don't need to distinguish whether you mean the version of @code{name}
1770 that takes an @code{int} parameter, @code{name(int)}, or the version
1771 that takes a @code{float} parameter, @code{name(float)}.
1774 (@value{GDBP}) b bubble( @kbd{M-?}
1775 bubble(int) bubble(double)
1776 (@value{GDBP}) b bubble(dou @kbd{M-?}
1780 See @ref{quoting names} for a description of other scenarios that
1783 For more information about overloaded functions, see @ref{C Plus Plus
1784 Expressions, ,C@t{++} Expressions}. You can use the command @code{set
1785 overload-resolution off} to disable overload resolution;
1786 see @ref{Debugging C Plus Plus, ,@value{GDBN} Features for C@t{++}}.
1788 @cindex completion of structure field names
1789 @cindex structure field name completion
1790 @cindex completion of union field names
1791 @cindex union field name completion
1792 When completing in an expression which looks up a field in a
1793 structure, @value{GDBN} also tries@footnote{The completer can be
1794 confused by certain kinds of invalid expressions. Also, it only
1795 examines the static type of the expression, not the dynamic type.} to
1796 limit completions to the field names available in the type of the
1800 (@value{GDBP}) p gdb_stdout.@kbd{M-?}
1801 magic to_fputs to_rewind
1802 to_data to_isatty to_write
1803 to_delete to_put to_write_async_safe
1808 This is because the @code{gdb_stdout} is a variable of the type
1809 @code{struct ui_file} that is defined in @value{GDBN} sources as
1816 ui_file_flush_ftype *to_flush;
1817 ui_file_write_ftype *to_write;
1818 ui_file_write_async_safe_ftype *to_write_async_safe;
1819 ui_file_fputs_ftype *to_fputs;
1820 ui_file_read_ftype *to_read;
1821 ui_file_delete_ftype *to_delete;
1822 ui_file_isatty_ftype *to_isatty;
1823 ui_file_rewind_ftype *to_rewind;
1824 ui_file_put_ftype *to_put;
1829 @node Command Options
1830 @section Command options
1832 @cindex command options
1833 Some commands accept options starting with a leading dash. For
1834 example, @code{print -pretty}. Similarly to command names, you can
1835 abbreviate a @value{GDBN} option to the first few letters of the
1836 option name, if that abbreviation is unambiguous, and you can also use
1837 the @key{TAB} key to get @value{GDBN} to fill out the rest of a word
1838 in an option (or to show you the alternatives available, if there is
1839 more than one possibility).
1841 @cindex command options, raw input
1842 Some commands take raw input as argument. For example, the print
1843 command processes arbitrary expressions in any of the languages
1844 supported by @value{GDBN}. With such commands, because raw input may
1845 start with a leading dash that would be confused with an option or any
1846 of its abbreviations, e.g.@: @code{print -r} (short for @code{print
1847 -raw} or printing negative @code{r}?), if you specify any command
1848 option, then you must use a double-dash (@code{--}) delimiter to
1849 indicate the end of options.
1851 @cindex command options, boolean
1853 Some options are described as accepting an argument which can be
1854 either @code{on} or @code{off}. These are known as @dfn{boolean
1855 options}. Similarly to boolean settings commands---@code{on} and
1856 @code{off} are the typical values, but any of @code{1}, @code{yes} and
1857 @code{enable} can also be used as ``true'' value, and any of @code{0},
1858 @code{no} and @code{disable} can also be used as ``false'' value. You
1859 can also omit a ``true'' value, as it is implied by default.
1861 For example, these are equivalent:
1864 (@value{GDBP}) print -object on -pretty off -element unlimited -- *myptr
1865 (@value{GDBP}) p -o -p 0 -e u -- *myptr
1868 You can discover the set of options some command accepts by completing
1869 on @code{-} after the command name. For example:
1872 (@value{GDBP}) print -@key{TAB}@key{TAB}
1873 -address -max-depth -repeats -vtbl
1874 -array -null-stop -static-members
1875 -array-indexes -object -symbol
1876 -elements -pretty -union
1879 Completion will in some cases guide you with a suggestion of what kind
1880 of argument an option expects. For example:
1883 (@value{GDBP}) print -elements @key{TAB}@key{TAB}
1887 Here, the option expects a number (e.g., @code{100}), not literal
1888 @code{NUMBER}. Such metasyntactical arguments are always presented in
1891 (For more on using the @code{print} command, see @ref{Data, ,Examining
1895 @section Getting Help
1896 @cindex online documentation
1899 You can always ask @value{GDBN} itself for information on its commands,
1900 using the command @code{help}.
1903 @kindex h @r{(@code{help})}
1906 You can use @code{help} (abbreviated @code{h}) with no arguments to
1907 display a short list of named classes of commands:
1911 List of classes of commands:
1913 aliases -- Aliases of other commands
1914 breakpoints -- Making program stop at certain points
1915 data -- Examining data
1916 files -- Specifying and examining files
1917 internals -- Maintenance commands
1918 obscure -- Obscure features
1919 running -- Running the program
1920 stack -- Examining the stack
1921 status -- Status inquiries
1922 support -- Support facilities
1923 tracepoints -- Tracing of program execution without
1924 stopping the program
1925 user-defined -- User-defined commands
1927 Type "help" followed by a class name for a list of
1928 commands in that class.
1929 Type "help" followed by command name for full
1931 Command name abbreviations are allowed if unambiguous.
1934 @c the above line break eliminates huge line overfull...
1936 @item help @var{class}
1937 Using one of the general help classes as an argument, you can get a
1938 list of the individual commands in that class. For example, here is the
1939 help display for the class @code{status}:
1942 (@value{GDBP}) help status
1947 @c Line break in "show" line falsifies real output, but needed
1948 @c to fit in smallbook page size.
1949 info -- Generic command for showing things
1950 about the program being debugged
1951 show -- Generic command for showing things
1954 Type "help" followed by command name for full
1956 Command name abbreviations are allowed if unambiguous.
1960 @item help @var{command}
1961 With a command name as @code{help} argument, @value{GDBN} displays a
1962 short paragraph on how to use that command.
1965 @item apropos [-v] @var{regexp}
1966 The @code{apropos} command searches through all of the @value{GDBN}
1967 commands, and their documentation, for the regular expression specified in
1968 @var{args}. It prints out all matches found. The optional flag @samp{-v},
1969 which stands for @samp{verbose}, indicates to output the full documentation
1970 of the matching commands and highlight the parts of the documentation
1971 matching @var{regexp}. For example:
1982 alias -- Define a new command that is an alias of an existing command
1983 aliases -- Aliases of other commands
1984 d -- Delete some breakpoints or auto-display expressions
1985 del -- Delete some breakpoints or auto-display expressions
1986 delete -- Delete some breakpoints or auto-display expressions
1994 apropos -v cut.*thread apply
1998 results in the below output, where @samp{cut for 'thread apply}
1999 is highlighted if styling is enabled.
2003 taas -- Apply a command to all threads (ignoring errors
2006 shortcut for 'thread apply all -s COMMAND'
2008 tfaas -- Apply a command to all frames of all threads
2009 (ignoring errors and empty output).
2010 Usage: tfaas COMMAND
2011 shortcut for 'thread apply all -s frame apply all -s COMMAND'
2016 @item complete @var{args}
2017 The @code{complete @var{args}} command lists all the possible completions
2018 for the beginning of a command. Use @var{args} to specify the beginning of the
2019 command you want completed. For example:
2025 @noindent results in:
2036 @noindent This is intended for use by @sc{gnu} Emacs.
2039 In addition to @code{help}, you can use the @value{GDBN} commands @code{info}
2040 and @code{show} to inquire about the state of your program, or the state
2041 of @value{GDBN} itself. Each command supports many topics of inquiry; this
2042 manual introduces each of them in the appropriate context. The listings
2043 under @code{info} and under @code{show} in the Command, Variable, and
2044 Function Index point to all the sub-commands. @xref{Command and Variable
2050 @kindex i @r{(@code{info})}
2052 This command (abbreviated @code{i}) is for describing the state of your
2053 program. For example, you can show the arguments passed to a function
2054 with @code{info args}, list the registers currently in use with @code{info
2055 registers}, or list the breakpoints you have set with @code{info breakpoints}.
2056 You can get a complete list of the @code{info} sub-commands with
2057 @w{@code{help info}}.
2061 You can assign the result of an expression to an environment variable with
2062 @code{set}. For example, you can set the @value{GDBN} prompt to a $-sign with
2063 @code{set prompt $}.
2067 In contrast to @code{info}, @code{show} is for describing the state of
2068 @value{GDBN} itself.
2069 You can change most of the things you can @code{show}, by using the
2070 related command @code{set}; for example, you can control what number
2071 system is used for displays with @code{set radix}, or simply inquire
2072 which is currently in use with @code{show radix}.
2075 To display all the settable parameters and their current
2076 values, you can use @code{show} with no arguments; you may also use
2077 @code{info set}. Both commands produce the same display.
2078 @c FIXME: "info set" violates the rule that "info" is for state of
2079 @c FIXME...program. Ck w/ GNU: "info set" to be called something else,
2080 @c FIXME...or change desc of rule---eg "state of prog and debugging session"?
2084 Here are several miscellaneous @code{show} subcommands, all of which are
2085 exceptional in lacking corresponding @code{set} commands:
2088 @kindex show version
2089 @cindex @value{GDBN} version number
2091 Show what version of @value{GDBN} is running. You should include this
2092 information in @value{GDBN} bug-reports. If multiple versions of
2093 @value{GDBN} are in use at your site, you may need to determine which
2094 version of @value{GDBN} you are running; as @value{GDBN} evolves, new
2095 commands are introduced, and old ones may wither away. Also, many
2096 system vendors ship variant versions of @value{GDBN}, and there are
2097 variant versions of @value{GDBN} in @sc{gnu}/Linux distributions as well.
2098 The version number is the same as the one announced when you start
2101 @kindex show copying
2102 @kindex info copying
2103 @cindex display @value{GDBN} copyright
2106 Display information about permission for copying @value{GDBN}.
2108 @kindex show warranty
2109 @kindex info warranty
2111 @itemx info warranty
2112 Display the @sc{gnu} ``NO WARRANTY'' statement, or a warranty,
2113 if your version of @value{GDBN} comes with one.
2115 @kindex show configuration
2116 @item show configuration
2117 Display detailed information about the way @value{GDBN} was configured
2118 when it was built. This displays the optional arguments passed to the
2119 @file{configure} script and also configuration parameters detected
2120 automatically by @command{configure}. When reporting a @value{GDBN}
2121 bug (@pxref{GDB Bugs}), it is important to include this information in
2127 @chapter Running Programs Under @value{GDBN}
2129 When you run a program under @value{GDBN}, you must first generate
2130 debugging information when you compile it.
2132 You may start @value{GDBN} with its arguments, if any, in an environment
2133 of your choice. If you are doing native debugging, you may redirect
2134 your program's input and output, debug an already running process, or
2135 kill a child process.
2138 * Compilation:: Compiling for debugging
2139 * Starting:: Starting your program
2140 * Arguments:: Your program's arguments
2141 * Environment:: Your program's environment
2143 * Working Directory:: Your program's working directory
2144 * Input/Output:: Your program's input and output
2145 * Attach:: Debugging an already-running process
2146 * Kill Process:: Killing the child process
2148 * Inferiors and Programs:: Debugging multiple inferiors and programs
2149 * Threads:: Debugging programs with multiple threads
2150 * Forks:: Debugging forks
2151 * Checkpoint/Restart:: Setting a @emph{bookmark} to return to later
2155 @section Compiling for Debugging
2157 In order to debug a program effectively, you need to generate
2158 debugging information when you compile it. This debugging information
2159 is stored in the object file; it describes the data type of each
2160 variable or function and the correspondence between source line numbers
2161 and addresses in the executable code.
2163 To request debugging information, specify the @samp{-g} option when you run
2166 Programs that are to be shipped to your customers are compiled with
2167 optimizations, using the @samp{-O} compiler option. However, some
2168 compilers are unable to handle the @samp{-g} and @samp{-O} options
2169 together. Using those compilers, you cannot generate optimized
2170 executables containing debugging information.
2172 @value{NGCC}, the @sc{gnu} C/C@t{++} compiler, supports @samp{-g} with or
2173 without @samp{-O}, making it possible to debug optimized code. We
2174 recommend that you @emph{always} use @samp{-g} whenever you compile a
2175 program. You may think your program is correct, but there is no sense
2176 in pushing your luck. For more information, see @ref{Optimized Code}.
2178 Older versions of the @sc{gnu} C compiler permitted a variant option
2179 @w{@samp{-gg}} for debugging information. @value{GDBN} no longer supports this
2180 format; if your @sc{gnu} C compiler has this option, do not use it.
2182 @value{GDBN} knows about preprocessor macros and can show you their
2183 expansion (@pxref{Macros}). Most compilers do not include information
2184 about preprocessor macros in the debugging information if you specify
2185 the @option{-g} flag alone. Version 3.1 and later of @value{NGCC},
2186 the @sc{gnu} C compiler, provides macro information if you are using
2187 the DWARF debugging format, and specify the option @option{-g3}.
2189 @xref{Debugging Options,,Options for Debugging Your Program or GCC,
2190 gcc, Using the @sc{gnu} Compiler Collection (GCC)}, for more
2191 information on @value{NGCC} options affecting debug information.
2193 You will have the best debugging experience if you use the latest
2194 version of the DWARF debugging format that your compiler supports.
2195 DWARF is currently the most expressive and best supported debugging
2196 format in @value{GDBN}.
2200 @section Starting your Program
2206 @kindex r @r{(@code{run})}
2209 Use the @code{run} command to start your program under @value{GDBN}.
2210 You must first specify the program name with an argument to
2211 @value{GDBN} (@pxref{Invocation, ,Getting In and Out of
2212 @value{GDBN}}), or by using the @code{file} or @code{exec-file}
2213 command (@pxref{Files, ,Commands to Specify Files}).
2217 If you are running your program in an execution environment that
2218 supports processes, @code{run} creates an inferior process and makes
2219 that process run your program. In some environments without processes,
2220 @code{run} jumps to the start of your program. Other targets,
2221 like @samp{remote}, are always running. If you get an error
2222 message like this one:
2225 The "remote" target does not support "run".
2226 Try "help target" or "continue".
2230 then use @code{continue} to run your program. You may need @code{load}
2231 first (@pxref{load}).
2233 The execution of a program is affected by certain information it
2234 receives from its superior. @value{GDBN} provides ways to specify this
2235 information, which you must do @emph{before} starting your program. (You
2236 can change it after starting your program, but such changes only affect
2237 your program the next time you start it.) This information may be
2238 divided into four categories:
2241 @item The @emph{arguments.}
2242 Specify the arguments to give your program as the arguments of the
2243 @code{run} command. If a shell is available on your target, the shell
2244 is used to pass the arguments, so that you may use normal conventions
2245 (such as wildcard expansion or variable substitution) in describing
2247 In Unix systems, you can control which shell is used with the
2248 @code{SHELL} environment variable. If you do not define @code{SHELL},
2249 @value{GDBN} uses the default shell (@file{/bin/sh}). You can disable
2250 use of any shell with the @code{set startup-with-shell} command (see
2253 @item The @emph{environment.}
2254 Your program normally inherits its environment from @value{GDBN}, but you can
2255 use the @value{GDBN} commands @code{set environment} and @code{unset
2256 environment} to change parts of the environment that affect
2257 your program. @xref{Environment, ,Your Program's Environment}.
2259 @item The @emph{working directory.}
2260 You can set your program's working directory with the command
2261 @kbd{set cwd}. If you do not set any working directory with this
2262 command, your program will inherit @value{GDBN}'s working directory if
2263 native debugging, or the remote server's working directory if remote
2264 debugging. @xref{Working Directory, ,Your Program's Working
2267 @item The @emph{standard input and output.}
2268 Your program normally uses the same device for standard input and
2269 standard output as @value{GDBN} is using. You can redirect input and output
2270 in the @code{run} command line, or you can use the @code{tty} command to
2271 set a different device for your program.
2272 @xref{Input/Output, ,Your Program's Input and Output}.
2275 @emph{Warning:} While input and output redirection work, you cannot use
2276 pipes to pass the output of the program you are debugging to another
2277 program; if you attempt this, @value{GDBN} is likely to wind up debugging the
2281 When you issue the @code{run} command, your program begins to execute
2282 immediately. @xref{Stopping, ,Stopping and Continuing}, for discussion
2283 of how to arrange for your program to stop. Once your program has
2284 stopped, you may call functions in your program, using the @code{print}
2285 or @code{call} commands. @xref{Data, ,Examining Data}.
2287 If the modification time of your symbol file has changed since the last
2288 time @value{GDBN} read its symbols, @value{GDBN} discards its symbol
2289 table, and reads it again. When it does this, @value{GDBN} tries to retain
2290 your current breakpoints.
2295 @cindex run to main procedure
2296 The name of the main procedure can vary from language to language.
2297 With C or C@t{++}, the main procedure name is always @code{main}, but
2298 other languages such as Ada do not require a specific name for their
2299 main procedure. The debugger provides a convenient way to start the
2300 execution of the program and to stop at the beginning of the main
2301 procedure, depending on the language used.
2303 The @samp{start} command does the equivalent of setting a temporary
2304 breakpoint at the beginning of the main procedure and then invoking
2305 the @samp{run} command.
2307 @cindex elaboration phase
2308 Some programs contain an @dfn{elaboration} phase where some startup code is
2309 executed before the main procedure is called. This depends on the
2310 languages used to write your program. In C@t{++}, for instance,
2311 constructors for static and global objects are executed before
2312 @code{main} is called. It is therefore possible that the debugger stops
2313 before reaching the main procedure. However, the temporary breakpoint
2314 will remain to halt execution.
2316 Specify the arguments to give to your program as arguments to the
2317 @samp{start} command. These arguments will be given verbatim to the
2318 underlying @samp{run} command. Note that the same arguments will be
2319 reused if no argument is provided during subsequent calls to
2320 @samp{start} or @samp{run}.
2322 It is sometimes necessary to debug the program during elaboration. In
2323 these cases, using the @code{start} command would stop the execution
2324 of your program too late, as the program would have already completed
2325 the elaboration phase. Under these circumstances, either insert
2326 breakpoints in your elaboration code before running your program or
2327 use the @code{starti} command.
2331 @cindex run to first instruction
2332 The @samp{starti} command does the equivalent of setting a temporary
2333 breakpoint at the first instruction of a program's execution and then
2334 invoking the @samp{run} command. For programs containing an
2335 elaboration phase, the @code{starti} command will stop execution at
2336 the start of the elaboration phase.
2338 @anchor{set exec-wrapper}
2339 @kindex set exec-wrapper
2340 @item set exec-wrapper @var{wrapper}
2341 @itemx show exec-wrapper
2342 @itemx unset exec-wrapper
2343 When @samp{exec-wrapper} is set, the specified wrapper is used to
2344 launch programs for debugging. @value{GDBN} starts your program
2345 with a shell command of the form @kbd{exec @var{wrapper}
2346 @var{program}}. Quoting is added to @var{program} and its
2347 arguments, but not to @var{wrapper}, so you should add quotes if
2348 appropriate for your shell. The wrapper runs until it executes
2349 your program, and then @value{GDBN} takes control.
2351 You can use any program that eventually calls @code{execve} with
2352 its arguments as a wrapper. Several standard Unix utilities do
2353 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
2354 with @code{exec "$@@"} will also work.
2356 For example, you can use @code{env} to pass an environment variable to
2357 the debugged program, without setting the variable in your shell's
2361 (@value{GDBP}) set exec-wrapper env 'LD_PRELOAD=libtest.so'
2365 This command is available when debugging locally on most targets, excluding
2366 @sc{djgpp}, Cygwin, MS Windows, and QNX Neutrino.
2368 @kindex set startup-with-shell
2369 @anchor{set startup-with-shell}
2370 @item set startup-with-shell
2371 @itemx set startup-with-shell on
2372 @itemx set startup-with-shell off
2373 @itemx show startup-with-shell
2374 On Unix systems, by default, if a shell is available on your target,
2375 @value{GDBN}) uses it to start your program. Arguments of the
2376 @code{run} command are passed to the shell, which does variable
2377 substitution, expands wildcard characters and performs redirection of
2378 I/O. In some circumstances, it may be useful to disable such use of a
2379 shell, for example, when debugging the shell itself or diagnosing
2380 startup failures such as:
2384 Starting program: ./a.out
2385 During startup program terminated with signal SIGSEGV, Segmentation fault.
2389 which indicates the shell or the wrapper specified with
2390 @samp{exec-wrapper} crashed, not your program. Most often, this is
2391 caused by something odd in your shell's non-interactive mode
2392 initialization file---such as @file{.cshrc} for C-shell,
2393 $@file{.zshenv} for the Z shell, or the file specified in the
2394 @samp{BASH_ENV} environment variable for BASH.
2396 @anchor{set auto-connect-native-target}
2397 @kindex set auto-connect-native-target
2398 @item set auto-connect-native-target
2399 @itemx set auto-connect-native-target on
2400 @itemx set auto-connect-native-target off
2401 @itemx show auto-connect-native-target
2403 By default, if not connected to any target yet (e.g., with
2404 @code{target remote}), the @code{run} command starts your program as a
2405 native process under @value{GDBN}, on your local machine. If you're
2406 sure you don't want to debug programs on your local machine, you can
2407 tell @value{GDBN} to not connect to the native target automatically
2408 with the @code{set auto-connect-native-target off} command.
2410 If @code{on}, which is the default, and if @value{GDBN} is not
2411 connected to a target already, the @code{run} command automaticaly
2412 connects to the native target, if one is available.
2414 If @code{off}, and if @value{GDBN} is not connected to a target
2415 already, the @code{run} command fails with an error:
2419 Don't know how to run. Try "help target".
2422 If @value{GDBN} is already connected to a target, @value{GDBN} always
2423 uses it with the @code{run} command.
2425 In any case, you can explicitly connect to the native target with the
2426 @code{target native} command. For example,
2429 (@value{GDBP}) set auto-connect-native-target off
2431 Don't know how to run. Try "help target".
2432 (@value{GDBP}) target native
2434 Starting program: ./a.out
2435 [Inferior 1 (process 10421) exited normally]
2438 In case you connected explicitly to the @code{native} target,
2439 @value{GDBN} remains connected even if all inferiors exit, ready for
2440 the next @code{run} command. Use the @code{disconnect} command to
2443 Examples of other commands that likewise respect the
2444 @code{auto-connect-native-target} setting: @code{attach}, @code{info
2445 proc}, @code{info os}.
2447 @kindex set disable-randomization
2448 @item set disable-randomization
2449 @itemx set disable-randomization on
2450 This option (enabled by default in @value{GDBN}) will turn off the native
2451 randomization of the virtual address space of the started program. This option
2452 is useful for multiple debugging sessions to make the execution better
2453 reproducible and memory addresses reusable across debugging sessions.
2455 This feature is implemented only on certain targets, including @sc{gnu}/Linux.
2456 On @sc{gnu}/Linux you can get the same behavior using
2459 (@value{GDBP}) set exec-wrapper setarch `uname -m` -R
2462 @item set disable-randomization off
2463 Leave the behavior of the started executable unchanged. Some bugs rear their
2464 ugly heads only when the program is loaded at certain addresses. If your bug
2465 disappears when you run the program under @value{GDBN}, that might be because
2466 @value{GDBN} by default disables the address randomization on platforms, such
2467 as @sc{gnu}/Linux, which do that for stand-alone programs. Use @kbd{set
2468 disable-randomization off} to try to reproduce such elusive bugs.
2470 On targets where it is available, virtual address space randomization
2471 protects the programs against certain kinds of security attacks. In these
2472 cases the attacker needs to know the exact location of a concrete executable
2473 code. Randomizing its location makes it impossible to inject jumps misusing
2474 a code at its expected addresses.
2476 Prelinking shared libraries provides a startup performance advantage but it
2477 makes addresses in these libraries predictable for privileged processes by
2478 having just unprivileged access at the target system. Reading the shared
2479 library binary gives enough information for assembling the malicious code
2480 misusing it. Still even a prelinked shared library can get loaded at a new
2481 random address just requiring the regular relocation process during the
2482 startup. Shared libraries not already prelinked are always loaded at
2483 a randomly chosen address.
2485 Position independent executables (PIE) contain position independent code
2486 similar to the shared libraries and therefore such executables get loaded at
2487 a randomly chosen address upon startup. PIE executables always load even
2488 already prelinked shared libraries at a random address. You can build such
2489 executable using @command{gcc -fPIE -pie}.
2491 Heap (malloc storage), stack and custom mmap areas are always placed randomly
2492 (as long as the randomization is enabled).
2494 @item show disable-randomization
2495 Show the current setting of the explicit disable of the native randomization of
2496 the virtual address space of the started program.
2501 @section Your Program's Arguments
2503 @cindex arguments (to your program)
2504 The arguments to your program can be specified by the arguments of the
2506 They are passed to a shell, which expands wildcard characters and
2507 performs redirection of I/O, and thence to your program. Your
2508 @code{SHELL} environment variable (if it exists) specifies what shell
2509 @value{GDBN} uses. If you do not define @code{SHELL}, @value{GDBN} uses
2510 the default shell (@file{/bin/sh} on Unix).
2512 On non-Unix systems, the program is usually invoked directly by
2513 @value{GDBN}, which emulates I/O redirection via the appropriate system
2514 calls, and the wildcard characters are expanded by the startup code of
2515 the program, not by the shell.
2517 @code{run} with no arguments uses the same arguments used by the previous
2518 @code{run}, or those set by the @code{set args} command.
2523 Specify the arguments to be used the next time your program is run. If
2524 @code{set args} has no arguments, @code{run} executes your program
2525 with no arguments. Once you have run your program with arguments,
2526 using @code{set args} before the next @code{run} is the only way to run
2527 it again without arguments.
2531 Show the arguments to give your program when it is started.
2535 @section Your Program's Environment
2537 @cindex environment (of your program)
2538 The @dfn{environment} consists of a set of environment variables and
2539 their values. Environment variables conventionally record such things as
2540 your user name, your home directory, your terminal type, and your search
2541 path for programs to run. Usually you set up environment variables with
2542 the shell and they are inherited by all the other programs you run. When
2543 debugging, it can be useful to try running your program with a modified
2544 environment without having to start @value{GDBN} over again.
2548 @item path @var{directory}
2549 Add @var{directory} to the front of the @code{PATH} environment variable
2550 (the search path for executables) that will be passed to your program.
2551 The value of @code{PATH} used by @value{GDBN} does not change.
2552 You may specify several directory names, separated by whitespace or by a
2553 system-dependent separator character (@samp{:} on Unix, @samp{;} on
2554 MS-DOS and MS-Windows). If @var{directory} is already in the path, it
2555 is moved to the front, so it is searched sooner.
2557 You can use the string @samp{$cwd} to refer to whatever is the current
2558 working directory at the time @value{GDBN} searches the path. If you
2559 use @samp{.} instead, it refers to the directory where you executed the
2560 @code{path} command. @value{GDBN} replaces @samp{.} in the
2561 @var{directory} argument (with the current path) before adding
2562 @var{directory} to the search path.
2563 @c 'path' is explicitly nonrepeatable, but RMS points out it is silly to
2564 @c document that, since repeating it would be a no-op.
2568 Display the list of search paths for executables (the @code{PATH}
2569 environment variable).
2571 @kindex show environment
2572 @item show environment @r{[}@var{varname}@r{]}
2573 Print the value of environment variable @var{varname} to be given to
2574 your program when it starts. If you do not supply @var{varname},
2575 print the names and values of all environment variables to be given to
2576 your program. You can abbreviate @code{environment} as @code{env}.
2578 @kindex set environment
2579 @anchor{set environment}
2580 @item set environment @var{varname} @r{[}=@var{value}@r{]}
2581 Set environment variable @var{varname} to @var{value}. The value
2582 changes for your program (and the shell @value{GDBN} uses to launch
2583 it), not for @value{GDBN} itself. The @var{value} may be any string; the
2584 values of environment variables are just strings, and any
2585 interpretation is supplied by your program itself. The @var{value}
2586 parameter is optional; if it is eliminated, the variable is set to a
2588 @c "any string" here does not include leading, trailing
2589 @c blanks. Gnu asks: does anyone care?
2591 For example, this command:
2598 tells the debugged program, when subsequently run, that its user is named
2599 @samp{foo}. (The spaces around @samp{=} are used for clarity here; they
2600 are not actually required.)
2602 Note that on Unix systems, @value{GDBN} runs your program via a shell,
2603 which also inherits the environment set with @code{set environment}.
2604 If necessary, you can avoid that by using the @samp{env} program as a
2605 wrapper instead of using @code{set environment}. @xref{set
2606 exec-wrapper}, for an example doing just that.
2608 Environment variables that are set by the user are also transmitted to
2609 @command{gdbserver} to be used when starting the remote inferior.
2610 @pxref{QEnvironmentHexEncoded}.
2612 @kindex unset environment
2613 @anchor{unset environment}
2614 @item unset environment @var{varname}
2615 Remove variable @var{varname} from the environment to be passed to your
2616 program. This is different from @samp{set env @var{varname} =};
2617 @code{unset environment} removes the variable from the environment,
2618 rather than assigning it an empty value.
2620 Environment variables that are unset by the user are also unset on
2621 @command{gdbserver} when starting the remote inferior.
2622 @pxref{QEnvironmentUnset}.
2625 @emph{Warning:} On Unix systems, @value{GDBN} runs your program using
2626 the shell indicated by your @code{SHELL} environment variable if it
2627 exists (or @code{/bin/sh} if not). If your @code{SHELL} variable
2628 names a shell that runs an initialization file when started
2629 non-interactively---such as @file{.cshrc} for C-shell, $@file{.zshenv}
2630 for the Z shell, or the file specified in the @samp{BASH_ENV}
2631 environment variable for BASH---any variables you set in that file
2632 affect your program. You may wish to move setting of environment
2633 variables to files that are only run when you sign on, such as
2634 @file{.login} or @file{.profile}.
2636 @node Working Directory
2637 @section Your Program's Working Directory
2639 @cindex working directory (of your program)
2640 Each time you start your program with @code{run}, the inferior will be
2641 initialized with the current working directory specified by the
2642 @kbd{set cwd} command. If no directory has been specified by this
2643 command, then the inferior will inherit @value{GDBN}'s current working
2644 directory as its working directory if native debugging, or it will
2645 inherit the remote server's current working directory if remote
2650 @cindex change inferior's working directory
2651 @anchor{set cwd command}
2652 @item set cwd @r{[}@var{directory}@r{]}
2653 Set the inferior's working directory to @var{directory}, which will be
2654 @code{glob}-expanded in order to resolve tildes (@file{~}). If no
2655 argument has been specified, the command clears the setting and resets
2656 it to an empty state. This setting has no effect on @value{GDBN}'s
2657 working directory, and it only takes effect the next time you start
2658 the inferior. The @file{~} in @var{directory} is a short for the
2659 @dfn{home directory}, usually pointed to by the @env{HOME} environment
2660 variable. On MS-Windows, if @env{HOME} is not defined, @value{GDBN}
2661 uses the concatenation of @env{HOMEDRIVE} and @env{HOMEPATH} as
2664 You can also change @value{GDBN}'s current working directory by using
2665 the @code{cd} command.
2669 @cindex show inferior's working directory
2671 Show the inferior's working directory. If no directory has been
2672 specified by @kbd{set cwd}, then the default inferior's working
2673 directory is the same as @value{GDBN}'s working directory.
2676 @cindex change @value{GDBN}'s working directory
2678 @item cd @r{[}@var{directory}@r{]}
2679 Set the @value{GDBN} working directory to @var{directory}. If not
2680 given, @var{directory} uses @file{'~'}.
2682 The @value{GDBN} working directory serves as a default for the
2683 commands that specify files for @value{GDBN} to operate on.
2684 @xref{Files, ,Commands to Specify Files}.
2685 @xref{set cwd command}.
2689 Print the @value{GDBN} working directory.
2692 It is generally impossible to find the current working directory of
2693 the process being debugged (since a program can change its directory
2694 during its run). If you work on a system where @value{GDBN} supports
2695 the @code{info proc} command (@pxref{Process Information}), you can
2696 use the @code{info proc} command to find out the
2697 current working directory of the debuggee.
2700 @section Your Program's Input and Output
2705 By default, the program you run under @value{GDBN} does input and output to
2706 the same terminal that @value{GDBN} uses. @value{GDBN} switches the terminal
2707 to its own terminal modes to interact with you, but it records the terminal
2708 modes your program was using and switches back to them when you continue
2709 running your program.
2712 @kindex info terminal
2714 Displays information recorded by @value{GDBN} about the terminal modes your
2718 You can redirect your program's input and/or output using shell
2719 redirection with the @code{run} command. For example,
2726 starts your program, diverting its output to the file @file{outfile}.
2729 @cindex controlling terminal
2730 Another way to specify where your program should do input and output is
2731 with the @code{tty} command. This command accepts a file name as
2732 argument, and causes this file to be the default for future @code{run}
2733 commands. It also resets the controlling terminal for the child
2734 process, for future @code{run} commands. For example,
2741 directs that processes started with subsequent @code{run} commands
2742 default to do input and output on the terminal @file{/dev/ttyb} and have
2743 that as their controlling terminal.
2745 An explicit redirection in @code{run} overrides the @code{tty} command's
2746 effect on the input/output device, but not its effect on the controlling
2749 When you use the @code{tty} command or redirect input in the @code{run}
2750 command, only the input @emph{for your program} is affected. The input
2751 for @value{GDBN} still comes from your terminal. @code{tty} is an alias
2752 for @code{set inferior-tty}.
2754 @cindex inferior tty
2755 @cindex set inferior controlling terminal
2756 You can use the @code{show inferior-tty} command to tell @value{GDBN} to
2757 display the name of the terminal that will be used for future runs of your
2761 @item set inferior-tty [ @var{tty} ]
2762 @kindex set inferior-tty
2763 Set the tty for the program being debugged to @var{tty}. Omitting @var{tty}
2764 restores the default behavior, which is to use the same terminal as
2767 @item show inferior-tty
2768 @kindex show inferior-tty
2769 Show the current tty for the program being debugged.
2773 @section Debugging an Already-running Process
2778 @item attach @var{process-id}
2779 This command attaches to a running process---one that was started
2780 outside @value{GDBN}. (@code{info files} shows your active
2781 targets.) The command takes as argument a process ID. The usual way to
2782 find out the @var{process-id} of a Unix process is with the @code{ps} utility,
2783 or with the @samp{jobs -l} shell command.
2785 @code{attach} does not repeat if you press @key{RET} a second time after
2786 executing the command.
2789 To use @code{attach}, your program must be running in an environment
2790 which supports processes; for example, @code{attach} does not work for
2791 programs on bare-board targets that lack an operating system. You must
2792 also have permission to send the process a signal.
2794 When you use @code{attach}, the debugger finds the program running in
2795 the process first by looking in the current working directory, then (if
2796 the program is not found) by using the source file search path
2797 (@pxref{Source Path, ,Specifying Source Directories}). You can also use
2798 the @code{file} command to load the program. @xref{Files, ,Commands to
2801 The first thing @value{GDBN} does after arranging to debug the specified
2802 process is to stop it. You can examine and modify an attached process
2803 with all the @value{GDBN} commands that are ordinarily available when
2804 you start processes with @code{run}. You can insert breakpoints; you
2805 can step and continue; you can modify storage. If you would rather the
2806 process continue running, you may use the @code{continue} command after
2807 attaching @value{GDBN} to the process.
2812 When you have finished debugging the attached process, you can use the
2813 @code{detach} command to release it from @value{GDBN} control. Detaching
2814 the process continues its execution. After the @code{detach} command,
2815 that process and @value{GDBN} become completely independent once more, and you
2816 are ready to @code{attach} another process or start one with @code{run}.
2817 @code{detach} does not repeat if you press @key{RET} again after
2818 executing the command.
2821 If you exit @value{GDBN} while you have an attached process, you detach
2822 that process. If you use the @code{run} command, you kill that process.
2823 By default, @value{GDBN} asks for confirmation if you try to do either of these
2824 things; you can control whether or not you need to confirm by using the
2825 @code{set confirm} command (@pxref{Messages/Warnings, ,Optional Warnings and
2829 @section Killing the Child Process
2834 Kill the child process in which your program is running under @value{GDBN}.
2837 This command is useful if you wish to debug a core dump instead of a
2838 running process. @value{GDBN} ignores any core dump file while your program
2841 On some operating systems, a program cannot be executed outside @value{GDBN}
2842 while you have breakpoints set on it inside @value{GDBN}. You can use the
2843 @code{kill} command in this situation to permit running your program
2844 outside the debugger.
2846 The @code{kill} command is also useful if you wish to recompile and
2847 relink your program, since on many systems it is impossible to modify an
2848 executable file while it is running in a process. In this case, when you
2849 next type @code{run}, @value{GDBN} notices that the file has changed, and
2850 reads the symbol table again (while trying to preserve your current
2851 breakpoint settings).
2853 @node Inferiors and Programs
2854 @section Debugging Multiple Inferiors and Programs
2856 @value{GDBN} lets you run and debug multiple programs in a single
2857 session. In addition, @value{GDBN} on some systems may let you run
2858 several programs simultaneously (otherwise you have to exit from one
2859 before starting another). In the most general case, you can have
2860 multiple threads of execution in each of multiple processes, launched
2861 from multiple executables.
2864 @value{GDBN} represents the state of each program execution with an
2865 object called an @dfn{inferior}. An inferior typically corresponds to
2866 a process, but is more general and applies also to targets that do not
2867 have processes. Inferiors may be created before a process runs, and
2868 may be retained after a process exits. Inferiors have unique
2869 identifiers that are different from process ids. Usually each
2870 inferior will also have its own distinct address space, although some
2871 embedded targets may have several inferiors running in different parts
2872 of a single address space. Each inferior may in turn have multiple
2873 threads running in it.
2875 To find out what inferiors exist at any moment, use @w{@code{info
2879 @kindex info inferiors [ @var{id}@dots{} ]
2880 @item info inferiors
2881 Print a list of all inferiors currently being managed by @value{GDBN}.
2882 By default all inferiors are printed, but the argument @var{id}@dots{}
2883 -- a space separated list of inferior numbers -- can be used to limit
2884 the display to just the requested inferiors.
2886 @value{GDBN} displays for each inferior (in this order):
2890 the inferior number assigned by @value{GDBN}
2893 the target system's inferior identifier
2896 the name of the executable the inferior is running.
2901 An asterisk @samp{*} preceding the @value{GDBN} inferior number
2902 indicates the current inferior.
2906 @c end table here to get a little more width for example
2909 (@value{GDBP}) info inferiors
2910 Num Description Executable
2911 2 process 2307 hello
2912 * 1 process 3401 goodbye
2915 To switch focus between inferiors, use the @code{inferior} command:
2918 @kindex inferior @var{infno}
2919 @item inferior @var{infno}
2920 Make inferior number @var{infno} the current inferior. The argument
2921 @var{infno} is the inferior number assigned by @value{GDBN}, as shown
2922 in the first field of the @samp{info inferiors} display.
2925 @vindex $_inferior@r{, convenience variable}
2926 The debugger convenience variable @samp{$_inferior} contains the
2927 number of the current inferior. You may find this useful in writing
2928 breakpoint conditional expressions, command scripts, and so forth.
2929 @xref{Convenience Vars,, Convenience Variables}, for general
2930 information on convenience variables.
2932 You can get multiple executables into a debugging session via the
2933 @code{add-inferior} and @w{@code{clone-inferior}} commands. On some
2934 systems @value{GDBN} can add inferiors to the debug session
2935 automatically by following calls to @code{fork} and @code{exec}. To
2936 remove inferiors from the debugging session use the
2937 @w{@code{remove-inferiors}} command.
2940 @kindex add-inferior
2941 @item add-inferior [ -copies @var{n} ] [ -exec @var{executable} ]
2942 Adds @var{n} inferiors to be run using @var{executable} as the
2943 executable; @var{n} defaults to 1. If no executable is specified,
2944 the inferiors begins empty, with no program. You can still assign or
2945 change the program assigned to the inferior at any time by using the
2946 @code{file} command with the executable name as its argument.
2948 @kindex clone-inferior
2949 @item clone-inferior [ -copies @var{n} ] [ @var{infno} ]
2950 Adds @var{n} inferiors ready to execute the same program as inferior
2951 @var{infno}; @var{n} defaults to 1, and @var{infno} defaults to the
2952 number of the current inferior. This is a convenient command when you
2953 want to run another instance of the inferior you are debugging.
2956 (@value{GDBP}) info inferiors
2957 Num Description Executable
2958 * 1 process 29964 helloworld
2959 (@value{GDBP}) clone-inferior
2962 (@value{GDBP}) info inferiors
2963 Num Description Executable
2965 * 1 process 29964 helloworld
2968 You can now simply switch focus to inferior 2 and run it.
2970 @kindex remove-inferiors
2971 @item remove-inferiors @var{infno}@dots{}
2972 Removes the inferior or inferiors @var{infno}@dots{}. It is not
2973 possible to remove an inferior that is running with this command. For
2974 those, use the @code{kill} or @code{detach} command first.
2978 To quit debugging one of the running inferiors that is not the current
2979 inferior, you can either detach from it by using the @w{@code{detach
2980 inferior}} command (allowing it to run independently), or kill it
2981 using the @w{@code{kill inferiors}} command:
2984 @kindex detach inferiors @var{infno}@dots{}
2985 @item detach inferior @var{infno}@dots{}
2986 Detach from the inferior or inferiors identified by @value{GDBN}
2987 inferior number(s) @var{infno}@dots{}. Note that the inferior's entry
2988 still stays on the list of inferiors shown by @code{info inferiors},
2989 but its Description will show @samp{<null>}.
2991 @kindex kill inferiors @var{infno}@dots{}
2992 @item kill inferiors @var{infno}@dots{}
2993 Kill the inferior or inferiors identified by @value{GDBN} inferior
2994 number(s) @var{infno}@dots{}. Note that the inferior's entry still
2995 stays on the list of inferiors shown by @code{info inferiors}, but its
2996 Description will show @samp{<null>}.
2999 After the successful completion of a command such as @code{detach},
3000 @code{detach inferiors}, @code{kill} or @code{kill inferiors}, or after
3001 a normal process exit, the inferior is still valid and listed with
3002 @code{info inferiors}, ready to be restarted.
3005 To be notified when inferiors are started or exit under @value{GDBN}'s
3006 control use @w{@code{set print inferior-events}}:
3009 @kindex set print inferior-events
3010 @cindex print messages on inferior start and exit
3011 @item set print inferior-events
3012 @itemx set print inferior-events on
3013 @itemx set print inferior-events off
3014 The @code{set print inferior-events} command allows you to enable or
3015 disable printing of messages when @value{GDBN} notices that new
3016 inferiors have started or that inferiors have exited or have been
3017 detached. By default, these messages will not be printed.
3019 @kindex show print inferior-events
3020 @item show print inferior-events
3021 Show whether messages will be printed when @value{GDBN} detects that
3022 inferiors have started, exited or have been detached.
3025 Many commands will work the same with multiple programs as with a
3026 single program: e.g., @code{print myglobal} will simply display the
3027 value of @code{myglobal} in the current inferior.
3030 Occasionaly, when debugging @value{GDBN} itself, it may be useful to
3031 get more info about the relationship of inferiors, programs, address
3032 spaces in a debug session. You can do that with the @w{@code{maint
3033 info program-spaces}} command.
3036 @kindex maint info program-spaces
3037 @item maint info program-spaces
3038 Print a list of all program spaces currently being managed by
3041 @value{GDBN} displays for each program space (in this order):
3045 the program space number assigned by @value{GDBN}
3048 the name of the executable loaded into the program space, with e.g.,
3049 the @code{file} command.
3054 An asterisk @samp{*} preceding the @value{GDBN} program space number
3055 indicates the current program space.
3057 In addition, below each program space line, @value{GDBN} prints extra
3058 information that isn't suitable to display in tabular form. For
3059 example, the list of inferiors bound to the program space.
3062 (@value{GDBP}) maint info program-spaces
3066 Bound inferiors: ID 1 (process 21561)
3069 Here we can see that no inferior is running the program @code{hello},
3070 while @code{process 21561} is running the program @code{goodbye}. On
3071 some targets, it is possible that multiple inferiors are bound to the
3072 same program space. The most common example is that of debugging both
3073 the parent and child processes of a @code{vfork} call. For example,
3076 (@value{GDBP}) maint info program-spaces
3079 Bound inferiors: ID 2 (process 18050), ID 1 (process 18045)
3082 Here, both inferior 2 and inferior 1 are running in the same program
3083 space as a result of inferior 1 having executed a @code{vfork} call.
3087 @section Debugging Programs with Multiple Threads
3089 @cindex threads of execution
3090 @cindex multiple threads
3091 @cindex switching threads
3092 In some operating systems, such as GNU/Linux and Solaris, a single program
3093 may have more than one @dfn{thread} of execution. The precise semantics
3094 of threads differ from one operating system to another, but in general
3095 the threads of a single program are akin to multiple processes---except
3096 that they share one address space (that is, they can all examine and
3097 modify the same variables). On the other hand, each thread has its own
3098 registers and execution stack, and perhaps private memory.
3100 @value{GDBN} provides these facilities for debugging multi-thread
3104 @item automatic notification of new threads
3105 @item @samp{thread @var{thread-id}}, a command to switch among threads
3106 @item @samp{info threads}, a command to inquire about existing threads
3107 @item @samp{thread apply [@var{thread-id-list} | all] @var{args}},
3108 a command to apply a command to a list of threads
3109 @item thread-specific breakpoints
3110 @item @samp{set print thread-events}, which controls printing of
3111 messages on thread start and exit.
3112 @item @samp{set libthread-db-search-path @var{path}}, which lets
3113 the user specify which @code{libthread_db} to use if the default choice
3114 isn't compatible with the program.
3117 @cindex focus of debugging
3118 @cindex current thread
3119 The @value{GDBN} thread debugging facility allows you to observe all
3120 threads while your program runs---but whenever @value{GDBN} takes
3121 control, one thread in particular is always the focus of debugging.
3122 This thread is called the @dfn{current thread}. Debugging commands show
3123 program information from the perspective of the current thread.
3125 @cindex @code{New} @var{systag} message
3126 @cindex thread identifier (system)
3127 @c FIXME-implementors!! It would be more helpful if the [New...] message
3128 @c included GDB's numeric thread handle, so you could just go to that
3129 @c thread without first checking `info threads'.
3130 Whenever @value{GDBN} detects a new thread in your program, it displays
3131 the target system's identification for the thread with a message in the
3132 form @samp{[New @var{systag}]}, where @var{systag} is a thread identifier
3133 whose form varies depending on the particular system. For example, on
3134 @sc{gnu}/Linux, you might see
3137 [New Thread 0x41e02940 (LWP 25582)]
3141 when @value{GDBN} notices a new thread. In contrast, on other systems,
3142 the @var{systag} is simply something like @samp{process 368}, with no
3145 @c FIXME!! (1) Does the [New...] message appear even for the very first
3146 @c thread of a program, or does it only appear for the
3147 @c second---i.e.@: when it becomes obvious we have a multithread
3149 @c (2) *Is* there necessarily a first thread always? Or do some
3150 @c multithread systems permit starting a program with multiple
3151 @c threads ab initio?
3153 @anchor{thread numbers}
3154 @cindex thread number, per inferior
3155 @cindex thread identifier (GDB)
3156 For debugging purposes, @value{GDBN} associates its own thread number
3157 ---always a single integer---with each thread of an inferior. This
3158 number is unique between all threads of an inferior, but not unique
3159 between threads of different inferiors.
3161 @cindex qualified thread ID
3162 You can refer to a given thread in an inferior using the qualified
3163 @var{inferior-num}.@var{thread-num} syntax, also known as
3164 @dfn{qualified thread ID}, with @var{inferior-num} being the inferior
3165 number and @var{thread-num} being the thread number of the given
3166 inferior. For example, thread @code{2.3} refers to thread number 3 of
3167 inferior 2. If you omit @var{inferior-num} (e.g., @code{thread 3}),
3168 then @value{GDBN} infers you're referring to a thread of the current
3171 Until you create a second inferior, @value{GDBN} does not show the
3172 @var{inferior-num} part of thread IDs, even though you can always use
3173 the full @var{inferior-num}.@var{thread-num} form to refer to threads
3174 of inferior 1, the initial inferior.
3176 @anchor{thread ID lists}
3177 @cindex thread ID lists
3178 Some commands accept a space-separated @dfn{thread ID list} as
3179 argument. A list element can be:
3183 A thread ID as shown in the first field of the @samp{info threads}
3184 display, with or without an inferior qualifier. E.g., @samp{2.1} or
3188 A range of thread numbers, again with or without an inferior
3189 qualifier, as in @var{inf}.@var{thr1}-@var{thr2} or
3190 @var{thr1}-@var{thr2}. E.g., @samp{1.2-4} or @samp{2-4}.
3193 All threads of an inferior, specified with a star wildcard, with or
3194 without an inferior qualifier, as in @var{inf}.@code{*} (e.g.,
3195 @samp{1.*}) or @code{*}. The former refers to all threads of the
3196 given inferior, and the latter form without an inferior qualifier
3197 refers to all threads of the current inferior.
3201 For example, if the current inferior is 1, and inferior 7 has one
3202 thread with ID 7.1, the thread list @samp{1 2-3 4.5 6.7-9 7.*}
3203 includes threads 1 to 3 of inferior 1, thread 5 of inferior 4, threads
3204 7 to 9 of inferior 6 and all threads of inferior 7. That is, in
3205 expanded qualified form, the same as @samp{1.1 1.2 1.3 4.5 6.7 6.8 6.9
3209 @anchor{global thread numbers}
3210 @cindex global thread number
3211 @cindex global thread identifier (GDB)
3212 In addition to a @emph{per-inferior} number, each thread is also
3213 assigned a unique @emph{global} number, also known as @dfn{global
3214 thread ID}, a single integer. Unlike the thread number component of
3215 the thread ID, no two threads have the same global ID, even when
3216 you're debugging multiple inferiors.
3218 From @value{GDBN}'s perspective, a process always has at least one
3219 thread. In other words, @value{GDBN} assigns a thread number to the
3220 program's ``main thread'' even if the program is not multi-threaded.
3222 @vindex $_thread@r{, convenience variable}
3223 @vindex $_gthread@r{, convenience variable}
3224 The debugger convenience variables @samp{$_thread} and
3225 @samp{$_gthread} contain, respectively, the per-inferior thread number
3226 and the global thread number of the current thread. You may find this
3227 useful in writing breakpoint conditional expressions, command scripts,
3228 and so forth. @xref{Convenience Vars,, Convenience Variables}, for
3229 general information on convenience variables.
3231 If @value{GDBN} detects the program is multi-threaded, it augments the
3232 usual message about stopping at a breakpoint with the ID and name of
3233 the thread that hit the breakpoint.
3236 Thread 2 "client" hit Breakpoint 1, send_message () at client.c:68
3239 Likewise when the program receives a signal:
3242 Thread 1 "main" received signal SIGINT, Interrupt.
3246 @kindex info threads
3247 @item info threads @r{[}@var{thread-id-list}@r{]}
3249 Display information about one or more threads. With no arguments
3250 displays information about all threads. You can specify the list of
3251 threads that you want to display using the thread ID list syntax
3252 (@pxref{thread ID lists}).
3254 @value{GDBN} displays for each thread (in this order):
3258 the per-inferior thread number assigned by @value{GDBN}
3261 the global thread number assigned by @value{GDBN}, if the @samp{-gid}
3262 option was specified
3265 the target system's thread identifier (@var{systag})
3268 the thread's name, if one is known. A thread can either be named by
3269 the user (see @code{thread name}, below), or, in some cases, by the
3273 the current stack frame summary for that thread
3277 An asterisk @samp{*} to the left of the @value{GDBN} thread number
3278 indicates the current thread.
3282 @c end table here to get a little more width for example
3285 (@value{GDBP}) info threads
3287 * 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3288 2 process 35 thread 23 0x34e5 in sigpause ()
3289 3 process 35 thread 27 0x34e5 in sigpause ()
3293 If you're debugging multiple inferiors, @value{GDBN} displays thread
3294 IDs using the qualified @var{inferior-num}.@var{thread-num} format.
3295 Otherwise, only @var{thread-num} is shown.
3297 If you specify the @samp{-gid} option, @value{GDBN} displays a column
3298 indicating each thread's global thread ID:
3301 (@value{GDBP}) info threads
3302 Id GId Target Id Frame
3303 1.1 1 process 35 thread 13 main (argc=1, argv=0x7ffffff8)
3304 1.2 3 process 35 thread 23 0x34e5 in sigpause ()
3305 1.3 4 process 35 thread 27 0x34e5 in sigpause ()
3306 * 2.1 2 process 65 thread 1 main (argc=1, argv=0x7ffffff8)
3309 On Solaris, you can display more information about user threads with a
3310 Solaris-specific command:
3313 @item maint info sol-threads
3314 @kindex maint info sol-threads
3315 @cindex thread info (Solaris)
3316 Display info on Solaris user threads.
3320 @kindex thread @var{thread-id}
3321 @item thread @var{thread-id}
3322 Make thread ID @var{thread-id} the current thread. The command
3323 argument @var{thread-id} is the @value{GDBN} thread ID, as shown in
3324 the first field of the @samp{info threads} display, with or without an
3325 inferior qualifier (e.g., @samp{2.1} or @samp{1}).
3327 @value{GDBN} responds by displaying the system identifier of the
3328 thread you selected, and its current stack frame summary:
3331 (@value{GDBP}) thread 2
3332 [Switching to thread 2 (Thread 0xb7fdab70 (LWP 12747))]
3333 #0 some_function (ignore=0x0) at example.c:8
3334 8 printf ("hello\n");
3338 As with the @samp{[New @dots{}]} message, the form of the text after
3339 @samp{Switching to} depends on your system's conventions for identifying
3342 @anchor{thread apply all}
3343 @kindex thread apply
3344 @cindex apply command to several threads
3345 @item thread apply [@var{thread-id-list} | all [-ascending]] [@var{flag}]@dots{} @var{command}
3346 The @code{thread apply} command allows you to apply the named
3347 @var{command} to one or more threads. Specify the threads that you
3348 want affected using the thread ID list syntax (@pxref{thread ID
3349 lists}), or specify @code{all} to apply to all threads. To apply a
3350 command to all threads in descending order, type @kbd{thread apply all
3351 @var{command}}. To apply a command to all threads in ascending order,
3352 type @kbd{thread apply all -ascending @var{command}}.
3354 The @var{flag} arguments control what output to produce and how to handle
3355 errors raised when applying @var{command} to a thread. @var{flag}
3356 must start with a @code{-} directly followed by one letter in
3357 @code{qcs}. If several flags are provided, they must be given
3358 individually, such as @code{-c -q}.
3360 By default, @value{GDBN} displays some thread information before the
3361 output produced by @var{command}, and an error raised during the
3362 execution of a @var{command} will abort @code{thread apply}. The
3363 following flags can be used to fine-tune this behavior:
3367 The flag @code{-c}, which stands for @samp{continue}, causes any
3368 errors in @var{command} to be displayed, and the execution of
3369 @code{thread apply} then continues.
3371 The flag @code{-s}, which stands for @samp{silent}, causes any errors
3372 or empty output produced by a @var{command} to be silently ignored.
3373 That is, the execution continues, but the thread information and errors
3376 The flag @code{-q} (@samp{quiet}) disables printing the thread
3380 Flags @code{-c} and @code{-s} cannot be used together.
3383 @cindex apply command to all threads (ignoring errors and empty output)
3384 @item taas [@var{option}]@dots{} @var{command}
3385 Shortcut for @code{thread apply all -s [@var{option}]@dots{} @var{command}}.
3386 Applies @var{command} on all threads, ignoring errors and empty output.
3388 The @code{taas} command accepts the same options as the @code{thread
3389 apply all} command. @xref{thread apply all}.
3392 @cindex apply a command to all frames of all threads (ignoring errors and empty output)
3393 @item tfaas [@var{option}]@dots{} @var{command}
3394 Shortcut for @code{thread apply all -s -- frame apply all -s [@var{option}]@dots{} @var{command}}.
3395 Applies @var{command} on all frames of all threads, ignoring errors
3396 and empty output. Note that the flag @code{-s} is specified twice:
3397 The first @code{-s} ensures that @code{thread apply} only shows the thread
3398 information of the threads for which @code{frame apply} produces
3399 some output. The second @code{-s} is needed to ensure that @code{frame
3400 apply} shows the frame information of a frame only if the
3401 @var{command} successfully produced some output.
3403 It can for example be used to print a local variable or a function
3404 argument without knowing the thread or frame where this variable or argument
3407 (@value{GDBP}) tfaas p some_local_var_i_do_not_remember_where_it_is
3410 The @code{tfaas} command accepts the same options as the @code{frame
3411 apply} command. @xref{frame apply}.
3414 @cindex name a thread
3415 @item thread name [@var{name}]
3416 This command assigns a name to the current thread. If no argument is
3417 given, any existing user-specified name is removed. The thread name
3418 appears in the @samp{info threads} display.
3420 On some systems, such as @sc{gnu}/Linux, @value{GDBN} is able to
3421 determine the name of the thread as given by the OS. On these
3422 systems, a name specified with @samp{thread name} will override the
3423 system-give name, and removing the user-specified name will cause
3424 @value{GDBN} to once again display the system-specified name.
3427 @cindex search for a thread
3428 @item thread find [@var{regexp}]
3429 Search for and display thread ids whose name or @var{systag}
3430 matches the supplied regular expression.
3432 As well as being the complement to the @samp{thread name} command,
3433 this command also allows you to identify a thread by its target
3434 @var{systag}. For instance, on @sc{gnu}/Linux, the target @var{systag}
3438 (@value{GDBN}) thread find 26688
3439 Thread 4 has target id 'Thread 0x41e02940 (LWP 26688)'
3440 (@value{GDBN}) info thread 4
3442 4 Thread 0x41e02940 (LWP 26688) 0x00000031ca6cd372 in select ()
3445 @kindex set print thread-events
3446 @cindex print messages on thread start and exit
3447 @item set print thread-events
3448 @itemx set print thread-events on
3449 @itemx set print thread-events off
3450 The @code{set print thread-events} command allows you to enable or
3451 disable printing of messages when @value{GDBN} notices that new threads have
3452 started or that threads have exited. By default, these messages will
3453 be printed if detection of these events is supported by the target.
3454 Note that these messages cannot be disabled on all targets.
3456 @kindex show print thread-events
3457 @item show print thread-events
3458 Show whether messages will be printed when @value{GDBN} detects that threads
3459 have started and exited.
3462 @xref{Thread Stops,,Stopping and Starting Multi-thread Programs}, for
3463 more information about how @value{GDBN} behaves when you stop and start
3464 programs with multiple threads.
3466 @xref{Set Watchpoints,,Setting Watchpoints}, for information about
3467 watchpoints in programs with multiple threads.
3469 @anchor{set libthread-db-search-path}
3471 @kindex set libthread-db-search-path
3472 @cindex search path for @code{libthread_db}
3473 @item set libthread-db-search-path @r{[}@var{path}@r{]}
3474 If this variable is set, @var{path} is a colon-separated list of
3475 directories @value{GDBN} will use to search for @code{libthread_db}.
3476 If you omit @var{path}, @samp{libthread-db-search-path} will be reset to
3477 its default value (@code{$sdir:$pdir} on @sc{gnu}/Linux and Solaris systems).
3478 Internally, the default value comes from the @code{LIBTHREAD_DB_SEARCH_PATH}
3481 On @sc{gnu}/Linux and Solaris systems, @value{GDBN} uses a ``helper''
3482 @code{libthread_db} library to obtain information about threads in the
3483 inferior process. @value{GDBN} will use @samp{libthread-db-search-path}
3484 to find @code{libthread_db}. @value{GDBN} also consults first if inferior
3485 specific thread debugging library loading is enabled
3486 by @samp{set auto-load libthread-db} (@pxref{libthread_db.so.1 file}).
3488 A special entry @samp{$sdir} for @samp{libthread-db-search-path}
3489 refers to the default system directories that are
3490 normally searched for loading shared libraries. The @samp{$sdir} entry
3491 is the only kind not needing to be enabled by @samp{set auto-load libthread-db}
3492 (@pxref{libthread_db.so.1 file}).
3494 A special entry @samp{$pdir} for @samp{libthread-db-search-path}
3495 refers to the directory from which @code{libpthread}
3496 was loaded in the inferior process.
3498 For any @code{libthread_db} library @value{GDBN} finds in above directories,
3499 @value{GDBN} attempts to initialize it with the current inferior process.
3500 If this initialization fails (which could happen because of a version
3501 mismatch between @code{libthread_db} and @code{libpthread}), @value{GDBN}
3502 will unload @code{libthread_db}, and continue with the next directory.
3503 If none of @code{libthread_db} libraries initialize successfully,
3504 @value{GDBN} will issue a warning and thread debugging will be disabled.
3506 Setting @code{libthread-db-search-path} is currently implemented
3507 only on some platforms.
3509 @kindex show libthread-db-search-path
3510 @item show libthread-db-search-path
3511 Display current libthread_db search path.
3513 @kindex set debug libthread-db
3514 @kindex show debug libthread-db
3515 @cindex debugging @code{libthread_db}
3516 @item set debug libthread-db
3517 @itemx show debug libthread-db
3518 Turns on or off display of @code{libthread_db}-related events.
3519 Use @code{1} to enable, @code{0} to disable.
3523 @section Debugging Forks
3525 @cindex fork, debugging programs which call
3526 @cindex multiple processes
3527 @cindex processes, multiple
3528 On most systems, @value{GDBN} has no special support for debugging
3529 programs which create additional processes using the @code{fork}
3530 function. When a program forks, @value{GDBN} will continue to debug the
3531 parent process and the child process will run unimpeded. If you have
3532 set a breakpoint in any code which the child then executes, the child
3533 will get a @code{SIGTRAP} signal which (unless it catches the signal)
3534 will cause it to terminate.
3536 However, if you want to debug the child process there is a workaround
3537 which isn't too painful. Put a call to @code{sleep} in the code which
3538 the child process executes after the fork. It may be useful to sleep
3539 only if a certain environment variable is set, or a certain file exists,
3540 so that the delay need not occur when you don't want to run @value{GDBN}
3541 on the child. While the child is sleeping, use the @code{ps} program to
3542 get its process ID. Then tell @value{GDBN} (a new invocation of
3543 @value{GDBN} if you are also debugging the parent process) to attach to
3544 the child process (@pxref{Attach}). From that point on you can debug
3545 the child process just like any other process which you attached to.
3547 On some systems, @value{GDBN} provides support for debugging programs
3548 that create additional processes using the @code{fork} or @code{vfork}
3549 functions. On @sc{gnu}/Linux platforms, this feature is supported
3550 with kernel version 2.5.46 and later.
3552 The fork debugging commands are supported in native mode and when
3553 connected to @code{gdbserver} in either @code{target remote} mode or
3554 @code{target extended-remote} mode.
3556 By default, when a program forks, @value{GDBN} will continue to debug
3557 the parent process and the child process will run unimpeded.
3559 If you want to follow the child process instead of the parent process,
3560 use the command @w{@code{set follow-fork-mode}}.
3563 @kindex set follow-fork-mode
3564 @item set follow-fork-mode @var{mode}
3565 Set the debugger response to a program call of @code{fork} or
3566 @code{vfork}. A call to @code{fork} or @code{vfork} creates a new
3567 process. The @var{mode} argument can be:
3571 The original process is debugged after a fork. The child process runs
3572 unimpeded. This is the default.
3575 The new process is debugged after a fork. The parent process runs
3580 @kindex show follow-fork-mode
3581 @item show follow-fork-mode
3582 Display the current debugger response to a @code{fork} or @code{vfork} call.
3585 @cindex debugging multiple processes
3586 On Linux, if you want to debug both the parent and child processes, use the
3587 command @w{@code{set detach-on-fork}}.
3590 @kindex set detach-on-fork
3591 @item set detach-on-fork @var{mode}
3592 Tells gdb whether to detach one of the processes after a fork, or
3593 retain debugger control over them both.
3597 The child process (or parent process, depending on the value of
3598 @code{follow-fork-mode}) will be detached and allowed to run
3599 independently. This is the default.
3602 Both processes will be held under the control of @value{GDBN}.
3603 One process (child or parent, depending on the value of
3604 @code{follow-fork-mode}) is debugged as usual, while the other
3609 @kindex show detach-on-fork
3610 @item show detach-on-fork
3611 Show whether detach-on-fork mode is on/off.
3614 If you choose to set @samp{detach-on-fork} mode off, then @value{GDBN}
3615 will retain control of all forked processes (including nested forks).
3616 You can list the forked processes under the control of @value{GDBN} by
3617 using the @w{@code{info inferiors}} command, and switch from one fork
3618 to another by using the @code{inferior} command (@pxref{Inferiors and
3619 Programs, ,Debugging Multiple Inferiors and Programs}).
3621 To quit debugging one of the forked processes, you can either detach
3622 from it by using the @w{@code{detach inferiors}} command (allowing it
3623 to run independently), or kill it using the @w{@code{kill inferiors}}
3624 command. @xref{Inferiors and Programs, ,Debugging Multiple Inferiors
3627 If you ask to debug a child process and a @code{vfork} is followed by an
3628 @code{exec}, @value{GDBN} executes the new target up to the first
3629 breakpoint in the new target. If you have a breakpoint set on
3630 @code{main} in your original program, the breakpoint will also be set on
3631 the child process's @code{main}.
3633 On some systems, when a child process is spawned by @code{vfork}, you
3634 cannot debug the child or parent until an @code{exec} call completes.
3636 If you issue a @code{run} command to @value{GDBN} after an @code{exec}
3637 call executes, the new target restarts. To restart the parent
3638 process, use the @code{file} command with the parent executable name
3639 as its argument. By default, after an @code{exec} call executes,
3640 @value{GDBN} discards the symbols of the previous executable image.
3641 You can change this behaviour with the @w{@code{set follow-exec-mode}}
3645 @kindex set follow-exec-mode
3646 @item set follow-exec-mode @var{mode}
3648 Set debugger response to a program call of @code{exec}. An
3649 @code{exec} call replaces the program image of a process.
3651 @code{follow-exec-mode} can be:
3655 @value{GDBN} creates a new inferior and rebinds the process to this
3656 new inferior. The program the process was running before the
3657 @code{exec} call can be restarted afterwards by restarting the
3663 (@value{GDBP}) info inferiors
3665 Id Description Executable
3668 process 12020 is executing new program: prog2
3669 Program exited normally.
3670 (@value{GDBP}) info inferiors
3671 Id Description Executable
3677 @value{GDBN} keeps the process bound to the same inferior. The new
3678 executable image replaces the previous executable loaded in the
3679 inferior. Restarting the inferior after the @code{exec} call, with
3680 e.g., the @code{run} command, restarts the executable the process was
3681 running after the @code{exec} call. This is the default mode.
3686 (@value{GDBP}) info inferiors
3687 Id Description Executable
3690 process 12020 is executing new program: prog2
3691 Program exited normally.
3692 (@value{GDBP}) info inferiors
3693 Id Description Executable
3700 @code{follow-exec-mode} is supported in native mode and
3701 @code{target extended-remote} mode.
3703 You can use the @code{catch} command to make @value{GDBN} stop whenever
3704 a @code{fork}, @code{vfork}, or @code{exec} call is made. @xref{Set
3705 Catchpoints, ,Setting Catchpoints}.
3707 @node Checkpoint/Restart
3708 @section Setting a @emph{Bookmark} to Return to Later
3713 @cindex snapshot of a process
3714 @cindex rewind program state
3716 On certain operating systems@footnote{Currently, only
3717 @sc{gnu}/Linux.}, @value{GDBN} is able to save a @dfn{snapshot} of a
3718 program's state, called a @dfn{checkpoint}, and come back to it
3721 Returning to a checkpoint effectively undoes everything that has
3722 happened in the program since the @code{checkpoint} was saved. This
3723 includes changes in memory, registers, and even (within some limits)
3724 system state. Effectively, it is like going back in time to the
3725 moment when the checkpoint was saved.
3727 Thus, if you're stepping thru a program and you think you're
3728 getting close to the point where things go wrong, you can save
3729 a checkpoint. Then, if you accidentally go too far and miss
3730 the critical statement, instead of having to restart your program
3731 from the beginning, you can just go back to the checkpoint and
3732 start again from there.
3734 This can be especially useful if it takes a lot of time or
3735 steps to reach the point where you think the bug occurs.
3737 To use the @code{checkpoint}/@code{restart} method of debugging:
3742 Save a snapshot of the debugged program's current execution state.
3743 The @code{checkpoint} command takes no arguments, but each checkpoint
3744 is assigned a small integer id, similar to a breakpoint id.
3746 @kindex info checkpoints
3747 @item info checkpoints
3748 List the checkpoints that have been saved in the current debugging
3749 session. For each checkpoint, the following information will be
3756 @item Source line, or label
3759 @kindex restart @var{checkpoint-id}
3760 @item restart @var{checkpoint-id}
3761 Restore the program state that was saved as checkpoint number
3762 @var{checkpoint-id}. All program variables, registers, stack frames
3763 etc.@: will be returned to the values that they had when the checkpoint
3764 was saved. In essence, gdb will ``wind back the clock'' to the point
3765 in time when the checkpoint was saved.
3767 Note that breakpoints, @value{GDBN} variables, command history etc.
3768 are not affected by restoring a checkpoint. In general, a checkpoint
3769 only restores things that reside in the program being debugged, not in
3772 @kindex delete checkpoint @var{checkpoint-id}
3773 @item delete checkpoint @var{checkpoint-id}
3774 Delete the previously-saved checkpoint identified by @var{checkpoint-id}.
3778 Returning to a previously saved checkpoint will restore the user state
3779 of the program being debugged, plus a significant subset of the system
3780 (OS) state, including file pointers. It won't ``un-write'' data from
3781 a file, but it will rewind the file pointer to the previous location,
3782 so that the previously written data can be overwritten. For files
3783 opened in read mode, the pointer will also be restored so that the
3784 previously read data can be read again.
3786 Of course, characters that have been sent to a printer (or other
3787 external device) cannot be ``snatched back'', and characters received
3788 from eg.@: a serial device can be removed from internal program buffers,
3789 but they cannot be ``pushed back'' into the serial pipeline, ready to
3790 be received again. Similarly, the actual contents of files that have
3791 been changed cannot be restored (at this time).
3793 However, within those constraints, you actually can ``rewind'' your
3794 program to a previously saved point in time, and begin debugging it
3795 again --- and you can change the course of events so as to debug a
3796 different execution path this time.
3798 @cindex checkpoints and process id
3799 Finally, there is one bit of internal program state that will be
3800 different when you return to a checkpoint --- the program's process
3801 id. Each checkpoint will have a unique process id (or @var{pid}),
3802 and each will be different from the program's original @var{pid}.
3803 If your program has saved a local copy of its process id, this could
3804 potentially pose a problem.
3806 @subsection A Non-obvious Benefit of Using Checkpoints
3808 On some systems such as @sc{gnu}/Linux, address space randomization
3809 is performed on new processes for security reasons. This makes it
3810 difficult or impossible to set a breakpoint, or watchpoint, on an
3811 absolute address if you have to restart the program, since the
3812 absolute location of a symbol will change from one execution to the
3815 A checkpoint, however, is an @emph{identical} copy of a process.
3816 Therefore if you create a checkpoint at (eg.@:) the start of main,
3817 and simply return to that checkpoint instead of restarting the
3818 process, you can avoid the effects of address randomization and
3819 your symbols will all stay in the same place.
3822 @chapter Stopping and Continuing
3824 The principal purposes of using a debugger are so that you can stop your
3825 program before it terminates; or so that, if your program runs into
3826 trouble, you can investigate and find out why.
3828 Inside @value{GDBN}, your program may stop for any of several reasons,
3829 such as a signal, a breakpoint, or reaching a new line after a
3830 @value{GDBN} command such as @code{step}. You may then examine and
3831 change variables, set new breakpoints or remove old ones, and then
3832 continue execution. Usually, the messages shown by @value{GDBN} provide
3833 ample explanation of the status of your program---but you can also
3834 explicitly request this information at any time.
3837 @kindex info program
3839 Display information about the status of your program: whether it is
3840 running or not, what process it is, and why it stopped.
3844 * Breakpoints:: Breakpoints, watchpoints, and catchpoints
3845 * Continuing and Stepping:: Resuming execution
3846 * Skipping Over Functions and Files::
3847 Skipping over functions and files
3849 * Thread Stops:: Stopping and starting multi-thread programs
3853 @section Breakpoints, Watchpoints, and Catchpoints
3856 A @dfn{breakpoint} makes your program stop whenever a certain point in
3857 the program is reached. For each breakpoint, you can add conditions to
3858 control in finer detail whether your program stops. You can set
3859 breakpoints with the @code{break} command and its variants (@pxref{Set
3860 Breaks, ,Setting Breakpoints}), to specify the place where your program
3861 should stop by line number, function name or exact address in the
3864 On some systems, you can set breakpoints in shared libraries before
3865 the executable is run.
3868 @cindex data breakpoints
3869 @cindex memory tracing
3870 @cindex breakpoint on memory address
3871 @cindex breakpoint on variable modification
3872 A @dfn{watchpoint} is a special breakpoint that stops your program
3873 when the value of an expression changes. The expression may be a value
3874 of a variable, or it could involve values of one or more variables
3875 combined by operators, such as @samp{a + b}. This is sometimes called
3876 @dfn{data breakpoints}. You must use a different command to set
3877 watchpoints (@pxref{Set Watchpoints, ,Setting Watchpoints}), but aside
3878 from that, you can manage a watchpoint like any other breakpoint: you
3879 enable, disable, and delete both breakpoints and watchpoints using the
3882 You can arrange to have values from your program displayed automatically
3883 whenever @value{GDBN} stops at a breakpoint. @xref{Auto Display,,
3887 @cindex breakpoint on events
3888 A @dfn{catchpoint} is another special breakpoint that stops your program
3889 when a certain kind of event occurs, such as the throwing of a C@t{++}
3890 exception or the loading of a library. As with watchpoints, you use a
3891 different command to set a catchpoint (@pxref{Set Catchpoints, ,Setting
3892 Catchpoints}), but aside from that, you can manage a catchpoint like any
3893 other breakpoint. (To stop when your program receives a signal, use the
3894 @code{handle} command; see @ref{Signals, ,Signals}.)
3896 @cindex breakpoint numbers
3897 @cindex numbers for breakpoints
3898 @value{GDBN} assigns a number to each breakpoint, watchpoint, or
3899 catchpoint when you create it; these numbers are successive integers
3900 starting with one. In many of the commands for controlling various
3901 features of breakpoints you use the breakpoint number to say which
3902 breakpoint you want to change. Each breakpoint may be @dfn{enabled} or
3903 @dfn{disabled}; if disabled, it has no effect on your program until you
3906 @cindex breakpoint ranges
3907 @cindex breakpoint lists
3908 @cindex ranges of breakpoints
3909 @cindex lists of breakpoints
3910 Some @value{GDBN} commands accept a space-separated list of breakpoints
3911 on which to operate. A list element can be either a single breakpoint number,
3912 like @samp{5}, or a range of such numbers, like @samp{5-7}.
3913 When a breakpoint list is given to a command, all breakpoints in that list
3917 * Set Breaks:: Setting breakpoints
3918 * Set Watchpoints:: Setting watchpoints
3919 * Set Catchpoints:: Setting catchpoints
3920 * Delete Breaks:: Deleting breakpoints
3921 * Disabling:: Disabling breakpoints
3922 * Conditions:: Break conditions
3923 * Break Commands:: Breakpoint command lists
3924 * Dynamic Printf:: Dynamic printf
3925 * Save Breakpoints:: How to save breakpoints in a file
3926 * Static Probe Points:: Listing static probe points
3927 * Error in Breakpoints:: ``Cannot insert breakpoints''
3928 * Breakpoint-related Warnings:: ``Breakpoint address adjusted...''
3932 @subsection Setting Breakpoints
3934 @c FIXME LMB what does GDB do if no code on line of breakpt?
3935 @c consider in particular declaration with/without initialization.
3937 @c FIXME 2 is there stuff on this already? break at fun start, already init?
3940 @kindex b @r{(@code{break})}
3941 @vindex $bpnum@r{, convenience variable}
3942 @cindex latest breakpoint
3943 Breakpoints are set with the @code{break} command (abbreviated
3944 @code{b}). The debugger convenience variable @samp{$bpnum} records the
3945 number of the breakpoint you've set most recently; see @ref{Convenience
3946 Vars,, Convenience Variables}, for a discussion of what you can do with
3947 convenience variables.
3950 @item break @var{location}
3951 Set a breakpoint at the given @var{location}, which can specify a
3952 function name, a line number, or an address of an instruction.
3953 (@xref{Specify Location}, for a list of all the possible ways to
3954 specify a @var{location}.) The breakpoint will stop your program just
3955 before it executes any of the code in the specified @var{location}.
3957 When using source languages that permit overloading of symbols, such as
3958 C@t{++}, a function name may refer to more than one possible place to break.
3959 @xref{Ambiguous Expressions,,Ambiguous Expressions}, for a discussion of
3962 It is also possible to insert a breakpoint that will stop the program
3963 only if a specific thread (@pxref{Thread-Specific Breakpoints})
3964 or a specific task (@pxref{Ada Tasks}) hits that breakpoint.
3967 When called without any arguments, @code{break} sets a breakpoint at
3968 the next instruction to be executed in the selected stack frame
3969 (@pxref{Stack, ,Examining the Stack}). In any selected frame but the
3970 innermost, this makes your program stop as soon as control
3971 returns to that frame. This is similar to the effect of a
3972 @code{finish} command in the frame inside the selected frame---except
3973 that @code{finish} does not leave an active breakpoint. If you use
3974 @code{break} without an argument in the innermost frame, @value{GDBN} stops
3975 the next time it reaches the current location; this may be useful
3978 @value{GDBN} normally ignores breakpoints when it resumes execution, until at
3979 least one instruction has been executed. If it did not do this, you
3980 would be unable to proceed past a breakpoint without first disabling the
3981 breakpoint. This rule applies whether or not the breakpoint already
3982 existed when your program stopped.
3984 @item break @dots{} if @var{cond}
3985 Set a breakpoint with condition @var{cond}; evaluate the expression
3986 @var{cond} each time the breakpoint is reached, and stop only if the
3987 value is nonzero---that is, if @var{cond} evaluates as true.
3988 @samp{@dots{}} stands for one of the possible arguments described
3989 above (or no argument) specifying where to break. @xref{Conditions,
3990 ,Break Conditions}, for more information on breakpoint conditions.
3993 @item tbreak @var{args}
3994 Set a breakpoint enabled only for one stop. The @var{args} are the
3995 same as for the @code{break} command, and the breakpoint is set in the same
3996 way, but the breakpoint is automatically deleted after the first time your
3997 program stops there. @xref{Disabling, ,Disabling Breakpoints}.
4000 @cindex hardware breakpoints
4001 @item hbreak @var{args}
4002 Set a hardware-assisted breakpoint. The @var{args} are the same as for the
4003 @code{break} command and the breakpoint is set in the same way, but the
4004 breakpoint requires hardware support and some target hardware may not
4005 have this support. The main purpose of this is EPROM/ROM code
4006 debugging, so you can set a breakpoint at an instruction without
4007 changing the instruction. This can be used with the new trap-generation
4008 provided by SPARClite DSU and most x86-based targets. These targets
4009 will generate traps when a program accesses some data or instruction
4010 address that is assigned to the debug registers. However the hardware
4011 breakpoint registers can take a limited number of breakpoints. For
4012 example, on the DSU, only two data breakpoints can be set at a time, and
4013 @value{GDBN} will reject this command if more than two are used. Delete
4014 or disable unused hardware breakpoints before setting new ones
4015 (@pxref{Disabling, ,Disabling Breakpoints}).
4016 @xref{Conditions, ,Break Conditions}.
4017 For remote targets, you can restrict the number of hardware
4018 breakpoints @value{GDBN} will use, see @ref{set remote
4019 hardware-breakpoint-limit}.
4022 @item thbreak @var{args}
4023 Set a hardware-assisted breakpoint enabled only for one stop. The @var{args}
4024 are the same as for the @code{hbreak} command and the breakpoint is set in
4025 the same way. However, like the @code{tbreak} command,
4026 the breakpoint is automatically deleted after the
4027 first time your program stops there. Also, like the @code{hbreak}
4028 command, the breakpoint requires hardware support and some target hardware
4029 may not have this support. @xref{Disabling, ,Disabling Breakpoints}.
4030 See also @ref{Conditions, ,Break Conditions}.
4033 @cindex regular expression
4034 @cindex breakpoints at functions matching a regexp
4035 @cindex set breakpoints in many functions
4036 @item rbreak @var{regex}
4037 Set breakpoints on all functions matching the regular expression
4038 @var{regex}. This command sets an unconditional breakpoint on all
4039 matches, printing a list of all breakpoints it set. Once these
4040 breakpoints are set, they are treated just like the breakpoints set with
4041 the @code{break} command. You can delete them, disable them, or make
4042 them conditional the same way as any other breakpoint.
4044 In programs using different languages, @value{GDBN} chooses the syntax
4045 to print the list of all breakpoints it sets according to the
4046 @samp{set language} value: using @samp{set language auto}
4047 (see @ref{Automatically, ,Set Language Automatically}) means to use the
4048 language of the breakpoint's function, other values mean to use
4049 the manually specified language (see @ref{Manually, ,Set Language Manually}).
4051 The syntax of the regular expression is the standard one used with tools
4052 like @file{grep}. Note that this is different from the syntax used by
4053 shells, so for instance @code{foo*} matches all functions that include
4054 an @code{fo} followed by zero or more @code{o}s. There is an implicit
4055 @code{.*} leading and trailing the regular expression you supply, so to
4056 match only functions that begin with @code{foo}, use @code{^foo}.
4058 @cindex non-member C@t{++} functions, set breakpoint in
4059 When debugging C@t{++} programs, @code{rbreak} is useful for setting
4060 breakpoints on overloaded functions that are not members of any special
4063 @cindex set breakpoints on all functions
4064 The @code{rbreak} command can be used to set breakpoints in
4065 @strong{all} the functions in a program, like this:
4068 (@value{GDBP}) rbreak .
4071 @item rbreak @var{file}:@var{regex}
4072 If @code{rbreak} is called with a filename qualification, it limits
4073 the search for functions matching the given regular expression to the
4074 specified @var{file}. This can be used, for example, to set breakpoints on
4075 every function in a given file:
4078 (@value{GDBP}) rbreak file.c:.
4081 The colon separating the filename qualifier from the regex may
4082 optionally be surrounded by spaces.
4084 @kindex info breakpoints
4085 @cindex @code{$_} and @code{info breakpoints}
4086 @item info breakpoints @r{[}@var{list}@dots{}@r{]}
4087 @itemx info break @r{[}@var{list}@dots{}@r{]}
4088 Print a table of all breakpoints, watchpoints, and catchpoints set and
4089 not deleted. Optional argument @var{n} means print information only
4090 about the specified breakpoint(s) (or watchpoint(s) or catchpoint(s)).
4091 For each breakpoint, following columns are printed:
4094 @item Breakpoint Numbers
4096 Breakpoint, watchpoint, or catchpoint.
4098 Whether the breakpoint is marked to be disabled or deleted when hit.
4099 @item Enabled or Disabled
4100 Enabled breakpoints are marked with @samp{y}. @samp{n} marks breakpoints
4101 that are not enabled.
4103 Where the breakpoint is in your program, as a memory address. For a
4104 pending breakpoint whose address is not yet known, this field will
4105 contain @samp{<PENDING>}. Such breakpoint won't fire until a shared
4106 library that has the symbol or line referred by breakpoint is loaded.
4107 See below for details. A breakpoint with several locations will
4108 have @samp{<MULTIPLE>} in this field---see below for details.
4110 Where the breakpoint is in the source for your program, as a file and
4111 line number. For a pending breakpoint, the original string passed to
4112 the breakpoint command will be listed as it cannot be resolved until
4113 the appropriate shared library is loaded in the future.
4117 If a breakpoint is conditional, there are two evaluation modes: ``host'' and
4118 ``target''. If mode is ``host'', breakpoint condition evaluation is done by
4119 @value{GDBN} on the host's side. If it is ``target'', then the condition
4120 is evaluated by the target. The @code{info break} command shows
4121 the condition on the line following the affected breakpoint, together with
4122 its condition evaluation mode in between parentheses.
4124 Breakpoint commands, if any, are listed after that. A pending breakpoint is
4125 allowed to have a condition specified for it. The condition is not parsed for
4126 validity until a shared library is loaded that allows the pending
4127 breakpoint to resolve to a valid location.
4130 @code{info break} with a breakpoint
4131 number @var{n} as argument lists only that breakpoint. The
4132 convenience variable @code{$_} and the default examining-address for
4133 the @code{x} command are set to the address of the last breakpoint
4134 listed (@pxref{Memory, ,Examining Memory}).
4137 @code{info break} displays a count of the number of times the breakpoint
4138 has been hit. This is especially useful in conjunction with the
4139 @code{ignore} command. You can ignore a large number of breakpoint
4140 hits, look at the breakpoint info to see how many times the breakpoint
4141 was hit, and then run again, ignoring one less than that number. This
4142 will get you quickly to the last hit of that breakpoint.
4145 For a breakpoints with an enable count (xref) greater than 1,
4146 @code{info break} also displays that count.
4150 @value{GDBN} allows you to set any number of breakpoints at the same place in
4151 your program. There is nothing silly or meaningless about this. When
4152 the breakpoints are conditional, this is even useful
4153 (@pxref{Conditions, ,Break Conditions}).
4155 @cindex multiple locations, breakpoints
4156 @cindex breakpoints, multiple locations
4157 It is possible that a breakpoint corresponds to several locations
4158 in your program. Examples of this situation are:
4162 Multiple functions in the program may have the same name.
4165 For a C@t{++} constructor, the @value{NGCC} compiler generates several
4166 instances of the function body, used in different cases.
4169 For a C@t{++} template function, a given line in the function can
4170 correspond to any number of instantiations.
4173 For an inlined function, a given source line can correspond to
4174 several places where that function is inlined.
4177 In all those cases, @value{GDBN} will insert a breakpoint at all
4178 the relevant locations.
4180 A breakpoint with multiple locations is displayed in the breakpoint
4181 table using several rows---one header row, followed by one row for
4182 each breakpoint location. The header row has @samp{<MULTIPLE>} in the
4183 address column. The rows for individual locations contain the actual
4184 addresses for locations, and show the functions to which those
4185 locations belong. The number column for a location is of the form
4186 @var{breakpoint-number}.@var{location-number}.
4191 Num Type Disp Enb Address What
4192 1 breakpoint keep y <MULTIPLE>
4194 breakpoint already hit 1 time
4195 1.1 y 0x080486a2 in void foo<int>() at t.cc:8
4196 1.2 y 0x080486ca in void foo<double>() at t.cc:8
4199 You cannot delete the individual locations from a breakpoint. However,
4200 each location can be individually enabled or disabled by passing
4201 @var{breakpoint-number}.@var{location-number} as argument to the
4202 @code{enable} and @code{disable} commands. It's also possible to
4203 @code{enable} and @code{disable} a range of @var{location-number}
4204 locations using a @var{breakpoint-number} and two @var{location-number}s,
4205 in increasing order, separated by a hyphen, like
4206 @kbd{@var{breakpoint-number}.@var{location-number1}-@var{location-number2}},
4207 in which case @value{GDBN} acts on all the locations in the range (inclusive).
4208 Disabling or enabling the parent breakpoint (@pxref{Disabling}) affects
4209 all of the locations that belong to that breakpoint.
4211 @cindex pending breakpoints
4212 It's quite common to have a breakpoint inside a shared library.
4213 Shared libraries can be loaded and unloaded explicitly,
4214 and possibly repeatedly, as the program is executed. To support
4215 this use case, @value{GDBN} updates breakpoint locations whenever
4216 any shared library is loaded or unloaded. Typically, you would
4217 set a breakpoint in a shared library at the beginning of your
4218 debugging session, when the library is not loaded, and when the
4219 symbols from the library are not available. When you try to set
4220 breakpoint, @value{GDBN} will ask you if you want to set
4221 a so called @dfn{pending breakpoint}---breakpoint whose address
4222 is not yet resolved.
4224 After the program is run, whenever a new shared library is loaded,
4225 @value{GDBN} reevaluates all the breakpoints. When a newly loaded
4226 shared library contains the symbol or line referred to by some
4227 pending breakpoint, that breakpoint is resolved and becomes an
4228 ordinary breakpoint. When a library is unloaded, all breakpoints
4229 that refer to its symbols or source lines become pending again.
4231 This logic works for breakpoints with multiple locations, too. For
4232 example, if you have a breakpoint in a C@t{++} template function, and
4233 a newly loaded shared library has an instantiation of that template,
4234 a new location is added to the list of locations for the breakpoint.
4236 Except for having unresolved address, pending breakpoints do not
4237 differ from regular breakpoints. You can set conditions or commands,
4238 enable and disable them and perform other breakpoint operations.
4240 @value{GDBN} provides some additional commands for controlling what
4241 happens when the @samp{break} command cannot resolve breakpoint
4242 address specification to an address:
4244 @kindex set breakpoint pending
4245 @kindex show breakpoint pending
4247 @item set breakpoint pending auto
4248 This is the default behavior. When @value{GDBN} cannot find the breakpoint
4249 location, it queries you whether a pending breakpoint should be created.
4251 @item set breakpoint pending on
4252 This indicates that an unrecognized breakpoint location should automatically
4253 result in a pending breakpoint being created.
4255 @item set breakpoint pending off
4256 This indicates that pending breakpoints are not to be created. Any
4257 unrecognized breakpoint location results in an error. This setting does
4258 not affect any pending breakpoints previously created.
4260 @item show breakpoint pending
4261 Show the current behavior setting for creating pending breakpoints.
4264 The settings above only affect the @code{break} command and its
4265 variants. Once breakpoint is set, it will be automatically updated
4266 as shared libraries are loaded and unloaded.
4268 @cindex automatic hardware breakpoints
4269 For some targets, @value{GDBN} can automatically decide if hardware or
4270 software breakpoints should be used, depending on whether the
4271 breakpoint address is read-only or read-write. This applies to
4272 breakpoints set with the @code{break} command as well as to internal
4273 breakpoints set by commands like @code{next} and @code{finish}. For
4274 breakpoints set with @code{hbreak}, @value{GDBN} will always use hardware
4277 You can control this automatic behaviour with the following commands:
4279 @kindex set breakpoint auto-hw
4280 @kindex show breakpoint auto-hw
4282 @item set breakpoint auto-hw on
4283 This is the default behavior. When @value{GDBN} sets a breakpoint, it
4284 will try to use the target memory map to decide if software or hardware
4285 breakpoint must be used.
4287 @item set breakpoint auto-hw off
4288 This indicates @value{GDBN} should not automatically select breakpoint
4289 type. If the target provides a memory map, @value{GDBN} will warn when
4290 trying to set software breakpoint at a read-only address.
4293 @value{GDBN} normally implements breakpoints by replacing the program code
4294 at the breakpoint address with a special instruction, which, when
4295 executed, given control to the debugger. By default, the program
4296 code is so modified only when the program is resumed. As soon as
4297 the program stops, @value{GDBN} restores the original instructions. This
4298 behaviour guards against leaving breakpoints inserted in the
4299 target should gdb abrubptly disconnect. However, with slow remote
4300 targets, inserting and removing breakpoint can reduce the performance.
4301 This behavior can be controlled with the following commands::
4303 @kindex set breakpoint always-inserted
4304 @kindex show breakpoint always-inserted
4306 @item set breakpoint always-inserted off
4307 All breakpoints, including newly added by the user, are inserted in
4308 the target only when the target is resumed. All breakpoints are
4309 removed from the target when it stops. This is the default mode.
4311 @item set breakpoint always-inserted on
4312 Causes all breakpoints to be inserted in the target at all times. If
4313 the user adds a new breakpoint, or changes an existing breakpoint, the
4314 breakpoints in the target are updated immediately. A breakpoint is
4315 removed from the target only when breakpoint itself is deleted.
4318 @value{GDBN} handles conditional breakpoints by evaluating these conditions
4319 when a breakpoint breaks. If the condition is true, then the process being
4320 debugged stops, otherwise the process is resumed.
4322 If the target supports evaluating conditions on its end, @value{GDBN} may
4323 download the breakpoint, together with its conditions, to it.
4325 This feature can be controlled via the following commands:
4327 @kindex set breakpoint condition-evaluation
4328 @kindex show breakpoint condition-evaluation
4330 @item set breakpoint condition-evaluation host
4331 This option commands @value{GDBN} to evaluate the breakpoint
4332 conditions on the host's side. Unconditional breakpoints are sent to
4333 the target which in turn receives the triggers and reports them back to GDB
4334 for condition evaluation. This is the standard evaluation mode.
4336 @item set breakpoint condition-evaluation target
4337 This option commands @value{GDBN} to download breakpoint conditions
4338 to the target at the moment of their insertion. The target
4339 is responsible for evaluating the conditional expression and reporting
4340 breakpoint stop events back to @value{GDBN} whenever the condition
4341 is true. Due to limitations of target-side evaluation, some conditions
4342 cannot be evaluated there, e.g., conditions that depend on local data
4343 that is only known to the host. Examples include
4344 conditional expressions involving convenience variables, complex types
4345 that cannot be handled by the agent expression parser and expressions
4346 that are too long to be sent over to the target, specially when the
4347 target is a remote system. In these cases, the conditions will be
4348 evaluated by @value{GDBN}.
4350 @item set breakpoint condition-evaluation auto
4351 This is the default mode. If the target supports evaluating breakpoint
4352 conditions on its end, @value{GDBN} will download breakpoint conditions to
4353 the target (limitations mentioned previously apply). If the target does
4354 not support breakpoint condition evaluation, then @value{GDBN} will fallback
4355 to evaluating all these conditions on the host's side.
4359 @cindex negative breakpoint numbers
4360 @cindex internal @value{GDBN} breakpoints
4361 @value{GDBN} itself sometimes sets breakpoints in your program for
4362 special purposes, such as proper handling of @code{longjmp} (in C
4363 programs). These internal breakpoints are assigned negative numbers,
4364 starting with @code{-1}; @samp{info breakpoints} does not display them.
4365 You can see these breakpoints with the @value{GDBN} maintenance command
4366 @samp{maint info breakpoints} (@pxref{maint info breakpoints}).
4369 @node Set Watchpoints
4370 @subsection Setting Watchpoints
4372 @cindex setting watchpoints
4373 You can use a watchpoint to stop execution whenever the value of an
4374 expression changes, without having to predict a particular place where
4375 this may happen. (This is sometimes called a @dfn{data breakpoint}.)
4376 The expression may be as simple as the value of a single variable, or
4377 as complex as many variables combined by operators. Examples include:
4381 A reference to the value of a single variable.
4384 An address cast to an appropriate data type. For example,
4385 @samp{*(int *)0x12345678} will watch a 4-byte region at the specified
4386 address (assuming an @code{int} occupies 4 bytes).
4389 An arbitrarily complex expression, such as @samp{a*b + c/d}. The
4390 expression can use any operators valid in the program's native
4391 language (@pxref{Languages}).
4394 You can set a watchpoint on an expression even if the expression can
4395 not be evaluated yet. For instance, you can set a watchpoint on
4396 @samp{*global_ptr} before @samp{global_ptr} is initialized.
4397 @value{GDBN} will stop when your program sets @samp{global_ptr} and
4398 the expression produces a valid value. If the expression becomes
4399 valid in some other way than changing a variable (e.g.@: if the memory
4400 pointed to by @samp{*global_ptr} becomes readable as the result of a
4401 @code{malloc} call), @value{GDBN} may not stop until the next time
4402 the expression changes.
4404 @cindex software watchpoints
4405 @cindex hardware watchpoints
4406 Depending on your system, watchpoints may be implemented in software or
4407 hardware. @value{GDBN} does software watchpointing by single-stepping your
4408 program and testing the variable's value each time, which is hundreds of
4409 times slower than normal execution. (But this may still be worth it, to
4410 catch errors where you have no clue what part of your program is the
4413 On some systems, such as most PowerPC or x86-based targets,
4414 @value{GDBN} includes support for hardware watchpoints, which do not
4415 slow down the running of your program.
4419 @item watch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4420 Set a watchpoint for an expression. @value{GDBN} will break when the
4421 expression @var{expr} is written into by the program and its value
4422 changes. The simplest (and the most popular) use of this command is
4423 to watch the value of a single variable:
4426 (@value{GDBP}) watch foo
4429 If the command includes a @code{@r{[}thread @var{thread-id}@r{]}}
4430 argument, @value{GDBN} breaks only when the thread identified by
4431 @var{thread-id} changes the value of @var{expr}. If any other threads
4432 change the value of @var{expr}, @value{GDBN} will not break. Note
4433 that watchpoints restricted to a single thread in this way only work
4434 with Hardware Watchpoints.
4436 Ordinarily a watchpoint respects the scope of variables in @var{expr}
4437 (see below). The @code{-location} argument tells @value{GDBN} to
4438 instead watch the memory referred to by @var{expr}. In this case,
4439 @value{GDBN} will evaluate @var{expr}, take the address of the result,
4440 and watch the memory at that address. The type of the result is used
4441 to determine the size of the watched memory. If the expression's
4442 result does not have an address, then @value{GDBN} will print an
4445 The @code{@r{[}mask @var{maskvalue}@r{]}} argument allows creation
4446 of masked watchpoints, if the current architecture supports this
4447 feature (e.g., PowerPC Embedded architecture, see @ref{PowerPC
4448 Embedded}.) A @dfn{masked watchpoint} specifies a mask in addition
4449 to an address to watch. The mask specifies that some bits of an address
4450 (the bits which are reset in the mask) should be ignored when matching
4451 the address accessed by the inferior against the watchpoint address.
4452 Thus, a masked watchpoint watches many addresses simultaneously---those
4453 addresses whose unmasked bits are identical to the unmasked bits in the
4454 watchpoint address. The @code{mask} argument implies @code{-location}.
4458 (@value{GDBP}) watch foo mask 0xffff00ff
4459 (@value{GDBP}) watch *0xdeadbeef mask 0xffffff00
4463 @item rwatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4464 Set a watchpoint that will break when the value of @var{expr} is read
4468 @item awatch @r{[}-l@r{|}-location@r{]} @var{expr} @r{[}thread @var{thread-id}@r{]} @r{[}mask @var{maskvalue}@r{]}
4469 Set a watchpoint that will break when @var{expr} is either read from
4470 or written into by the program.
4472 @kindex info watchpoints @r{[}@var{list}@dots{}@r{]}
4473 @item info watchpoints @r{[}@var{list}@dots{}@r{]}
4474 This command prints a list of watchpoints, using the same format as
4475 @code{info break} (@pxref{Set Breaks}).
4478 If you watch for a change in a numerically entered address you need to
4479 dereference it, as the address itself is just a constant number which will
4480 never change. @value{GDBN} refuses to create a watchpoint that watches
4481 a never-changing value:
4484 (@value{GDBP}) watch 0x600850
4485 Cannot watch constant value 0x600850.
4486 (@value{GDBP}) watch *(int *) 0x600850
4487 Watchpoint 1: *(int *) 6293584
4490 @value{GDBN} sets a @dfn{hardware watchpoint} if possible. Hardware
4491 watchpoints execute very quickly, and the debugger reports a change in
4492 value at the exact instruction where the change occurs. If @value{GDBN}
4493 cannot set a hardware watchpoint, it sets a software watchpoint, which
4494 executes more slowly and reports the change in value at the next
4495 @emph{statement}, not the instruction, after the change occurs.
4497 @cindex use only software watchpoints
4498 You can force @value{GDBN} to use only software watchpoints with the
4499 @kbd{set can-use-hw-watchpoints 0} command. With this variable set to
4500 zero, @value{GDBN} will never try to use hardware watchpoints, even if
4501 the underlying system supports them. (Note that hardware-assisted
4502 watchpoints that were set @emph{before} setting
4503 @code{can-use-hw-watchpoints} to zero will still use the hardware
4504 mechanism of watching expression values.)
4507 @item set can-use-hw-watchpoints
4508 @kindex set can-use-hw-watchpoints
4509 Set whether or not to use hardware watchpoints.
4511 @item show can-use-hw-watchpoints
4512 @kindex show can-use-hw-watchpoints
4513 Show the current mode of using hardware watchpoints.
4516 For remote targets, you can restrict the number of hardware
4517 watchpoints @value{GDBN} will use, see @ref{set remote
4518 hardware-breakpoint-limit}.
4520 When you issue the @code{watch} command, @value{GDBN} reports
4523 Hardware watchpoint @var{num}: @var{expr}
4527 if it was able to set a hardware watchpoint.
4529 Currently, the @code{awatch} and @code{rwatch} commands can only set
4530 hardware watchpoints, because accesses to data that don't change the
4531 value of the watched expression cannot be detected without examining
4532 every instruction as it is being executed, and @value{GDBN} does not do
4533 that currently. If @value{GDBN} finds that it is unable to set a
4534 hardware breakpoint with the @code{awatch} or @code{rwatch} command, it
4535 will print a message like this:
4538 Expression cannot be implemented with read/access watchpoint.
4541 Sometimes, @value{GDBN} cannot set a hardware watchpoint because the
4542 data type of the watched expression is wider than what a hardware
4543 watchpoint on the target machine can handle. For example, some systems
4544 can only watch regions that are up to 4 bytes wide; on such systems you
4545 cannot set hardware watchpoints for an expression that yields a
4546 double-precision floating-point number (which is typically 8 bytes
4547 wide). As a work-around, it might be possible to break the large region
4548 into a series of smaller ones and watch them with separate watchpoints.
4550 If you set too many hardware watchpoints, @value{GDBN} might be unable
4551 to insert all of them when you resume the execution of your program.
4552 Since the precise number of active watchpoints is unknown until such
4553 time as the program is about to be resumed, @value{GDBN} might not be
4554 able to warn you about this when you set the watchpoints, and the
4555 warning will be printed only when the program is resumed:
4558 Hardware watchpoint @var{num}: Could not insert watchpoint
4562 If this happens, delete or disable some of the watchpoints.
4564 Watching complex expressions that reference many variables can also
4565 exhaust the resources available for hardware-assisted watchpoints.
4566 That's because @value{GDBN} needs to watch every variable in the
4567 expression with separately allocated resources.
4569 If you call a function interactively using @code{print} or @code{call},
4570 any watchpoints you have set will be inactive until @value{GDBN} reaches another
4571 kind of breakpoint or the call completes.
4573 @value{GDBN} automatically deletes watchpoints that watch local
4574 (automatic) variables, or expressions that involve such variables, when
4575 they go out of scope, that is, when the execution leaves the block in
4576 which these variables were defined. In particular, when the program
4577 being debugged terminates, @emph{all} local variables go out of scope,
4578 and so only watchpoints that watch global variables remain set. If you
4579 rerun the program, you will need to set all such watchpoints again. One
4580 way of doing that would be to set a code breakpoint at the entry to the
4581 @code{main} function and when it breaks, set all the watchpoints.
4583 @cindex watchpoints and threads
4584 @cindex threads and watchpoints
4585 In multi-threaded programs, watchpoints will detect changes to the
4586 watched expression from every thread.
4589 @emph{Warning:} In multi-threaded programs, software watchpoints
4590 have only limited usefulness. If @value{GDBN} creates a software
4591 watchpoint, it can only watch the value of an expression @emph{in a
4592 single thread}. If you are confident that the expression can only
4593 change due to the current thread's activity (and if you are also
4594 confident that no other thread can become current), then you can use
4595 software watchpoints as usual. However, @value{GDBN} may not notice
4596 when a non-current thread's activity changes the expression. (Hardware
4597 watchpoints, in contrast, watch an expression in all threads.)
4600 @xref{set remote hardware-watchpoint-limit}.
4602 @node Set Catchpoints
4603 @subsection Setting Catchpoints
4604 @cindex catchpoints, setting
4605 @cindex exception handlers
4606 @cindex event handling
4608 You can use @dfn{catchpoints} to cause the debugger to stop for certain
4609 kinds of program events, such as C@t{++} exceptions or the loading of a
4610 shared library. Use the @code{catch} command to set a catchpoint.
4614 @item catch @var{event}
4615 Stop when @var{event} occurs. The @var{event} can be any of the following:
4618 @item throw @r{[}@var{regexp}@r{]}
4619 @itemx rethrow @r{[}@var{regexp}@r{]}
4620 @itemx catch @r{[}@var{regexp}@r{]}
4622 @kindex catch rethrow
4624 @cindex stop on C@t{++} exceptions
4625 The throwing, re-throwing, or catching of a C@t{++} exception.
4627 If @var{regexp} is given, then only exceptions whose type matches the
4628 regular expression will be caught.
4630 @vindex $_exception@r{, convenience variable}
4631 The convenience variable @code{$_exception} is available at an
4632 exception-related catchpoint, on some systems. This holds the
4633 exception being thrown.
4635 There are currently some limitations to C@t{++} exception handling in
4640 The support for these commands is system-dependent. Currently, only
4641 systems using the @samp{gnu-v3} C@t{++} ABI (@pxref{ABI}) are
4645 The regular expression feature and the @code{$_exception} convenience
4646 variable rely on the presence of some SDT probes in @code{libstdc++}.
4647 If these probes are not present, then these features cannot be used.
4648 These probes were first available in the GCC 4.8 release, but whether
4649 or not they are available in your GCC also depends on how it was
4653 The @code{$_exception} convenience variable is only valid at the
4654 instruction at which an exception-related catchpoint is set.
4657 When an exception-related catchpoint is hit, @value{GDBN} stops at a
4658 location in the system library which implements runtime exception
4659 support for C@t{++}, usually @code{libstdc++}. You can use @code{up}
4660 (@pxref{Selection}) to get to your code.
4663 If you call a function interactively, @value{GDBN} normally returns
4664 control to you when the function has finished executing. If the call
4665 raises an exception, however, the call may bypass the mechanism that
4666 returns control to you and cause your program either to abort or to
4667 simply continue running until it hits a breakpoint, catches a signal
4668 that @value{GDBN} is listening for, or exits. This is the case even if
4669 you set a catchpoint for the exception; catchpoints on exceptions are
4670 disabled within interactive calls. @xref{Calling}, for information on
4671 controlling this with @code{set unwind-on-terminating-exception}.
4674 You cannot raise an exception interactively.
4677 You cannot install an exception handler interactively.
4680 @item exception @r{[}@var{name}@r{]}
4681 @kindex catch exception
4682 @cindex Ada exception catching
4683 @cindex catch Ada exceptions
4684 An Ada exception being raised. If an exception name is specified
4685 at the end of the command (eg @code{catch exception Program_Error}),
4686 the debugger will stop only when this specific exception is raised.
4687 Otherwise, the debugger stops execution when any Ada exception is raised.
4689 When inserting an exception catchpoint on a user-defined exception whose
4690 name is identical to one of the exceptions defined by the language, the
4691 fully qualified name must be used as the exception name. Otherwise,
4692 @value{GDBN} will assume that it should stop on the pre-defined exception
4693 rather than the user-defined one. For instance, assuming an exception
4694 called @code{Constraint_Error} is defined in package @code{Pck}, then
4695 the command to use to catch such exceptions is @kbd{catch exception
4696 Pck.Constraint_Error}.
4698 @item exception unhandled
4699 @kindex catch exception unhandled
4700 An exception that was raised but is not handled by the program.
4702 @item handlers @r{[}@var{name}@r{]}
4703 @kindex catch handlers
4704 @cindex Ada exception handlers catching
4705 @cindex catch Ada exceptions when handled
4706 An Ada exception being handled. If an exception name is
4707 specified at the end of the command
4708 (eg @kbd{catch handlers Program_Error}), the debugger will stop
4709 only when this specific exception is handled.
4710 Otherwise, the debugger stops execution when any Ada exception is handled.
4712 When inserting a handlers catchpoint on a user-defined
4713 exception whose name is identical to one of the exceptions
4714 defined by the language, the fully qualified name must be used
4715 as the exception name. Otherwise, @value{GDBN} will assume that it
4716 should stop on the pre-defined exception rather than the
4717 user-defined one. For instance, assuming an exception called
4718 @code{Constraint_Error} is defined in package @code{Pck}, then the
4719 command to use to catch such exceptions handling is
4720 @kbd{catch handlers Pck.Constraint_Error}.
4723 @kindex catch assert
4724 A failed Ada assertion.
4728 @cindex break on fork/exec
4729 A call to @code{exec}.
4731 @anchor{catch syscall}
4733 @itemx syscall @r{[}@var{name} @r{|} @var{number} @r{|} @r{group:}@var{groupname} @r{|} @r{g:}@var{groupname}@r{]} @dots{}
4734 @kindex catch syscall
4735 @cindex break on a system call.
4736 A call to or return from a system call, a.k.a.@: @dfn{syscall}. A
4737 syscall is a mechanism for application programs to request a service
4738 from the operating system (OS) or one of the OS system services.
4739 @value{GDBN} can catch some or all of the syscalls issued by the
4740 debuggee, and show the related information for each syscall. If no
4741 argument is specified, calls to and returns from all system calls
4744 @var{name} can be any system call name that is valid for the
4745 underlying OS. Just what syscalls are valid depends on the OS. On
4746 GNU and Unix systems, you can find the full list of valid syscall
4747 names on @file{/usr/include/asm/unistd.h}.
4749 @c For MS-Windows, the syscall names and the corresponding numbers
4750 @c can be found, e.g., on this URL:
4751 @c http://www.metasploit.com/users/opcode/syscalls.html
4752 @c but we don't support Windows syscalls yet.
4754 Normally, @value{GDBN} knows in advance which syscalls are valid for
4755 each OS, so you can use the @value{GDBN} command-line completion
4756 facilities (@pxref{Completion,, command completion}) to list the
4759 You may also specify the system call numerically. A syscall's
4760 number is the value passed to the OS's syscall dispatcher to
4761 identify the requested service. When you specify the syscall by its
4762 name, @value{GDBN} uses its database of syscalls to convert the name
4763 into the corresponding numeric code, but using the number directly
4764 may be useful if @value{GDBN}'s database does not have the complete
4765 list of syscalls on your system (e.g., because @value{GDBN} lags
4766 behind the OS upgrades).
4768 You may specify a group of related syscalls to be caught at once using
4769 the @code{group:} syntax (@code{g:} is a shorter equivalent). For
4770 instance, on some platforms @value{GDBN} allows you to catch all
4771 network related syscalls, by passing the argument @code{group:network}
4772 to @code{catch syscall}. Note that not all syscall groups are
4773 available in every system. You can use the command completion
4774 facilities (@pxref{Completion,, command completion}) to list the
4775 syscall groups available on your environment.
4777 The example below illustrates how this command works if you don't provide
4781 (@value{GDBP}) catch syscall
4782 Catchpoint 1 (syscall)
4784 Starting program: /tmp/catch-syscall
4786 Catchpoint 1 (call to syscall 'close'), \
4787 0xffffe424 in __kernel_vsyscall ()
4791 Catchpoint 1 (returned from syscall 'close'), \
4792 0xffffe424 in __kernel_vsyscall ()
4796 Here is an example of catching a system call by name:
4799 (@value{GDBP}) catch syscall chroot
4800 Catchpoint 1 (syscall 'chroot' [61])
4802 Starting program: /tmp/catch-syscall
4804 Catchpoint 1 (call to syscall 'chroot'), \
4805 0xffffe424 in __kernel_vsyscall ()
4809 Catchpoint 1 (returned from syscall 'chroot'), \
4810 0xffffe424 in __kernel_vsyscall ()
4814 An example of specifying a system call numerically. In the case
4815 below, the syscall number has a corresponding entry in the XML
4816 file, so @value{GDBN} finds its name and prints it:
4819 (@value{GDBP}) catch syscall 252
4820 Catchpoint 1 (syscall(s) 'exit_group')
4822 Starting program: /tmp/catch-syscall
4824 Catchpoint 1 (call to syscall 'exit_group'), \
4825 0xffffe424 in __kernel_vsyscall ()
4829 Program exited normally.
4833 Here is an example of catching a syscall group:
4836 (@value{GDBP}) catch syscall group:process
4837 Catchpoint 1 (syscalls 'exit' [1] 'fork' [2] 'waitpid' [7]
4838 'execve' [11] 'wait4' [114] 'clone' [120] 'vfork' [190]
4839 'exit_group' [252] 'waitid' [284] 'unshare' [310])
4841 Starting program: /tmp/catch-syscall
4843 Catchpoint 1 (call to syscall fork), 0x00007ffff7df4e27 in open64 ()
4844 from /lib64/ld-linux-x86-64.so.2
4850 However, there can be situations when there is no corresponding name
4851 in XML file for that syscall number. In this case, @value{GDBN} prints
4852 a warning message saying that it was not able to find the syscall name,
4853 but the catchpoint will be set anyway. See the example below:
4856 (@value{GDBP}) catch syscall 764
4857 warning: The number '764' does not represent a known syscall.
4858 Catchpoint 2 (syscall 764)
4862 If you configure @value{GDBN} using the @samp{--without-expat} option,
4863 it will not be able to display syscall names. Also, if your
4864 architecture does not have an XML file describing its system calls,
4865 you will not be able to see the syscall names. It is important to
4866 notice that these two features are used for accessing the syscall
4867 name database. In either case, you will see a warning like this:
4870 (@value{GDBP}) catch syscall
4871 warning: Could not open "syscalls/i386-linux.xml"
4872 warning: Could not load the syscall XML file 'syscalls/i386-linux.xml'.
4873 GDB will not be able to display syscall names.
4874 Catchpoint 1 (syscall)
4878 Of course, the file name will change depending on your architecture and system.
4880 Still using the example above, you can also try to catch a syscall by its
4881 number. In this case, you would see something like:
4884 (@value{GDBP}) catch syscall 252
4885 Catchpoint 1 (syscall(s) 252)
4888 Again, in this case @value{GDBN} would not be able to display syscall's names.
4892 A call to @code{fork}.
4896 A call to @code{vfork}.
4898 @item load @r{[}@var{regexp}@r{]}
4899 @itemx unload @r{[}@var{regexp}@r{]}
4901 @kindex catch unload
4902 The loading or unloading of a shared library. If @var{regexp} is
4903 given, then the catchpoint will stop only if the regular expression
4904 matches one of the affected libraries.
4906 @item signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
4907 @kindex catch signal
4908 The delivery of a signal.
4910 With no arguments, this catchpoint will catch any signal that is not
4911 used internally by @value{GDBN}, specifically, all signals except
4912 @samp{SIGTRAP} and @samp{SIGINT}.
4914 With the argument @samp{all}, all signals, including those used by
4915 @value{GDBN}, will be caught. This argument cannot be used with other
4918 Otherwise, the arguments are a list of signal names as given to
4919 @code{handle} (@pxref{Signals}). Only signals specified in this list
4922 One reason that @code{catch signal} can be more useful than
4923 @code{handle} is that you can attach commands and conditions to the
4926 When a signal is caught by a catchpoint, the signal's @code{stop} and
4927 @code{print} settings, as specified by @code{handle}, are ignored.
4928 However, whether the signal is still delivered to the inferior depends
4929 on the @code{pass} setting; this can be changed in the catchpoint's
4934 @item tcatch @var{event}
4936 Set a catchpoint that is enabled only for one stop. The catchpoint is
4937 automatically deleted after the first time the event is caught.
4941 Use the @code{info break} command to list the current catchpoints.
4945 @subsection Deleting Breakpoints
4947 @cindex clearing breakpoints, watchpoints, catchpoints
4948 @cindex deleting breakpoints, watchpoints, catchpoints
4949 It is often necessary to eliminate a breakpoint, watchpoint, or
4950 catchpoint once it has done its job and you no longer want your program
4951 to stop there. This is called @dfn{deleting} the breakpoint. A
4952 breakpoint that has been deleted no longer exists; it is forgotten.
4954 With the @code{clear} command you can delete breakpoints according to
4955 where they are in your program. With the @code{delete} command you can
4956 delete individual breakpoints, watchpoints, or catchpoints by specifying
4957 their breakpoint numbers.
4959 It is not necessary to delete a breakpoint to proceed past it. @value{GDBN}
4960 automatically ignores breakpoints on the first instruction to be executed
4961 when you continue execution without changing the execution address.
4966 Delete any breakpoints at the next instruction to be executed in the
4967 selected stack frame (@pxref{Selection, ,Selecting a Frame}). When
4968 the innermost frame is selected, this is a good way to delete a
4969 breakpoint where your program just stopped.
4971 @item clear @var{location}
4972 Delete any breakpoints set at the specified @var{location}.
4973 @xref{Specify Location}, for the various forms of @var{location}; the
4974 most useful ones are listed below:
4977 @item clear @var{function}
4978 @itemx clear @var{filename}:@var{function}
4979 Delete any breakpoints set at entry to the named @var{function}.
4981 @item clear @var{linenum}
4982 @itemx clear @var{filename}:@var{linenum}
4983 Delete any breakpoints set at or within the code of the specified
4984 @var{linenum} of the specified @var{filename}.
4987 @cindex delete breakpoints
4989 @kindex d @r{(@code{delete})}
4990 @item delete @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
4991 Delete the breakpoints, watchpoints, or catchpoints of the breakpoint
4992 list specified as argument. If no argument is specified, delete all
4993 breakpoints (@value{GDBN} asks confirmation, unless you have @code{set
4994 confirm off}). You can abbreviate this command as @code{d}.
4998 @subsection Disabling Breakpoints
5000 @cindex enable/disable a breakpoint
5001 Rather than deleting a breakpoint, watchpoint, or catchpoint, you might
5002 prefer to @dfn{disable} it. This makes the breakpoint inoperative as if
5003 it had been deleted, but remembers the information on the breakpoint so
5004 that you can @dfn{enable} it again later.
5006 You disable and enable breakpoints, watchpoints, and catchpoints with
5007 the @code{enable} and @code{disable} commands, optionally specifying
5008 one or more breakpoint numbers as arguments. Use @code{info break} to
5009 print a list of all breakpoints, watchpoints, and catchpoints if you
5010 do not know which numbers to use.
5012 Disabling and enabling a breakpoint that has multiple locations
5013 affects all of its locations.
5015 A breakpoint, watchpoint, or catchpoint can have any of several
5016 different states of enablement:
5020 Enabled. The breakpoint stops your program. A breakpoint set
5021 with the @code{break} command starts out in this state.
5023 Disabled. The breakpoint has no effect on your program.
5025 Enabled once. The breakpoint stops your program, but then becomes
5028 Enabled for a count. The breakpoint stops your program for the next
5029 N times, then becomes disabled.
5031 Enabled for deletion. The breakpoint stops your program, but
5032 immediately after it does so it is deleted permanently. A breakpoint
5033 set with the @code{tbreak} command starts out in this state.
5036 You can use the following commands to enable or disable breakpoints,
5037 watchpoints, and catchpoints:
5041 @kindex dis @r{(@code{disable})}
5042 @item disable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5043 Disable the specified breakpoints---or all breakpoints, if none are
5044 listed. A disabled breakpoint has no effect but is not forgotten. All
5045 options such as ignore-counts, conditions and commands are remembered in
5046 case the breakpoint is enabled again later. You may abbreviate
5047 @code{disable} as @code{dis}.
5050 @item enable @r{[}breakpoints@r{]} @r{[}@var{list}@dots{}@r{]}
5051 Enable the specified breakpoints (or all defined breakpoints). They
5052 become effective once again in stopping your program.
5054 @item enable @r{[}breakpoints@r{]} once @var{list}@dots{}
5055 Enable the specified breakpoints temporarily. @value{GDBN} disables any
5056 of these breakpoints immediately after stopping your program.
5058 @item enable @r{[}breakpoints@r{]} count @var{count} @var{list}@dots{}
5059 Enable the specified breakpoints temporarily. @value{GDBN} records
5060 @var{count} with each of the specified breakpoints, and decrements a
5061 breakpoint's count when it is hit. When any count reaches 0,
5062 @value{GDBN} disables that breakpoint. If a breakpoint has an ignore
5063 count (@pxref{Conditions, ,Break Conditions}), that will be
5064 decremented to 0 before @var{count} is affected.
5066 @item enable @r{[}breakpoints@r{]} delete @var{list}@dots{}
5067 Enable the specified breakpoints to work once, then die. @value{GDBN}
5068 deletes any of these breakpoints as soon as your program stops there.
5069 Breakpoints set by the @code{tbreak} command start out in this state.
5072 @c FIXME: I think the following ``Except for [...] @code{tbreak}'' is
5073 @c confusing: tbreak is also initially enabled.
5074 Except for a breakpoint set with @code{tbreak} (@pxref{Set Breaks,
5075 ,Setting Breakpoints}), breakpoints that you set are initially enabled;
5076 subsequently, they become disabled or enabled only when you use one of
5077 the commands above. (The command @code{until} can set and delete a
5078 breakpoint of its own, but it does not change the state of your other
5079 breakpoints; see @ref{Continuing and Stepping, ,Continuing and
5083 @subsection Break Conditions
5084 @cindex conditional breakpoints
5085 @cindex breakpoint conditions
5087 @c FIXME what is scope of break condition expr? Context where wanted?
5088 @c in particular for a watchpoint?
5089 The simplest sort of breakpoint breaks every time your program reaches a
5090 specified place. You can also specify a @dfn{condition} for a
5091 breakpoint. A condition is just a Boolean expression in your
5092 programming language (@pxref{Expressions, ,Expressions}). A breakpoint with
5093 a condition evaluates the expression each time your program reaches it,
5094 and your program stops only if the condition is @emph{true}.
5096 This is the converse of using assertions for program validation; in that
5097 situation, you want to stop when the assertion is violated---that is,
5098 when the condition is false. In C, if you want to test an assertion expressed
5099 by the condition @var{assert}, you should set the condition
5100 @samp{! @var{assert}} on the appropriate breakpoint.
5102 Conditions are also accepted for watchpoints; you may not need them,
5103 since a watchpoint is inspecting the value of an expression anyhow---but
5104 it might be simpler, say, to just set a watchpoint on a variable name,
5105 and specify a condition that tests whether the new value is an interesting
5108 Break conditions can have side effects, and may even call functions in
5109 your program. This can be useful, for example, to activate functions
5110 that log program progress, or to use your own print functions to
5111 format special data structures. The effects are completely predictable
5112 unless there is another enabled breakpoint at the same address. (In
5113 that case, @value{GDBN} might see the other breakpoint first and stop your
5114 program without checking the condition of this one.) Note that
5115 breakpoint commands are usually more convenient and flexible than break
5117 purpose of performing side effects when a breakpoint is reached
5118 (@pxref{Break Commands, ,Breakpoint Command Lists}).
5120 Breakpoint conditions can also be evaluated on the target's side if
5121 the target supports it. Instead of evaluating the conditions locally,
5122 @value{GDBN} encodes the expression into an agent expression
5123 (@pxref{Agent Expressions}) suitable for execution on the target,
5124 independently of @value{GDBN}. Global variables become raw memory
5125 locations, locals become stack accesses, and so forth.
5127 In this case, @value{GDBN} will only be notified of a breakpoint trigger
5128 when its condition evaluates to true. This mechanism may provide faster
5129 response times depending on the performance characteristics of the target
5130 since it does not need to keep @value{GDBN} informed about
5131 every breakpoint trigger, even those with false conditions.
5133 Break conditions can be specified when a breakpoint is set, by using
5134 @samp{if} in the arguments to the @code{break} command. @xref{Set
5135 Breaks, ,Setting Breakpoints}. They can also be changed at any time
5136 with the @code{condition} command.
5138 You can also use the @code{if} keyword with the @code{watch} command.
5139 The @code{catch} command does not recognize the @code{if} keyword;
5140 @code{condition} is the only way to impose a further condition on a
5145 @item condition @var{bnum} @var{expression}
5146 Specify @var{expression} as the break condition for breakpoint,
5147 watchpoint, or catchpoint number @var{bnum}. After you set a condition,
5148 breakpoint @var{bnum} stops your program only if the value of
5149 @var{expression} is true (nonzero, in C). When you use
5150 @code{condition}, @value{GDBN} checks @var{expression} immediately for
5151 syntactic correctness, and to determine whether symbols in it have
5152 referents in the context of your breakpoint. If @var{expression} uses
5153 symbols not referenced in the context of the breakpoint, @value{GDBN}
5154 prints an error message:
5157 No symbol "foo" in current context.
5162 not actually evaluate @var{expression} at the time the @code{condition}
5163 command (or a command that sets a breakpoint with a condition, like
5164 @code{break if @dots{}}) is given, however. @xref{Expressions, ,Expressions}.
5166 @item condition @var{bnum}
5167 Remove the condition from breakpoint number @var{bnum}. It becomes
5168 an ordinary unconditional breakpoint.
5171 @cindex ignore count (of breakpoint)
5172 A special case of a breakpoint condition is to stop only when the
5173 breakpoint has been reached a certain number of times. This is so
5174 useful that there is a special way to do it, using the @dfn{ignore
5175 count} of the breakpoint. Every breakpoint has an ignore count, which
5176 is an integer. Most of the time, the ignore count is zero, and
5177 therefore has no effect. But if your program reaches a breakpoint whose
5178 ignore count is positive, then instead of stopping, it just decrements
5179 the ignore count by one and continues. As a result, if the ignore count
5180 value is @var{n}, the breakpoint does not stop the next @var{n} times
5181 your program reaches it.
5185 @item ignore @var{bnum} @var{count}
5186 Set the ignore count of breakpoint number @var{bnum} to @var{count}.
5187 The next @var{count} times the breakpoint is reached, your program's
5188 execution does not stop; other than to decrement the ignore count, @value{GDBN}
5191 To make the breakpoint stop the next time it is reached, specify
5194 When you use @code{continue} to resume execution of your program from a
5195 breakpoint, you can specify an ignore count directly as an argument to
5196 @code{continue}, rather than using @code{ignore}. @xref{Continuing and
5197 Stepping,,Continuing and Stepping}.
5199 If a breakpoint has a positive ignore count and a condition, the
5200 condition is not checked. Once the ignore count reaches zero,
5201 @value{GDBN} resumes checking the condition.
5203 You could achieve the effect of the ignore count with a condition such
5204 as @w{@samp{$foo-- <= 0}} using a debugger convenience variable that
5205 is decremented each time. @xref{Convenience Vars, ,Convenience
5209 Ignore counts apply to breakpoints, watchpoints, and catchpoints.
5212 @node Break Commands
5213 @subsection Breakpoint Command Lists
5215 @cindex breakpoint commands
5216 You can give any breakpoint (or watchpoint or catchpoint) a series of
5217 commands to execute when your program stops due to that breakpoint. For
5218 example, you might want to print the values of certain expressions, or
5219 enable other breakpoints.
5223 @kindex end@r{ (breakpoint commands)}
5224 @item commands @r{[}@var{list}@dots{}@r{]}
5225 @itemx @dots{} @var{command-list} @dots{}
5227 Specify a list of commands for the given breakpoints. The commands
5228 themselves appear on the following lines. Type a line containing just
5229 @code{end} to terminate the commands.
5231 To remove all commands from a breakpoint, type @code{commands} and
5232 follow it immediately with @code{end}; that is, give no commands.
5234 With no argument, @code{commands} refers to the last breakpoint,
5235 watchpoint, or catchpoint set (not to the breakpoint most recently
5236 encountered). If the most recent breakpoints were set with a single
5237 command, then the @code{commands} will apply to all the breakpoints
5238 set by that command. This applies to breakpoints set by
5239 @code{rbreak}, and also applies when a single @code{break} command
5240 creates multiple breakpoints (@pxref{Ambiguous Expressions,,Ambiguous
5244 Pressing @key{RET} as a means of repeating the last @value{GDBN} command is
5245 disabled within a @var{command-list}.
5247 You can use breakpoint commands to start your program up again. Simply
5248 use the @code{continue} command, or @code{step}, or any other command
5249 that resumes execution.
5251 Any other commands in the command list, after a command that resumes
5252 execution, are ignored. This is because any time you resume execution
5253 (even with a simple @code{next} or @code{step}), you may encounter
5254 another breakpoint---which could have its own command list, leading to
5255 ambiguities about which list to execute.
5258 If the first command you specify in a command list is @code{silent}, the
5259 usual message about stopping at a breakpoint is not printed. This may
5260 be desirable for breakpoints that are to print a specific message and
5261 then continue. If none of the remaining commands print anything, you
5262 see no sign that the breakpoint was reached. @code{silent} is
5263 meaningful only at the beginning of a breakpoint command list.
5265 The commands @code{echo}, @code{output}, and @code{printf} allow you to
5266 print precisely controlled output, and are often useful in silent
5267 breakpoints. @xref{Output, ,Commands for Controlled Output}.
5269 For example, here is how you could use breakpoint commands to print the
5270 value of @code{x} at entry to @code{foo} whenever @code{x} is positive.
5276 printf "x is %d\n",x
5281 One application for breakpoint commands is to compensate for one bug so
5282 you can test for another. Put a breakpoint just after the erroneous line
5283 of code, give it a condition to detect the case in which something
5284 erroneous has been done, and give it commands to assign correct values
5285 to any variables that need them. End with the @code{continue} command
5286 so that your program does not stop, and start with the @code{silent}
5287 command so that no output is produced. Here is an example:
5298 @node Dynamic Printf
5299 @subsection Dynamic Printf
5301 @cindex dynamic printf
5303 The dynamic printf command @code{dprintf} combines a breakpoint with
5304 formatted printing of your program's data to give you the effect of
5305 inserting @code{printf} calls into your program on-the-fly, without
5306 having to recompile it.
5308 In its most basic form, the output goes to the GDB console. However,
5309 you can set the variable @code{dprintf-style} for alternate handling.
5310 For instance, you can ask to format the output by calling your
5311 program's @code{printf} function. This has the advantage that the
5312 characters go to the program's output device, so they can recorded in
5313 redirects to files and so forth.
5315 If you are doing remote debugging with a stub or agent, you can also
5316 ask to have the printf handled by the remote agent. In addition to
5317 ensuring that the output goes to the remote program's device along
5318 with any other output the program might produce, you can also ask that
5319 the dprintf remain active even after disconnecting from the remote
5320 target. Using the stub/agent is also more efficient, as it can do
5321 everything without needing to communicate with @value{GDBN}.
5325 @item dprintf @var{location},@var{template},@var{expression}[,@var{expression}@dots{}]
5326 Whenever execution reaches @var{location}, print the values of one or
5327 more @var{expressions} under the control of the string @var{template}.
5328 To print several values, separate them with commas.
5330 @item set dprintf-style @var{style}
5331 Set the dprintf output to be handled in one of several different
5332 styles enumerated below. A change of style affects all existing
5333 dynamic printfs immediately. (If you need individual control over the
5334 print commands, simply define normal breakpoints with
5335 explicitly-supplied command lists.)
5339 @kindex dprintf-style gdb
5340 Handle the output using the @value{GDBN} @code{printf} command.
5343 @kindex dprintf-style call
5344 Handle the output by calling a function in your program (normally
5348 @kindex dprintf-style agent
5349 Have the remote debugging agent (such as @code{gdbserver}) handle
5350 the output itself. This style is only available for agents that
5351 support running commands on the target.
5354 @item set dprintf-function @var{function}
5355 Set the function to call if the dprintf style is @code{call}. By
5356 default its value is @code{printf}. You may set it to any expression.
5357 that @value{GDBN} can evaluate to a function, as per the @code{call}
5360 @item set dprintf-channel @var{channel}
5361 Set a ``channel'' for dprintf. If set to a non-empty value,
5362 @value{GDBN} will evaluate it as an expression and pass the result as
5363 a first argument to the @code{dprintf-function}, in the manner of
5364 @code{fprintf} and similar functions. Otherwise, the dprintf format
5365 string will be the first argument, in the manner of @code{printf}.
5367 As an example, if you wanted @code{dprintf} output to go to a logfile
5368 that is a standard I/O stream assigned to the variable @code{mylog},
5369 you could do the following:
5372 (gdb) set dprintf-style call
5373 (gdb) set dprintf-function fprintf
5374 (gdb) set dprintf-channel mylog
5375 (gdb) dprintf 25,"at line 25, glob=%d\n",glob
5376 Dprintf 1 at 0x123456: file main.c, line 25.
5378 1 dprintf keep y 0x00123456 in main at main.c:25
5379 call (void) fprintf (mylog,"at line 25, glob=%d\n",glob)
5384 Note that the @code{info break} displays the dynamic printf commands
5385 as normal breakpoint commands; you can thus easily see the effect of
5386 the variable settings.
5388 @item set disconnected-dprintf on
5389 @itemx set disconnected-dprintf off
5390 @kindex set disconnected-dprintf
5391 Choose whether @code{dprintf} commands should continue to run if
5392 @value{GDBN} has disconnected from the target. This only applies
5393 if the @code{dprintf-style} is @code{agent}.
5395 @item show disconnected-dprintf off
5396 @kindex show disconnected-dprintf
5397 Show the current choice for disconnected @code{dprintf}.
5401 @value{GDBN} does not check the validity of function and channel,
5402 relying on you to supply values that are meaningful for the contexts
5403 in which they are being used. For instance, the function and channel
5404 may be the values of local variables, but if that is the case, then
5405 all enabled dynamic prints must be at locations within the scope of
5406 those locals. If evaluation fails, @value{GDBN} will report an error.
5408 @node Save Breakpoints
5409 @subsection How to save breakpoints to a file
5411 To save breakpoint definitions to a file use the @w{@code{save
5412 breakpoints}} command.
5415 @kindex save breakpoints
5416 @cindex save breakpoints to a file for future sessions
5417 @item save breakpoints [@var{filename}]
5418 This command saves all current breakpoint definitions together with
5419 their commands and ignore counts, into a file @file{@var{filename}}
5420 suitable for use in a later debugging session. This includes all
5421 types of breakpoints (breakpoints, watchpoints, catchpoints,
5422 tracepoints). To read the saved breakpoint definitions, use the
5423 @code{source} command (@pxref{Command Files}). Note that watchpoints
5424 with expressions involving local variables may fail to be recreated
5425 because it may not be possible to access the context where the
5426 watchpoint is valid anymore. Because the saved breakpoint definitions
5427 are simply a sequence of @value{GDBN} commands that recreate the
5428 breakpoints, you can edit the file in your favorite editing program,
5429 and remove the breakpoint definitions you're not interested in, or
5430 that can no longer be recreated.
5433 @node Static Probe Points
5434 @subsection Static Probe Points
5436 @cindex static probe point, SystemTap
5437 @cindex static probe point, DTrace
5438 @value{GDBN} supports @dfn{SDT} probes in the code. @acronym{SDT} stands
5439 for Statically Defined Tracing, and the probes are designed to have a tiny
5440 runtime code and data footprint, and no dynamic relocations.
5442 Currently, the following types of probes are supported on
5443 ELF-compatible systems:
5447 @item @code{SystemTap} (@uref{http://sourceware.org/systemtap/})
5448 @acronym{SDT} probes@footnote{See
5449 @uref{http://sourceware.org/systemtap/wiki/AddingUserSpaceProbingToApps}
5450 for more information on how to add @code{SystemTap} @acronym{SDT}
5451 probes in your applications.}. @code{SystemTap} probes are usable
5452 from assembly, C and C@t{++} languages@footnote{See
5453 @uref{http://sourceware.org/systemtap/wiki/UserSpaceProbeImplementation}
5454 for a good reference on how the @acronym{SDT} probes are implemented.}.
5456 @item @code{DTrace} (@uref{http://oss.oracle.com/projects/DTrace})
5457 @acronym{USDT} probes. @code{DTrace} probes are usable from C and
5461 @cindex semaphores on static probe points
5462 Some @code{SystemTap} probes have an associated semaphore variable;
5463 for instance, this happens automatically if you defined your probe
5464 using a DTrace-style @file{.d} file. If your probe has a semaphore,
5465 @value{GDBN} will automatically enable it when you specify a
5466 breakpoint using the @samp{-probe-stap} notation. But, if you put a
5467 breakpoint at a probe's location by some other method (e.g.,
5468 @code{break file:line}), then @value{GDBN} will not automatically set
5469 the semaphore. @code{DTrace} probes do not support semaphores.
5471 You can examine the available static static probes using @code{info
5472 probes}, with optional arguments:
5476 @item info probes @r{[}@var{type}@r{]} @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5477 If given, @var{type} is either @code{stap} for listing
5478 @code{SystemTap} probes or @code{dtrace} for listing @code{DTrace}
5479 probes. If omitted all probes are listed regardless of their types.
5481 If given, @var{provider} is a regular expression used to match against provider
5482 names when selecting which probes to list. If omitted, probes by all
5483 probes from all providers are listed.
5485 If given, @var{name} is a regular expression to match against probe names
5486 when selecting which probes to list. If omitted, probe names are not
5487 considered when deciding whether to display them.
5489 If given, @var{objfile} is a regular expression used to select which
5490 object files (executable or shared libraries) to examine. If not
5491 given, all object files are considered.
5493 @item info probes all
5494 List the available static probes, from all types.
5497 @cindex enabling and disabling probes
5498 Some probe points can be enabled and/or disabled. The effect of
5499 enabling or disabling a probe depends on the type of probe being
5500 handled. Some @code{DTrace} probes can be enabled or
5501 disabled, but @code{SystemTap} probes cannot be disabled.
5503 You can enable (or disable) one or more probes using the following
5504 commands, with optional arguments:
5507 @kindex enable probes
5508 @item enable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5509 If given, @var{provider} is a regular expression used to match against
5510 provider names when selecting which probes to enable. If omitted,
5511 all probes from all providers are enabled.
5513 If given, @var{name} is a regular expression to match against probe
5514 names when selecting which probes to enable. If omitted, probe names
5515 are not considered when deciding whether to enable them.
5517 If given, @var{objfile} is a regular expression used to select which
5518 object files (executable or shared libraries) to examine. If not
5519 given, all object files are considered.
5521 @kindex disable probes
5522 @item disable probes @r{[}@var{provider} @r{[}@var{name} @r{[}@var{objfile}@r{]}@r{]}@r{]}
5523 See the @code{enable probes} command above for a description of the
5524 optional arguments accepted by this command.
5527 @vindex $_probe_arg@r{, convenience variable}
5528 A probe may specify up to twelve arguments. These are available at the
5529 point at which the probe is defined---that is, when the current PC is
5530 at the probe's location. The arguments are available using the
5531 convenience variables (@pxref{Convenience Vars})
5532 @code{$_probe_arg0}@dots{}@code{$_probe_arg11}. In @code{SystemTap}
5533 probes each probe argument is an integer of the appropriate size;
5534 types are not preserved. In @code{DTrace} probes types are preserved
5535 provided that they are recognized as such by @value{GDBN}; otherwise
5536 the value of the probe argument will be a long integer. The
5537 convenience variable @code{$_probe_argc} holds the number of arguments
5538 at the current probe point.
5540 These variables are always available, but attempts to access them at
5541 any location other than a probe point will cause @value{GDBN} to give
5545 @c @ifclear BARETARGET
5546 @node Error in Breakpoints
5547 @subsection ``Cannot insert breakpoints''
5549 If you request too many active hardware-assisted breakpoints and
5550 watchpoints, you will see this error message:
5552 @c FIXME: the precise wording of this message may change; the relevant
5553 @c source change is not committed yet (Sep 3, 1999).
5555 Stopped; cannot insert breakpoints.
5556 You may have requested too many hardware breakpoints and watchpoints.
5560 This message is printed when you attempt to resume the program, since
5561 only then @value{GDBN} knows exactly how many hardware breakpoints and
5562 watchpoints it needs to insert.
5564 When this message is printed, you need to disable or remove some of the
5565 hardware-assisted breakpoints and watchpoints, and then continue.
5567 @node Breakpoint-related Warnings
5568 @subsection ``Breakpoint address adjusted...''
5569 @cindex breakpoint address adjusted
5571 Some processor architectures place constraints on the addresses at
5572 which breakpoints may be placed. For architectures thus constrained,
5573 @value{GDBN} will attempt to adjust the breakpoint's address to comply
5574 with the constraints dictated by the architecture.
5576 One example of such an architecture is the Fujitsu FR-V. The FR-V is
5577 a VLIW architecture in which a number of RISC-like instructions may be
5578 bundled together for parallel execution. The FR-V architecture
5579 constrains the location of a breakpoint instruction within such a
5580 bundle to the instruction with the lowest address. @value{GDBN}
5581 honors this constraint by adjusting a breakpoint's address to the
5582 first in the bundle.
5584 It is not uncommon for optimized code to have bundles which contain
5585 instructions from different source statements, thus it may happen that
5586 a breakpoint's address will be adjusted from one source statement to
5587 another. Since this adjustment may significantly alter @value{GDBN}'s
5588 breakpoint related behavior from what the user expects, a warning is
5589 printed when the breakpoint is first set and also when the breakpoint
5592 A warning like the one below is printed when setting a breakpoint
5593 that's been subject to address adjustment:
5596 warning: Breakpoint address adjusted from 0x00010414 to 0x00010410.
5599 Such warnings are printed both for user settable and @value{GDBN}'s
5600 internal breakpoints. If you see one of these warnings, you should
5601 verify that a breakpoint set at the adjusted address will have the
5602 desired affect. If not, the breakpoint in question may be removed and
5603 other breakpoints may be set which will have the desired behavior.
5604 E.g., it may be sufficient to place the breakpoint at a later
5605 instruction. A conditional breakpoint may also be useful in some
5606 cases to prevent the breakpoint from triggering too often.
5608 @value{GDBN} will also issue a warning when stopping at one of these
5609 adjusted breakpoints:
5612 warning: Breakpoint 1 address previously adjusted from 0x00010414
5616 When this warning is encountered, it may be too late to take remedial
5617 action except in cases where the breakpoint is hit earlier or more
5618 frequently than expected.
5620 @node Continuing and Stepping
5621 @section Continuing and Stepping
5625 @cindex resuming execution
5626 @dfn{Continuing} means resuming program execution until your program
5627 completes normally. In contrast, @dfn{stepping} means executing just
5628 one more ``step'' of your program, where ``step'' may mean either one
5629 line of source code, or one machine instruction (depending on what
5630 particular command you use). Either when continuing or when stepping,
5631 your program may stop even sooner, due to a breakpoint or a signal. (If
5632 it stops due to a signal, you may want to use @code{handle}, or use
5633 @samp{signal 0} to resume execution (@pxref{Signals, ,Signals}),
5634 or you may step into the signal's handler (@pxref{stepping and signal
5639 @kindex c @r{(@code{continue})}
5640 @kindex fg @r{(resume foreground execution)}
5641 @item continue @r{[}@var{ignore-count}@r{]}
5642 @itemx c @r{[}@var{ignore-count}@r{]}
5643 @itemx fg @r{[}@var{ignore-count}@r{]}
5644 Resume program execution, at the address where your program last stopped;
5645 any breakpoints set at that address are bypassed. The optional argument
5646 @var{ignore-count} allows you to specify a further number of times to
5647 ignore a breakpoint at this location; its effect is like that of
5648 @code{ignore} (@pxref{Conditions, ,Break Conditions}).
5650 The argument @var{ignore-count} is meaningful only when your program
5651 stopped due to a breakpoint. At other times, the argument to
5652 @code{continue} is ignored.
5654 The synonyms @code{c} and @code{fg} (for @dfn{foreground}, as the
5655 debugged program is deemed to be the foreground program) are provided
5656 purely for convenience, and have exactly the same behavior as
5660 To resume execution at a different place, you can use @code{return}
5661 (@pxref{Returning, ,Returning from a Function}) to go back to the
5662 calling function; or @code{jump} (@pxref{Jumping, ,Continuing at a
5663 Different Address}) to go to an arbitrary location in your program.
5665 A typical technique for using stepping is to set a breakpoint
5666 (@pxref{Breakpoints, ,Breakpoints; Watchpoints; and Catchpoints}) at the
5667 beginning of the function or the section of your program where a problem
5668 is believed to lie, run your program until it stops at that breakpoint,
5669 and then step through the suspect area, examining the variables that are
5670 interesting, until you see the problem happen.
5674 @kindex s @r{(@code{step})}
5676 Continue running your program until control reaches a different source
5677 line, then stop it and return control to @value{GDBN}. This command is
5678 abbreviated @code{s}.
5681 @c "without debugging information" is imprecise; actually "without line
5682 @c numbers in the debugging information". (gcc -g1 has debugging info but
5683 @c not line numbers). But it seems complex to try to make that
5684 @c distinction here.
5685 @emph{Warning:} If you use the @code{step} command while control is
5686 within a function that was compiled without debugging information,
5687 execution proceeds until control reaches a function that does have
5688 debugging information. Likewise, it will not step into a function which
5689 is compiled without debugging information. To step through functions
5690 without debugging information, use the @code{stepi} command, described
5694 The @code{step} command only stops at the first instruction of a source
5695 line. This prevents the multiple stops that could otherwise occur in
5696 @code{switch} statements, @code{for} loops, etc. @code{step} continues
5697 to stop if a function that has debugging information is called within
5698 the line. In other words, @code{step} @emph{steps inside} any functions
5699 called within the line.
5701 Also, the @code{step} command only enters a function if there is line
5702 number information for the function. Otherwise it acts like the
5703 @code{next} command. This avoids problems when using @code{cc -gl}
5704 on @acronym{MIPS} machines. Previously, @code{step} entered subroutines if there
5705 was any debugging information about the routine.
5707 @item step @var{count}
5708 Continue running as in @code{step}, but do so @var{count} times. If a
5709 breakpoint is reached, or a signal not related to stepping occurs before
5710 @var{count} steps, stepping stops right away.
5713 @kindex n @r{(@code{next})}
5714 @item next @r{[}@var{count}@r{]}
5715 Continue to the next source line in the current (innermost) stack frame.
5716 This is similar to @code{step}, but function calls that appear within
5717 the line of code are executed without stopping. Execution stops when
5718 control reaches a different line of code at the original stack level
5719 that was executing when you gave the @code{next} command. This command
5720 is abbreviated @code{n}.
5722 An argument @var{count} is a repeat count, as for @code{step}.
5725 @c FIX ME!! Do we delete this, or is there a way it fits in with
5726 @c the following paragraph? --- Vctoria
5728 @c @code{next} within a function that lacks debugging information acts like
5729 @c @code{step}, but any function calls appearing within the code of the
5730 @c function are executed without stopping.
5732 The @code{next} command only stops at the first instruction of a
5733 source line. This prevents multiple stops that could otherwise occur in
5734 @code{switch} statements, @code{for} loops, etc.
5736 @kindex set step-mode
5738 @cindex functions without line info, and stepping
5739 @cindex stepping into functions with no line info
5740 @itemx set step-mode on
5741 The @code{set step-mode on} command causes the @code{step} command to
5742 stop at the first instruction of a function which contains no debug line
5743 information rather than stepping over it.
5745 This is useful in cases where you may be interested in inspecting the
5746 machine instructions of a function which has no symbolic info and do not
5747 want @value{GDBN} to automatically skip over this function.
5749 @item set step-mode off
5750 Causes the @code{step} command to step over any functions which contains no
5751 debug information. This is the default.
5753 @item show step-mode
5754 Show whether @value{GDBN} will stop in or step over functions without
5755 source line debug information.
5758 @kindex fin @r{(@code{finish})}
5760 Continue running until just after function in the selected stack frame
5761 returns. Print the returned value (if any). This command can be
5762 abbreviated as @code{fin}.
5764 Contrast this with the @code{return} command (@pxref{Returning,
5765 ,Returning from a Function}).
5767 @kindex set print finish
5768 @kindex show print finish
5769 @item set print finish @r{[}on|off@r{]}
5770 @itemx show print finish
5771 By default the @code{finish} command will show the value that is
5772 returned by the function. This can be disabled using @code{set print
5773 finish off}. When disabled, the value is still entered into the value
5774 history (@pxref{Value History}), but not displayed.
5777 @kindex u @r{(@code{until})}
5778 @cindex run until specified location
5781 Continue running until a source line past the current line, in the
5782 current stack frame, is reached. This command is used to avoid single
5783 stepping through a loop more than once. It is like the @code{next}
5784 command, except that when @code{until} encounters a jump, it
5785 automatically continues execution until the program counter is greater
5786 than the address of the jump.
5788 This means that when you reach the end of a loop after single stepping
5789 though it, @code{until} makes your program continue execution until it
5790 exits the loop. In contrast, a @code{next} command at the end of a loop
5791 simply steps back to the beginning of the loop, which forces you to step
5792 through the next iteration.
5794 @code{until} always stops your program if it attempts to exit the current
5797 @code{until} may produce somewhat counterintuitive results if the order
5798 of machine code does not match the order of the source lines. For
5799 example, in the following excerpt from a debugging session, the @code{f}
5800 (@code{frame}) command shows that execution is stopped at line
5801 @code{206}; yet when we use @code{until}, we get to line @code{195}:
5805 #0 main (argc=4, argv=0xf7fffae8) at m4.c:206
5807 (@value{GDBP}) until
5808 195 for ( ; argc > 0; NEXTARG) @{
5811 This happened because, for execution efficiency, the compiler had
5812 generated code for the loop closure test at the end, rather than the
5813 start, of the loop---even though the test in a C @code{for}-loop is
5814 written before the body of the loop. The @code{until} command appeared
5815 to step back to the beginning of the loop when it advanced to this
5816 expression; however, it has not really gone to an earlier
5817 statement---not in terms of the actual machine code.
5819 @code{until} with no argument works by means of single
5820 instruction stepping, and hence is slower than @code{until} with an
5823 @item until @var{location}
5824 @itemx u @var{location}
5825 Continue running your program until either the specified @var{location} is
5826 reached, or the current stack frame returns. The location is any of
5827 the forms described in @ref{Specify Location}.
5828 This form of the command uses temporary breakpoints, and
5829 hence is quicker than @code{until} without an argument. The specified
5830 location is actually reached only if it is in the current frame. This
5831 implies that @code{until} can be used to skip over recursive function
5832 invocations. For instance in the code below, if the current location is
5833 line @code{96}, issuing @code{until 99} will execute the program up to
5834 line @code{99} in the same invocation of factorial, i.e., after the inner
5835 invocations have returned.
5838 94 int factorial (int value)
5840 96 if (value > 1) @{
5841 97 value *= factorial (value - 1);
5848 @kindex advance @var{location}
5849 @item advance @var{location}
5850 Continue running the program up to the given @var{location}. An argument is
5851 required, which should be of one of the forms described in
5852 @ref{Specify Location}.
5853 Execution will also stop upon exit from the current stack
5854 frame. This command is similar to @code{until}, but @code{advance} will
5855 not skip over recursive function calls, and the target location doesn't
5856 have to be in the same frame as the current one.
5860 @kindex si @r{(@code{stepi})}
5862 @itemx stepi @var{arg}
5864 Execute one machine instruction, then stop and return to the debugger.
5866 It is often useful to do @samp{display/i $pc} when stepping by machine
5867 instructions. This makes @value{GDBN} automatically display the next
5868 instruction to be executed, each time your program stops. @xref{Auto
5869 Display,, Automatic Display}.
5871 An argument is a repeat count, as in @code{step}.
5875 @kindex ni @r{(@code{nexti})}
5877 @itemx nexti @var{arg}
5879 Execute one machine instruction, but if it is a function call,
5880 proceed until the function returns.
5882 An argument is a repeat count, as in @code{next}.
5886 @anchor{range stepping}
5887 @cindex range stepping
5888 @cindex target-assisted range stepping
5889 By default, and if available, @value{GDBN} makes use of
5890 target-assisted @dfn{range stepping}. In other words, whenever you
5891 use a stepping command (e.g., @code{step}, @code{next}), @value{GDBN}
5892 tells the target to step the corresponding range of instruction
5893 addresses instead of issuing multiple single-steps. This speeds up
5894 line stepping, particularly for remote targets. Ideally, there should
5895 be no reason you would want to turn range stepping off. However, it's
5896 possible that a bug in the debug info, a bug in the remote stub (for
5897 remote targets), or even a bug in @value{GDBN} could make line
5898 stepping behave incorrectly when target-assisted range stepping is
5899 enabled. You can use the following command to turn off range stepping
5903 @kindex set range-stepping
5904 @kindex show range-stepping
5905 @item set range-stepping
5906 @itemx show range-stepping
5907 Control whether range stepping is enabled.
5909 If @code{on}, and the target supports it, @value{GDBN} tells the
5910 target to step a range of addresses itself, instead of issuing
5911 multiple single-steps. If @code{off}, @value{GDBN} always issues
5912 single-steps, even if range stepping is supported by the target. The
5913 default is @code{on}.
5917 @node Skipping Over Functions and Files
5918 @section Skipping Over Functions and Files
5919 @cindex skipping over functions and files
5921 The program you are debugging may contain some functions which are
5922 uninteresting to debug. The @code{skip} command lets you tell @value{GDBN} to
5923 skip a function, all functions in a file or a particular function in
5924 a particular file when stepping.
5926 For example, consider the following C function:
5937 Suppose you wish to step into the functions @code{foo} and @code{bar}, but you
5938 are not interested in stepping through @code{boring}. If you run @code{step}
5939 at line 103, you'll enter @code{boring()}, but if you run @code{next}, you'll
5940 step over both @code{foo} and @code{boring}!
5942 One solution is to @code{step} into @code{boring} and use the @code{finish}
5943 command to immediately exit it. But this can become tedious if @code{boring}
5944 is called from many places.
5946 A more flexible solution is to execute @kbd{skip boring}. This instructs
5947 @value{GDBN} never to step into @code{boring}. Now when you execute
5948 @code{step} at line 103, you'll step over @code{boring} and directly into
5951 Functions may be skipped by providing either a function name, linespec
5952 (@pxref{Specify Location}), regular expression that matches the function's
5953 name, file name or a @code{glob}-style pattern that matches the file name.
5955 On Posix systems the form of the regular expression is
5956 ``Extended Regular Expressions''. See for example @samp{man 7 regex}
5957 on @sc{gnu}/Linux systems. On non-Posix systems the form of the regular
5958 expression is whatever is provided by the @code{regcomp} function of
5959 the underlying system.
5960 See for example @samp{man 7 glob} on @sc{gnu}/Linux systems for a
5961 description of @code{glob}-style patterns.
5965 @item skip @r{[}@var{options}@r{]}
5966 The basic form of the @code{skip} command takes zero or more options
5967 that specify what to skip.
5968 The @var{options} argument is any useful combination of the following:
5971 @item -file @var{file}
5972 @itemx -fi @var{file}
5973 Functions in @var{file} will be skipped over when stepping.
5975 @item -gfile @var{file-glob-pattern}
5976 @itemx -gfi @var{file-glob-pattern}
5977 @cindex skipping over files via glob-style patterns
5978 Functions in files matching @var{file-glob-pattern} will be skipped
5982 (gdb) skip -gfi utils/*.c
5985 @item -function @var{linespec}
5986 @itemx -fu @var{linespec}
5987 Functions named by @var{linespec} or the function containing the line
5988 named by @var{linespec} will be skipped over when stepping.
5989 @xref{Specify Location}.
5991 @item -rfunction @var{regexp}
5992 @itemx -rfu @var{regexp}
5993 @cindex skipping over functions via regular expressions
5994 Functions whose name matches @var{regexp} will be skipped over when stepping.
5996 This form is useful for complex function names.
5997 For example, there is generally no need to step into C@t{++} @code{std::string}
5998 constructors or destructors. Plus with C@t{++} templates it can be hard to
5999 write out the full name of the function, and often it doesn't matter what
6000 the template arguments are. Specifying the function to be skipped as a
6001 regular expression makes this easier.
6004 (gdb) skip -rfu ^std::(allocator|basic_string)<.*>::~?\1 *\(
6007 If you want to skip every templated C@t{++} constructor and destructor
6008 in the @code{std} namespace you can do:
6011 (gdb) skip -rfu ^std::([a-zA-z0-9_]+)<.*>::~?\1 *\(
6015 If no options are specified, the function you're currently debugging
6018 @kindex skip function
6019 @item skip function @r{[}@var{linespec}@r{]}
6020 After running this command, the function named by @var{linespec} or the
6021 function containing the line named by @var{linespec} will be skipped over when
6022 stepping. @xref{Specify Location}.
6024 If you do not specify @var{linespec}, the function you're currently debugging
6027 (If you have a function called @code{file} that you want to skip, use
6028 @kbd{skip function file}.)
6031 @item skip file @r{[}@var{filename}@r{]}
6032 After running this command, any function whose source lives in @var{filename}
6033 will be skipped over when stepping.
6036 (gdb) skip file boring.c
6037 File boring.c will be skipped when stepping.
6040 If you do not specify @var{filename}, functions whose source lives in the file
6041 you're currently debugging will be skipped.
6044 Skips can be listed, deleted, disabled, and enabled, much like breakpoints.
6045 These are the commands for managing your list of skips:
6049 @item info skip @r{[}@var{range}@r{]}
6050 Print details about the specified skip(s). If @var{range} is not specified,
6051 print a table with details about all functions and files marked for skipping.
6052 @code{info skip} prints the following information about each skip:
6056 A number identifying this skip.
6057 @item Enabled or Disabled
6058 Enabled skips are marked with @samp{y}.
6059 Disabled skips are marked with @samp{n}.
6061 If the file name is a @samp{glob} pattern this is @samp{y}.
6062 Otherwise it is @samp{n}.
6064 The name or @samp{glob} pattern of the file to be skipped.
6065 If no file is specified this is @samp{<none>}.
6067 If the function name is a @samp{regular expression} this is @samp{y}.
6068 Otherwise it is @samp{n}.
6070 The name or regular expression of the function to skip.
6071 If no function is specified this is @samp{<none>}.
6075 @item skip delete @r{[}@var{range}@r{]}
6076 Delete the specified skip(s). If @var{range} is not specified, delete all
6080 @item skip enable @r{[}@var{range}@r{]}
6081 Enable the specified skip(s). If @var{range} is not specified, enable all
6084 @kindex skip disable
6085 @item skip disable @r{[}@var{range}@r{]}
6086 Disable the specified skip(s). If @var{range} is not specified, disable all
6089 @kindex set debug skip
6090 @item set debug skip @r{[}on|off@r{]}
6091 Set whether to print the debug output about skipping files and functions.
6093 @kindex show debug skip
6094 @item show debug skip
6095 Show whether the debug output about skipping files and functions is printed.
6103 A signal is an asynchronous event that can happen in a program. The
6104 operating system defines the possible kinds of signals, and gives each
6105 kind a name and a number. For example, in Unix @code{SIGINT} is the
6106 signal a program gets when you type an interrupt character (often @kbd{Ctrl-c});
6107 @code{SIGSEGV} is the signal a program gets from referencing a place in
6108 memory far away from all the areas in use; @code{SIGALRM} occurs when
6109 the alarm clock timer goes off (which happens only if your program has
6110 requested an alarm).
6112 @cindex fatal signals
6113 Some signals, including @code{SIGALRM}, are a normal part of the
6114 functioning of your program. Others, such as @code{SIGSEGV}, indicate
6115 errors; these signals are @dfn{fatal} (they kill your program immediately) if the
6116 program has not specified in advance some other way to handle the signal.
6117 @code{SIGINT} does not indicate an error in your program, but it is normally
6118 fatal so it can carry out the purpose of the interrupt: to kill the program.
6120 @value{GDBN} has the ability to detect any occurrence of a signal in your
6121 program. You can tell @value{GDBN} in advance what to do for each kind of
6124 @cindex handling signals
6125 Normally, @value{GDBN} is set up to let the non-erroneous signals like
6126 @code{SIGALRM} be silently passed to your program
6127 (so as not to interfere with their role in the program's functioning)
6128 but to stop your program immediately whenever an error signal happens.
6129 You can change these settings with the @code{handle} command.
6132 @kindex info signals
6136 Print a table of all the kinds of signals and how @value{GDBN} has been told to
6137 handle each one. You can use this to see the signal numbers of all
6138 the defined types of signals.
6140 @item info signals @var{sig}
6141 Similar, but print information only about the specified signal number.
6143 @code{info handle} is an alias for @code{info signals}.
6145 @item catch signal @r{[}@var{signal}@dots{} @r{|} @samp{all}@r{]}
6146 Set a catchpoint for the indicated signals. @xref{Set Catchpoints},
6147 for details about this command.
6150 @item handle @var{signal} @r{[}@var{keywords}@dots{}@r{]}
6151 Change the way @value{GDBN} handles signal @var{signal}. The @var{signal}
6152 can be the number of a signal or its name (with or without the
6153 @samp{SIG} at the beginning); a list of signal numbers of the form
6154 @samp{@var{low}-@var{high}}; or the word @samp{all}, meaning all the
6155 known signals. Optional arguments @var{keywords}, described below,
6156 say what change to make.
6160 The keywords allowed by the @code{handle} command can be abbreviated.
6161 Their full names are:
6165 @value{GDBN} should not stop your program when this signal happens. It may
6166 still print a message telling you that the signal has come in.
6169 @value{GDBN} should stop your program when this signal happens. This implies
6170 the @code{print} keyword as well.
6173 @value{GDBN} should print a message when this signal happens.
6176 @value{GDBN} should not mention the occurrence of the signal at all. This
6177 implies the @code{nostop} keyword as well.
6181 @value{GDBN} should allow your program to see this signal; your program
6182 can handle the signal, or else it may terminate if the signal is fatal
6183 and not handled. @code{pass} and @code{noignore} are synonyms.
6187 @value{GDBN} should not allow your program to see this signal.
6188 @code{nopass} and @code{ignore} are synonyms.
6192 When a signal stops your program, the signal is not visible to the
6194 continue. Your program sees the signal then, if @code{pass} is in
6195 effect for the signal in question @emph{at that time}. In other words,
6196 after @value{GDBN} reports a signal, you can use the @code{handle}
6197 command with @code{pass} or @code{nopass} to control whether your
6198 program sees that signal when you continue.
6200 The default is set to @code{nostop}, @code{noprint}, @code{pass} for
6201 non-erroneous signals such as @code{SIGALRM}, @code{SIGWINCH} and
6202 @code{SIGCHLD}, and to @code{stop}, @code{print}, @code{pass} for the
6205 You can also use the @code{signal} command to prevent your program from
6206 seeing a signal, or cause it to see a signal it normally would not see,
6207 or to give it any signal at any time. For example, if your program stopped
6208 due to some sort of memory reference error, you might store correct
6209 values into the erroneous variables and continue, hoping to see more
6210 execution; but your program would probably terminate immediately as
6211 a result of the fatal signal once it saw the signal. To prevent this,
6212 you can continue with @samp{signal 0}. @xref{Signaling, ,Giving your
6215 @cindex stepping and signal handlers
6216 @anchor{stepping and signal handlers}
6218 @value{GDBN} optimizes for stepping the mainline code. If a signal
6219 that has @code{handle nostop} and @code{handle pass} set arrives while
6220 a stepping command (e.g., @code{stepi}, @code{step}, @code{next}) is
6221 in progress, @value{GDBN} lets the signal handler run and then resumes
6222 stepping the mainline code once the signal handler returns. In other
6223 words, @value{GDBN} steps over the signal handler. This prevents
6224 signals that you've specified as not interesting (with @code{handle
6225 nostop}) from changing the focus of debugging unexpectedly. Note that
6226 the signal handler itself may still hit a breakpoint, stop for another
6227 signal that has @code{handle stop} in effect, or for any other event
6228 that normally results in stopping the stepping command sooner. Also
6229 note that @value{GDBN} still informs you that the program received a
6230 signal if @code{handle print} is set.
6232 @anchor{stepping into signal handlers}
6234 If you set @code{handle pass} for a signal, and your program sets up a
6235 handler for it, then issuing a stepping command, such as @code{step}
6236 or @code{stepi}, when your program is stopped due to the signal will
6237 step @emph{into} the signal handler (if the target supports that).
6239 Likewise, if you use the @code{queue-signal} command to queue a signal
6240 to be delivered to the current thread when execution of the thread
6241 resumes (@pxref{Signaling, ,Giving your Program a Signal}), then a
6242 stepping command will step into the signal handler.
6244 Here's an example, using @code{stepi} to step to the first instruction
6245 of @code{SIGUSR1}'s handler:
6248 (@value{GDBP}) handle SIGUSR1
6249 Signal Stop Print Pass to program Description
6250 SIGUSR1 Yes Yes Yes User defined signal 1
6254 Program received signal SIGUSR1, User defined signal 1.
6255 main () sigusr1.c:28
6258 sigusr1_handler () at sigusr1.c:9
6262 The same, but using @code{queue-signal} instead of waiting for the
6263 program to receive the signal first:
6268 (@value{GDBP}) queue-signal SIGUSR1
6270 sigusr1_handler () at sigusr1.c:9
6275 @cindex extra signal information
6276 @anchor{extra signal information}
6278 On some targets, @value{GDBN} can inspect extra signal information
6279 associated with the intercepted signal, before it is actually
6280 delivered to the program being debugged. This information is exported
6281 by the convenience variable @code{$_siginfo}, and consists of data
6282 that is passed by the kernel to the signal handler at the time of the
6283 receipt of a signal. The data type of the information itself is
6284 target dependent. You can see the data type using the @code{ptype
6285 $_siginfo} command. On Unix systems, it typically corresponds to the
6286 standard @code{siginfo_t} type, as defined in the @file{signal.h}
6289 Here's an example, on a @sc{gnu}/Linux system, printing the stray
6290 referenced address that raised a segmentation fault.
6294 (@value{GDBP}) continue
6295 Program received signal SIGSEGV, Segmentation fault.
6296 0x0000000000400766 in main ()
6298 (@value{GDBP}) ptype $_siginfo
6305 struct @{...@} _kill;
6306 struct @{...@} _timer;
6308 struct @{...@} _sigchld;
6309 struct @{...@} _sigfault;
6310 struct @{...@} _sigpoll;
6313 (@value{GDBP}) ptype $_siginfo._sifields._sigfault
6317 (@value{GDBP}) p $_siginfo._sifields._sigfault.si_addr
6318 $1 = (void *) 0x7ffff7ff7000
6322 Depending on target support, @code{$_siginfo} may also be writable.
6324 @cindex Intel MPX boundary violations
6325 @cindex boundary violations, Intel MPX
6326 On some targets, a @code{SIGSEGV} can be caused by a boundary
6327 violation, i.e., accessing an address outside of the allowed range.
6328 In those cases @value{GDBN} may displays additional information,
6329 depending on how @value{GDBN} has been told to handle the signal.
6330 With @code{handle stop SIGSEGV}, @value{GDBN} displays the violation
6331 kind: "Upper" or "Lower", the memory address accessed and the
6332 bounds, while with @code{handle nostop SIGSEGV} no additional
6333 information is displayed.
6335 The usual output of a segfault is:
6337 Program received signal SIGSEGV, Segmentation fault
6338 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6339 68 value = *(p + len);
6342 While a bound violation is presented as:
6344 Program received signal SIGSEGV, Segmentation fault
6345 Upper bound violation while accessing address 0x7fffffffc3b3
6346 Bounds: [lower = 0x7fffffffc390, upper = 0x7fffffffc3a3]
6347 0x0000000000400d7c in upper () at i386-mpx-sigsegv.c:68
6348 68 value = *(p + len);
6352 @section Stopping and Starting Multi-thread Programs
6354 @cindex stopped threads
6355 @cindex threads, stopped
6357 @cindex continuing threads
6358 @cindex threads, continuing
6360 @value{GDBN} supports debugging programs with multiple threads
6361 (@pxref{Threads,, Debugging Programs with Multiple Threads}). There
6362 are two modes of controlling execution of your program within the
6363 debugger. In the default mode, referred to as @dfn{all-stop mode},
6364 when any thread in your program stops (for example, at a breakpoint
6365 or while being stepped), all other threads in the program are also stopped by
6366 @value{GDBN}. On some targets, @value{GDBN} also supports
6367 @dfn{non-stop mode}, in which other threads can continue to run freely while
6368 you examine the stopped thread in the debugger.
6371 * All-Stop Mode:: All threads stop when GDB takes control
6372 * Non-Stop Mode:: Other threads continue to execute
6373 * Background Execution:: Running your program asynchronously
6374 * Thread-Specific Breakpoints:: Controlling breakpoints
6375 * Interrupted System Calls:: GDB may interfere with system calls
6376 * Observer Mode:: GDB does not alter program behavior
6380 @subsection All-Stop Mode
6382 @cindex all-stop mode
6384 In all-stop mode, whenever your program stops under @value{GDBN} for any reason,
6385 @emph{all} threads of execution stop, not just the current thread. This
6386 allows you to examine the overall state of the program, including
6387 switching between threads, without worrying that things may change
6390 Conversely, whenever you restart the program, @emph{all} threads start
6391 executing. @emph{This is true even when single-stepping} with commands
6392 like @code{step} or @code{next}.
6394 In particular, @value{GDBN} cannot single-step all threads in lockstep.
6395 Since thread scheduling is up to your debugging target's operating
6396 system (not controlled by @value{GDBN}), other threads may
6397 execute more than one statement while the current thread completes a
6398 single step. Moreover, in general other threads stop in the middle of a
6399 statement, rather than at a clean statement boundary, when the program
6402 You might even find your program stopped in another thread after
6403 continuing or even single-stepping. This happens whenever some other
6404 thread runs into a breakpoint, a signal, or an exception before the
6405 first thread completes whatever you requested.
6407 @cindex automatic thread selection
6408 @cindex switching threads automatically
6409 @cindex threads, automatic switching
6410 Whenever @value{GDBN} stops your program, due to a breakpoint or a
6411 signal, it automatically selects the thread where that breakpoint or
6412 signal happened. @value{GDBN} alerts you to the context switch with a
6413 message such as @samp{[Switching to Thread @var{n}]} to identify the
6416 On some OSes, you can modify @value{GDBN}'s default behavior by
6417 locking the OS scheduler to allow only a single thread to run.
6420 @item set scheduler-locking @var{mode}
6421 @cindex scheduler locking mode
6422 @cindex lock scheduler
6423 Set the scheduler locking mode. It applies to normal execution,
6424 record mode, and replay mode. If it is @code{off}, then there is no
6425 locking and any thread may run at any time. If @code{on}, then only
6426 the current thread may run when the inferior is resumed. The
6427 @code{step} mode optimizes for single-stepping; it prevents other
6428 threads from preempting the current thread while you are stepping, so
6429 that the focus of debugging does not change unexpectedly. Other
6430 threads never get a chance to run when you step, and they are
6431 completely free to run when you use commands like @samp{continue},
6432 @samp{until}, or @samp{finish}. However, unless another thread hits a
6433 breakpoint during its timeslice, @value{GDBN} does not change the
6434 current thread away from the thread that you are debugging. The
6435 @code{replay} mode behaves like @code{off} in record mode and like
6436 @code{on} in replay mode.
6438 @item show scheduler-locking
6439 Display the current scheduler locking mode.
6442 @cindex resume threads of multiple processes simultaneously
6443 By default, when you issue one of the execution commands such as
6444 @code{continue}, @code{next} or @code{step}, @value{GDBN} allows only
6445 threads of the current inferior to run. For example, if @value{GDBN}
6446 is attached to two inferiors, each with two threads, the
6447 @code{continue} command resumes only the two threads of the current
6448 inferior. This is useful, for example, when you debug a program that
6449 forks and you want to hold the parent stopped (so that, for instance,
6450 it doesn't run to exit), while you debug the child. In other
6451 situations, you may not be interested in inspecting the current state
6452 of any of the processes @value{GDBN} is attached to, and you may want
6453 to resume them all until some breakpoint is hit. In the latter case,
6454 you can instruct @value{GDBN} to allow all threads of all the
6455 inferiors to run with the @w{@code{set schedule-multiple}} command.
6458 @kindex set schedule-multiple
6459 @item set schedule-multiple
6460 Set the mode for allowing threads of multiple processes to be resumed
6461 when an execution command is issued. When @code{on}, all threads of
6462 all processes are allowed to run. When @code{off}, only the threads
6463 of the current process are resumed. The default is @code{off}. The
6464 @code{scheduler-locking} mode takes precedence when set to @code{on},
6465 or while you are stepping and set to @code{step}.
6467 @item show schedule-multiple
6468 Display the current mode for resuming the execution of threads of
6473 @subsection Non-Stop Mode
6475 @cindex non-stop mode
6477 @c This section is really only a place-holder, and needs to be expanded
6478 @c with more details.
6480 For some multi-threaded targets, @value{GDBN} supports an optional
6481 mode of operation in which you can examine stopped program threads in
6482 the debugger while other threads continue to execute freely. This
6483 minimizes intrusion when debugging live systems, such as programs
6484 where some threads have real-time constraints or must continue to
6485 respond to external events. This is referred to as @dfn{non-stop} mode.
6487 In non-stop mode, when a thread stops to report a debugging event,
6488 @emph{only} that thread is stopped; @value{GDBN} does not stop other
6489 threads as well, in contrast to the all-stop mode behavior. Additionally,
6490 execution commands such as @code{continue} and @code{step} apply by default
6491 only to the current thread in non-stop mode, rather than all threads as
6492 in all-stop mode. This allows you to control threads explicitly in
6493 ways that are not possible in all-stop mode --- for example, stepping
6494 one thread while allowing others to run freely, stepping
6495 one thread while holding all others stopped, or stepping several threads
6496 independently and simultaneously.
6498 To enter non-stop mode, use this sequence of commands before you run
6499 or attach to your program:
6502 # If using the CLI, pagination breaks non-stop.
6505 # Finally, turn it on!
6509 You can use these commands to manipulate the non-stop mode setting:
6512 @kindex set non-stop
6513 @item set non-stop on
6514 Enable selection of non-stop mode.
6515 @item set non-stop off
6516 Disable selection of non-stop mode.
6517 @kindex show non-stop
6519 Show the current non-stop enablement setting.
6522 Note these commands only reflect whether non-stop mode is enabled,
6523 not whether the currently-executing program is being run in non-stop mode.
6524 In particular, the @code{set non-stop} preference is only consulted when
6525 @value{GDBN} starts or connects to the target program, and it is generally
6526 not possible to switch modes once debugging has started. Furthermore,
6527 since not all targets support non-stop mode, even when you have enabled
6528 non-stop mode, @value{GDBN} may still fall back to all-stop operation by
6531 In non-stop mode, all execution commands apply only to the current thread
6532 by default. That is, @code{continue} only continues one thread.
6533 To continue all threads, issue @code{continue -a} or @code{c -a}.
6535 You can use @value{GDBN}'s background execution commands
6536 (@pxref{Background Execution}) to run some threads in the background
6537 while you continue to examine or step others from @value{GDBN}.
6538 The MI execution commands (@pxref{GDB/MI Program Execution}) are
6539 always executed asynchronously in non-stop mode.
6541 Suspending execution is done with the @code{interrupt} command when
6542 running in the background, or @kbd{Ctrl-c} during foreground execution.
6543 In all-stop mode, this stops the whole process;
6544 but in non-stop mode the interrupt applies only to the current thread.
6545 To stop the whole program, use @code{interrupt -a}.
6547 Other execution commands do not currently support the @code{-a} option.
6549 In non-stop mode, when a thread stops, @value{GDBN} doesn't automatically make
6550 that thread current, as it does in all-stop mode. This is because the
6551 thread stop notifications are asynchronous with respect to @value{GDBN}'s
6552 command interpreter, and it would be confusing if @value{GDBN} unexpectedly
6553 changed to a different thread just as you entered a command to operate on the
6554 previously current thread.
6556 @node Background Execution
6557 @subsection Background Execution
6559 @cindex foreground execution
6560 @cindex background execution
6561 @cindex asynchronous execution
6562 @cindex execution, foreground, background and asynchronous
6564 @value{GDBN}'s execution commands have two variants: the normal
6565 foreground (synchronous) behavior, and a background
6566 (asynchronous) behavior. In foreground execution, @value{GDBN} waits for
6567 the program to report that some thread has stopped before prompting for
6568 another command. In background execution, @value{GDBN} immediately gives
6569 a command prompt so that you can issue other commands while your program runs.
6571 If the target doesn't support async mode, @value{GDBN} issues an error
6572 message if you attempt to use the background execution commands.
6574 @cindex @code{&}, background execution of commands
6575 To specify background execution, add a @code{&} to the command. For example,
6576 the background form of the @code{continue} command is @code{continue&}, or
6577 just @code{c&}. The execution commands that accept background execution
6583 @xref{Starting, , Starting your Program}.
6587 @xref{Attach, , Debugging an Already-running Process}.
6591 @xref{Continuing and Stepping, step}.
6595 @xref{Continuing and Stepping, stepi}.
6599 @xref{Continuing and Stepping, next}.
6603 @xref{Continuing and Stepping, nexti}.
6607 @xref{Continuing and Stepping, continue}.
6611 @xref{Continuing and Stepping, finish}.
6615 @xref{Continuing and Stepping, until}.
6619 Background execution is especially useful in conjunction with non-stop
6620 mode for debugging programs with multiple threads; see @ref{Non-Stop Mode}.
6621 However, you can also use these commands in the normal all-stop mode with
6622 the restriction that you cannot issue another execution command until the
6623 previous one finishes. Examples of commands that are valid in all-stop
6624 mode while the program is running include @code{help} and @code{info break}.
6626 You can interrupt your program while it is running in the background by
6627 using the @code{interrupt} command.
6634 Suspend execution of the running program. In all-stop mode,
6635 @code{interrupt} stops the whole process, but in non-stop mode, it stops
6636 only the current thread. To stop the whole program in non-stop mode,
6637 use @code{interrupt -a}.
6640 @node Thread-Specific Breakpoints
6641 @subsection Thread-Specific Breakpoints
6643 When your program has multiple threads (@pxref{Threads,, Debugging
6644 Programs with Multiple Threads}), you can choose whether to set
6645 breakpoints on all threads, or on a particular thread.
6648 @cindex breakpoints and threads
6649 @cindex thread breakpoints
6650 @kindex break @dots{} thread @var{thread-id}
6651 @item break @var{location} thread @var{thread-id}
6652 @itemx break @var{location} thread @var{thread-id} if @dots{}
6653 @var{location} specifies source lines; there are several ways of
6654 writing them (@pxref{Specify Location}), but the effect is always to
6655 specify some source line.
6657 Use the qualifier @samp{thread @var{thread-id}} with a breakpoint command
6658 to specify that you only want @value{GDBN} to stop the program when a
6659 particular thread reaches this breakpoint. The @var{thread-id} specifier
6660 is one of the thread identifiers assigned by @value{GDBN}, shown
6661 in the first column of the @samp{info threads} display.
6663 If you do not specify @samp{thread @var{thread-id}} when you set a
6664 breakpoint, the breakpoint applies to @emph{all} threads of your
6667 You can use the @code{thread} qualifier on conditional breakpoints as
6668 well; in this case, place @samp{thread @var{thread-id}} before or
6669 after the breakpoint condition, like this:
6672 (@value{GDBP}) break frik.c:13 thread 28 if bartab > lim
6677 Thread-specific breakpoints are automatically deleted when
6678 @value{GDBN} detects the corresponding thread is no longer in the
6679 thread list. For example:
6683 Thread-specific breakpoint 3 deleted - thread 28 no longer in the thread list.
6686 There are several ways for a thread to disappear, such as a regular
6687 thread exit, but also when you detach from the process with the
6688 @code{detach} command (@pxref{Attach, ,Debugging an Already-running
6689 Process}), or if @value{GDBN} loses the remote connection
6690 (@pxref{Remote Debugging}), etc. Note that with some targets,
6691 @value{GDBN} is only able to detect a thread has exited when the user
6692 explictly asks for the thread list with the @code{info threads}
6695 @node Interrupted System Calls
6696 @subsection Interrupted System Calls
6698 @cindex thread breakpoints and system calls
6699 @cindex system calls and thread breakpoints
6700 @cindex premature return from system calls
6701 There is an unfortunate side effect when using @value{GDBN} to debug
6702 multi-threaded programs. If one thread stops for a
6703 breakpoint, or for some other reason, and another thread is blocked in a
6704 system call, then the system call may return prematurely. This is a
6705 consequence of the interaction between multiple threads and the signals
6706 that @value{GDBN} uses to implement breakpoints and other events that
6709 To handle this problem, your program should check the return value of
6710 each system call and react appropriately. This is good programming
6713 For example, do not write code like this:
6719 The call to @code{sleep} will return early if a different thread stops
6720 at a breakpoint or for some other reason.
6722 Instead, write this:
6727 unslept = sleep (unslept);
6730 A system call is allowed to return early, so the system is still
6731 conforming to its specification. But @value{GDBN} does cause your
6732 multi-threaded program to behave differently than it would without
6735 Also, @value{GDBN} uses internal breakpoints in the thread library to
6736 monitor certain events such as thread creation and thread destruction.
6737 When such an event happens, a system call in another thread may return
6738 prematurely, even though your program does not appear to stop.
6741 @subsection Observer Mode
6743 If you want to build on non-stop mode and observe program behavior
6744 without any chance of disruption by @value{GDBN}, you can set
6745 variables to disable all of the debugger's attempts to modify state,
6746 whether by writing memory, inserting breakpoints, etc. These operate
6747 at a low level, intercepting operations from all commands.
6749 When all of these are set to @code{off}, then @value{GDBN} is said to
6750 be @dfn{observer mode}. As a convenience, the variable
6751 @code{observer} can be set to disable these, plus enable non-stop
6754 Note that @value{GDBN} will not prevent you from making nonsensical
6755 combinations of these settings. For instance, if you have enabled
6756 @code{may-insert-breakpoints} but disabled @code{may-write-memory},
6757 then breakpoints that work by writing trap instructions into the code
6758 stream will still not be able to be placed.
6763 @item set observer on
6764 @itemx set observer off
6765 When set to @code{on}, this disables all the permission variables
6766 below (except for @code{insert-fast-tracepoints}), plus enables
6767 non-stop debugging. Setting this to @code{off} switches back to
6768 normal debugging, though remaining in non-stop mode.
6771 Show whether observer mode is on or off.
6773 @kindex may-write-registers
6774 @item set may-write-registers on
6775 @itemx set may-write-registers off
6776 This controls whether @value{GDBN} will attempt to alter the values of
6777 registers, such as with assignment expressions in @code{print}, or the
6778 @code{jump} command. It defaults to @code{on}.
6780 @item show may-write-registers
6781 Show the current permission to write registers.
6783 @kindex may-write-memory
6784 @item set may-write-memory on
6785 @itemx set may-write-memory off
6786 This controls whether @value{GDBN} will attempt to alter the contents
6787 of memory, such as with assignment expressions in @code{print}. It
6788 defaults to @code{on}.
6790 @item show may-write-memory
6791 Show the current permission to write memory.
6793 @kindex may-insert-breakpoints
6794 @item set may-insert-breakpoints on
6795 @itemx set may-insert-breakpoints off
6796 This controls whether @value{GDBN} will attempt to insert breakpoints.
6797 This affects all breakpoints, including internal breakpoints defined
6798 by @value{GDBN}. It defaults to @code{on}.
6800 @item show may-insert-breakpoints
6801 Show the current permission to insert breakpoints.
6803 @kindex may-insert-tracepoints
6804 @item set may-insert-tracepoints on
6805 @itemx set may-insert-tracepoints off
6806 This controls whether @value{GDBN} will attempt to insert (regular)
6807 tracepoints at the beginning of a tracing experiment. It affects only
6808 non-fast tracepoints, fast tracepoints being under the control of
6809 @code{may-insert-fast-tracepoints}. It defaults to @code{on}.
6811 @item show may-insert-tracepoints
6812 Show the current permission to insert tracepoints.
6814 @kindex may-insert-fast-tracepoints
6815 @item set may-insert-fast-tracepoints on
6816 @itemx set may-insert-fast-tracepoints off
6817 This controls whether @value{GDBN} will attempt to insert fast
6818 tracepoints at the beginning of a tracing experiment. It affects only
6819 fast tracepoints, regular (non-fast) tracepoints being under the
6820 control of @code{may-insert-tracepoints}. It defaults to @code{on}.
6822 @item show may-insert-fast-tracepoints
6823 Show the current permission to insert fast tracepoints.
6825 @kindex may-interrupt
6826 @item set may-interrupt on
6827 @itemx set may-interrupt off
6828 This controls whether @value{GDBN} will attempt to interrupt or stop
6829 program execution. When this variable is @code{off}, the
6830 @code{interrupt} command will have no effect, nor will
6831 @kbd{Ctrl-c}. It defaults to @code{on}.
6833 @item show may-interrupt
6834 Show the current permission to interrupt or stop the program.
6838 @node Reverse Execution
6839 @chapter Running programs backward
6840 @cindex reverse execution
6841 @cindex running programs backward
6843 When you are debugging a program, it is not unusual to realize that
6844 you have gone too far, and some event of interest has already happened.
6845 If the target environment supports it, @value{GDBN} can allow you to
6846 ``rewind'' the program by running it backward.
6848 A target environment that supports reverse execution should be able
6849 to ``undo'' the changes in machine state that have taken place as the
6850 program was executing normally. Variables, registers etc.@: should
6851 revert to their previous values. Obviously this requires a great
6852 deal of sophistication on the part of the target environment; not
6853 all target environments can support reverse execution.
6855 When a program is executed in reverse, the instructions that
6856 have most recently been executed are ``un-executed'', in reverse
6857 order. The program counter runs backward, following the previous
6858 thread of execution in reverse. As each instruction is ``un-executed'',
6859 the values of memory and/or registers that were changed by that
6860 instruction are reverted to their previous states. After executing
6861 a piece of source code in reverse, all side effects of that code
6862 should be ``undone'', and all variables should be returned to their
6863 prior values@footnote{
6864 Note that some side effects are easier to undo than others. For instance,
6865 memory and registers are relatively easy, but device I/O is hard. Some
6866 targets may be able undo things like device I/O, and some may not.
6868 The contract between @value{GDBN} and the reverse executing target
6869 requires only that the target do something reasonable when
6870 @value{GDBN} tells it to execute backwards, and then report the
6871 results back to @value{GDBN}. Whatever the target reports back to
6872 @value{GDBN}, @value{GDBN} will report back to the user. @value{GDBN}
6873 assumes that the memory and registers that the target reports are in a
6874 consistant state, but @value{GDBN} accepts whatever it is given.
6877 On some platforms, @value{GDBN} has built-in support for reverse
6878 execution, activated with the @code{record} or @code{record btrace}
6879 commands. @xref{Process Record and Replay}. Some remote targets,
6880 typically full system emulators, support reverse execution directly
6881 without requiring any special command.
6883 If you are debugging in a target environment that supports
6884 reverse execution, @value{GDBN} provides the following commands.
6887 @kindex reverse-continue
6888 @kindex rc @r{(@code{reverse-continue})}
6889 @item reverse-continue @r{[}@var{ignore-count}@r{]}
6890 @itemx rc @r{[}@var{ignore-count}@r{]}
6891 Beginning at the point where your program last stopped, start executing
6892 in reverse. Reverse execution will stop for breakpoints and synchronous
6893 exceptions (signals), just like normal execution. Behavior of
6894 asynchronous signals depends on the target environment.
6896 @kindex reverse-step
6897 @kindex rs @r{(@code{step})}
6898 @item reverse-step @r{[}@var{count}@r{]}
6899 Run the program backward until control reaches the start of a
6900 different source line; then stop it, and return control to @value{GDBN}.
6902 Like the @code{step} command, @code{reverse-step} will only stop
6903 at the beginning of a source line. It ``un-executes'' the previously
6904 executed source line. If the previous source line included calls to
6905 debuggable functions, @code{reverse-step} will step (backward) into
6906 the called function, stopping at the beginning of the @emph{last}
6907 statement in the called function (typically a return statement).
6909 Also, as with the @code{step} command, if non-debuggable functions are
6910 called, @code{reverse-step} will run thru them backward without stopping.
6912 @kindex reverse-stepi
6913 @kindex rsi @r{(@code{reverse-stepi})}
6914 @item reverse-stepi @r{[}@var{count}@r{]}
6915 Reverse-execute one machine instruction. Note that the instruction
6916 to be reverse-executed is @emph{not} the one pointed to by the program
6917 counter, but the instruction executed prior to that one. For instance,
6918 if the last instruction was a jump, @code{reverse-stepi} will take you
6919 back from the destination of the jump to the jump instruction itself.
6921 @kindex reverse-next
6922 @kindex rn @r{(@code{reverse-next})}
6923 @item reverse-next @r{[}@var{count}@r{]}
6924 Run backward to the beginning of the previous line executed in
6925 the current (innermost) stack frame. If the line contains function
6926 calls, they will be ``un-executed'' without stopping. Starting from
6927 the first line of a function, @code{reverse-next} will take you back
6928 to the caller of that function, @emph{before} the function was called,
6929 just as the normal @code{next} command would take you from the last
6930 line of a function back to its return to its caller
6931 @footnote{Unless the code is too heavily optimized.}.
6933 @kindex reverse-nexti
6934 @kindex rni @r{(@code{reverse-nexti})}
6935 @item reverse-nexti @r{[}@var{count}@r{]}
6936 Like @code{nexti}, @code{reverse-nexti} executes a single instruction
6937 in reverse, except that called functions are ``un-executed'' atomically.
6938 That is, if the previously executed instruction was a return from
6939 another function, @code{reverse-nexti} will continue to execute
6940 in reverse until the call to that function (from the current stack
6943 @kindex reverse-finish
6944 @item reverse-finish
6945 Just as the @code{finish} command takes you to the point where the
6946 current function returns, @code{reverse-finish} takes you to the point
6947 where it was called. Instead of ending up at the end of the current
6948 function invocation, you end up at the beginning.
6950 @kindex set exec-direction
6951 @item set exec-direction
6952 Set the direction of target execution.
6953 @item set exec-direction reverse
6954 @cindex execute forward or backward in time
6955 @value{GDBN} will perform all execution commands in reverse, until the
6956 exec-direction mode is changed to ``forward''. Affected commands include
6957 @code{step, stepi, next, nexti, continue, and finish}. The @code{return}
6958 command cannot be used in reverse mode.
6959 @item set exec-direction forward
6960 @value{GDBN} will perform all execution commands in the normal fashion.
6961 This is the default.
6965 @node Process Record and Replay
6966 @chapter Recording Inferior's Execution and Replaying It
6967 @cindex process record and replay
6968 @cindex recording inferior's execution and replaying it
6970 On some platforms, @value{GDBN} provides a special @dfn{process record
6971 and replay} target that can record a log of the process execution, and
6972 replay it later with both forward and reverse execution commands.
6975 When this target is in use, if the execution log includes the record
6976 for the next instruction, @value{GDBN} will debug in @dfn{replay
6977 mode}. In the replay mode, the inferior does not really execute code
6978 instructions. Instead, all the events that normally happen during
6979 code execution are taken from the execution log. While code is not
6980 really executed in replay mode, the values of registers (including the
6981 program counter register) and the memory of the inferior are still
6982 changed as they normally would. Their contents are taken from the
6986 If the record for the next instruction is not in the execution log,
6987 @value{GDBN} will debug in @dfn{record mode}. In this mode, the
6988 inferior executes normally, and @value{GDBN} records the execution log
6991 The process record and replay target supports reverse execution
6992 (@pxref{Reverse Execution}), even if the platform on which the
6993 inferior runs does not. However, the reverse execution is limited in
6994 this case by the range of the instructions recorded in the execution
6995 log. In other words, reverse execution on platforms that don't
6996 support it directly can only be done in the replay mode.
6998 When debugging in the reverse direction, @value{GDBN} will work in
6999 replay mode as long as the execution log includes the record for the
7000 previous instruction; otherwise, it will work in record mode, if the
7001 platform supports reverse execution, or stop if not.
7003 Currently, process record and replay is supported on ARM, Aarch64,
7004 Moxie, PowerPC, PowerPC64, S/390, and x86 (i386/amd64) running
7005 GNU/Linux. Process record and replay can be used both when native
7006 debugging, and when remote debugging via @code{gdbserver}.
7008 For architecture environments that support process record and replay,
7009 @value{GDBN} provides the following commands:
7012 @kindex target record
7013 @kindex target record-full
7014 @kindex target record-btrace
7017 @kindex record btrace
7018 @kindex record btrace bts
7019 @kindex record btrace pt
7025 @kindex rec btrace bts
7026 @kindex rec btrace pt
7029 @item record @var{method}
7030 This command starts the process record and replay target. The
7031 recording method can be specified as parameter. Without a parameter
7032 the command uses the @code{full} recording method. The following
7033 recording methods are available:
7037 Full record/replay recording using @value{GDBN}'s software record and
7038 replay implementation. This method allows replaying and reverse
7041 @item btrace @var{format}
7042 Hardware-supported instruction recording, supported on Intel
7043 processors. This method does not record data. Further, the data is
7044 collected in a ring buffer so old data will be overwritten when the
7045 buffer is full. It allows limited reverse execution. Variables and
7046 registers are not available during reverse execution. In remote
7047 debugging, recording continues on disconnect. Recorded data can be
7048 inspected after reconnecting. The recording may be stopped using
7051 The recording format can be specified as parameter. Without a parameter
7052 the command chooses the recording format. The following recording
7053 formats are available:
7057 @cindex branch trace store
7058 Use the @dfn{Branch Trace Store} (@acronym{BTS}) recording format. In
7059 this format, the processor stores a from/to record for each executed
7060 branch in the btrace ring buffer.
7063 @cindex Intel Processor Trace
7064 Use the @dfn{Intel Processor Trace} recording format. In this
7065 format, the processor stores the execution trace in a compressed form
7066 that is afterwards decoded by @value{GDBN}.
7068 The trace can be recorded with very low overhead. The compressed
7069 trace format also allows small trace buffers to already contain a big
7070 number of instructions compared to @acronym{BTS}.
7072 Decoding the recorded execution trace, on the other hand, is more
7073 expensive than decoding @acronym{BTS} trace. This is mostly due to the
7074 increased number of instructions to process. You should increase the
7075 buffer-size with care.
7078 Not all recording formats may be available on all processors.
7081 The process record and replay target can only debug a process that is
7082 already running. Therefore, you need first to start the process with
7083 the @kbd{run} or @kbd{start} commands, and then start the recording
7084 with the @kbd{record @var{method}} command.
7086 @cindex displaced stepping, and process record and replay
7087 Displaced stepping (@pxref{Maintenance Commands,, displaced stepping})
7088 will be automatically disabled when process record and replay target
7089 is started. That's because the process record and replay target
7090 doesn't support displaced stepping.
7092 @cindex non-stop mode, and process record and replay
7093 @cindex asynchronous execution, and process record and replay
7094 If the inferior is in the non-stop mode (@pxref{Non-Stop Mode}) or in
7095 the asynchronous execution mode (@pxref{Background Execution}), not
7096 all recording methods are available. The @code{full} recording method
7097 does not support these two modes.
7102 Stop the process record and replay target. When process record and
7103 replay target stops, the entire execution log will be deleted and the
7104 inferior will either be terminated, or will remain in its final state.
7106 When you stop the process record and replay target in record mode (at
7107 the end of the execution log), the inferior will be stopped at the
7108 next instruction that would have been recorded. In other words, if
7109 you record for a while and then stop recording, the inferior process
7110 will be left in the same state as if the recording never happened.
7112 On the other hand, if the process record and replay target is stopped
7113 while in replay mode (that is, not at the end of the execution log,
7114 but at some earlier point), the inferior process will become ``live''
7115 at that earlier state, and it will then be possible to continue the
7116 usual ``live'' debugging of the process from that state.
7118 When the inferior process exits, or @value{GDBN} detaches from it,
7119 process record and replay target will automatically stop itself.
7123 Go to a specific location in the execution log. There are several
7124 ways to specify the location to go to:
7127 @item record goto begin
7128 @itemx record goto start
7129 Go to the beginning of the execution log.
7131 @item record goto end
7132 Go to the end of the execution log.
7134 @item record goto @var{n}
7135 Go to instruction number @var{n} in the execution log.
7139 @item record save @var{filename}
7140 Save the execution log to a file @file{@var{filename}}.
7141 Default filename is @file{gdb_record.@var{process_id}}, where
7142 @var{process_id} is the process ID of the inferior.
7144 This command may not be available for all recording methods.
7146 @kindex record restore
7147 @item record restore @var{filename}
7148 Restore the execution log from a file @file{@var{filename}}.
7149 File must have been created with @code{record save}.
7151 @kindex set record full
7152 @item set record full insn-number-max @var{limit}
7153 @itemx set record full insn-number-max unlimited
7154 Set the limit of instructions to be recorded for the @code{full}
7155 recording method. Default value is 200000.
7157 If @var{limit} is a positive number, then @value{GDBN} will start
7158 deleting instructions from the log once the number of the record
7159 instructions becomes greater than @var{limit}. For every new recorded
7160 instruction, @value{GDBN} will delete the earliest recorded
7161 instruction to keep the number of recorded instructions at the limit.
7162 (Since deleting recorded instructions loses information, @value{GDBN}
7163 lets you control what happens when the limit is reached, by means of
7164 the @code{stop-at-limit} option, described below.)
7166 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will never
7167 delete recorded instructions from the execution log. The number of
7168 recorded instructions is limited only by the available memory.
7170 @kindex show record full
7171 @item show record full insn-number-max
7172 Show the limit of instructions to be recorded with the @code{full}
7175 @item set record full stop-at-limit
7176 Control the behavior of the @code{full} recording method when the
7177 number of recorded instructions reaches the limit. If ON (the
7178 default), @value{GDBN} will stop when the limit is reached for the
7179 first time and ask you whether you want to stop the inferior or
7180 continue running it and recording the execution log. If you decide
7181 to continue recording, each new recorded instruction will cause the
7182 oldest one to be deleted.
7184 If this option is OFF, @value{GDBN} will automatically delete the
7185 oldest record to make room for each new one, without asking.
7187 @item show record full stop-at-limit
7188 Show the current setting of @code{stop-at-limit}.
7190 @item set record full memory-query
7191 Control the behavior when @value{GDBN} is unable to record memory
7192 changes caused by an instruction for the @code{full} recording method.
7193 If ON, @value{GDBN} will query whether to stop the inferior in that
7196 If this option is OFF (the default), @value{GDBN} will automatically
7197 ignore the effect of such instructions on memory. Later, when
7198 @value{GDBN} replays this execution log, it will mark the log of this
7199 instruction as not accessible, and it will not affect the replay
7202 @item show record full memory-query
7203 Show the current setting of @code{memory-query}.
7205 @kindex set record btrace
7206 The @code{btrace} record target does not trace data. As a
7207 convenience, when replaying, @value{GDBN} reads read-only memory off
7208 the live program directly, assuming that the addresses of the
7209 read-only areas don't change. This for example makes it possible to
7210 disassemble code while replaying, but not to print variables.
7211 In some cases, being able to inspect variables might be useful.
7212 You can use the following command for that:
7214 @item set record btrace replay-memory-access
7215 Control the behavior of the @code{btrace} recording method when
7216 accessing memory during replay. If @code{read-only} (the default),
7217 @value{GDBN} will only allow accesses to read-only memory.
7218 If @code{read-write}, @value{GDBN} will allow accesses to read-only
7219 and to read-write memory. Beware that the accessed memory corresponds
7220 to the live target and not necessarily to the current replay
7223 @item set record btrace cpu @var{identifier}
7224 Set the processor to be used for enabling workarounds for processor
7225 errata when decoding the trace.
7227 Processor errata are defects in processor operation, caused by its
7228 design or manufacture. They can cause a trace not to match the
7229 specification. This, in turn, may cause trace decode to fail.
7230 @value{GDBN} can detect erroneous trace packets and correct them, thus
7231 avoiding the decoding failures. These corrections are known as
7232 @dfn{errata workarounds}, and are enabled based on the processor on
7233 which the trace was recorded.
7235 By default, @value{GDBN} attempts to detect the processor
7236 automatically, and apply the necessary workarounds for it. However,
7237 you may need to specify the processor if @value{GDBN} does not yet
7238 support it. This command allows you to do that, and also allows to
7239 disable the workarounds.
7241 The argument @var{identifier} identifies the @sc{cpu} and is of the
7242 form: @code{@var{vendor}:@var{procesor identifier}}. In addition,
7243 there are two special identifiers, @code{none} and @code{auto}
7246 The following vendor identifiers and corresponding processor
7247 identifiers are currently supported:
7249 @multitable @columnfractions .1 .9
7252 @tab @var{family}/@var{model}[/@var{stepping}]
7256 On GNU/Linux systems, the processor @var{family}, @var{model}, and
7257 @var{stepping} can be obtained from @code{/proc/cpuinfo}.
7259 If @var{identifier} is @code{auto}, enable errata workarounds for the
7260 processor on which the trace was recorded. If @var{identifier} is
7261 @code{none}, errata workarounds are disabled.
7263 For example, when using an old @value{GDBN} on a new system, decode
7264 may fail because @value{GDBN} does not support the new processor. It
7265 often suffices to specify an older processor that @value{GDBN}
7270 Active record target: record-btrace
7271 Recording format: Intel Processor Trace.
7273 Failed to configure the Intel Processor Trace decoder: unknown cpu.
7274 (gdb) set record btrace cpu intel:6/158
7276 Active record target: record-btrace
7277 Recording format: Intel Processor Trace.
7279 Recorded 84872 instructions in 3189 functions (0 gaps) for thread 1 (...).
7282 @kindex show record btrace
7283 @item show record btrace replay-memory-access
7284 Show the current setting of @code{replay-memory-access}.
7286 @item show record btrace cpu
7287 Show the processor to be used for enabling trace decode errata
7290 @kindex set record btrace bts
7291 @item set record btrace bts buffer-size @var{size}
7292 @itemx set record btrace bts buffer-size unlimited
7293 Set the requested ring buffer size for branch tracing in @acronym{BTS}
7294 format. Default is 64KB.
7296 If @var{size} is a positive number, then @value{GDBN} will try to
7297 allocate a buffer of at least @var{size} bytes for each new thread
7298 that uses the btrace recording method and the @acronym{BTS} format.
7299 The actually obtained buffer size may differ from the requested
7300 @var{size}. Use the @code{info record} command to see the actual
7301 buffer size for each thread that uses the btrace recording method and
7302 the @acronym{BTS} format.
7304 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7305 allocate a buffer of 4MB.
7307 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7308 also need longer to process the branch trace data before it can be used.
7310 @item show record btrace bts buffer-size @var{size}
7311 Show the current setting of the requested ring buffer size for branch
7312 tracing in @acronym{BTS} format.
7314 @kindex set record btrace pt
7315 @item set record btrace pt buffer-size @var{size}
7316 @itemx set record btrace pt buffer-size unlimited
7317 Set the requested ring buffer size for branch tracing in Intel
7318 Processor Trace format. Default is 16KB.
7320 If @var{size} is a positive number, then @value{GDBN} will try to
7321 allocate a buffer of at least @var{size} bytes for each new thread
7322 that uses the btrace recording method and the Intel Processor Trace
7323 format. The actually obtained buffer size may differ from the
7324 requested @var{size}. Use the @code{info record} command to see the
7325 actual buffer size for each thread.
7327 If @var{limit} is @code{unlimited} or zero, @value{GDBN} will try to
7328 allocate a buffer of 4MB.
7330 Bigger buffers mean longer traces. On the other hand, @value{GDBN} will
7331 also need longer to process the branch trace data before it can be used.
7333 @item show record btrace pt buffer-size @var{size}
7334 Show the current setting of the requested ring buffer size for branch
7335 tracing in Intel Processor Trace format.
7339 Show various statistics about the recording depending on the recording
7344 For the @code{full} recording method, it shows the state of process
7345 record and its in-memory execution log buffer, including:
7349 Whether in record mode or replay mode.
7351 Lowest recorded instruction number (counting from when the current execution log started recording instructions).
7353 Highest recorded instruction number.
7355 Current instruction about to be replayed (if in replay mode).
7357 Number of instructions contained in the execution log.
7359 Maximum number of instructions that may be contained in the execution log.
7363 For the @code{btrace} recording method, it shows:
7369 Number of instructions that have been recorded.
7371 Number of blocks of sequential control-flow formed by the recorded
7374 Whether in record mode or replay mode.
7377 For the @code{bts} recording format, it also shows:
7380 Size of the perf ring buffer.
7383 For the @code{pt} recording format, it also shows:
7386 Size of the perf ring buffer.
7390 @kindex record delete
7393 When record target runs in replay mode (``in the past''), delete the
7394 subsequent execution log and begin to record a new execution log starting
7395 from the current address. This means you will abandon the previously
7396 recorded ``future'' and begin recording a new ``future''.
7398 @kindex record instruction-history
7399 @kindex rec instruction-history
7400 @item record instruction-history
7401 Disassembles instructions from the recorded execution log. By
7402 default, ten instructions are disassembled. This can be changed using
7403 the @code{set record instruction-history-size} command. Instructions
7404 are printed in execution order.
7406 It can also print mixed source+disassembly if you specify the the
7407 @code{/m} or @code{/s} modifier, and print the raw instructions in hex
7408 as well as in symbolic form by specifying the @code{/r} modifier.
7410 The current position marker is printed for the instruction at the
7411 current program counter value. This instruction can appear multiple
7412 times in the trace and the current position marker will be printed
7413 every time. To omit the current position marker, specify the
7416 To better align the printed instructions when the trace contains
7417 instructions from more than one function, the function name may be
7418 omitted by specifying the @code{/f} modifier.
7420 Speculatively executed instructions are prefixed with @samp{?}. This
7421 feature is not available for all recording formats.
7423 There are several ways to specify what part of the execution log to
7427 @item record instruction-history @var{insn}
7428 Disassembles ten instructions starting from instruction number
7431 @item record instruction-history @var{insn}, +/-@var{n}
7432 Disassembles @var{n} instructions around instruction number
7433 @var{insn}. If @var{n} is preceded with @code{+}, disassembles
7434 @var{n} instructions after instruction number @var{insn}. If
7435 @var{n} is preceded with @code{-}, disassembles @var{n}
7436 instructions before instruction number @var{insn}.
7438 @item record instruction-history
7439 Disassembles ten more instructions after the last disassembly.
7441 @item record instruction-history -
7442 Disassembles ten more instructions before the last disassembly.
7444 @item record instruction-history @var{begin}, @var{end}
7445 Disassembles instructions beginning with instruction number
7446 @var{begin} until instruction number @var{end}. The instruction
7447 number @var{end} is included.
7450 This command may not be available for all recording methods.
7453 @item set record instruction-history-size @var{size}
7454 @itemx set record instruction-history-size unlimited
7455 Define how many instructions to disassemble in the @code{record
7456 instruction-history} command. The default value is 10.
7457 A @var{size} of @code{unlimited} means unlimited instructions.
7460 @item show record instruction-history-size
7461 Show how many instructions to disassemble in the @code{record
7462 instruction-history} command.
7464 @kindex record function-call-history
7465 @kindex rec function-call-history
7466 @item record function-call-history
7467 Prints the execution history at function granularity. It prints one
7468 line for each sequence of instructions that belong to the same
7469 function giving the name of that function, the source lines
7470 for this instruction sequence (if the @code{/l} modifier is
7471 specified), and the instructions numbers that form the sequence (if
7472 the @code{/i} modifier is specified). The function names are indented
7473 to reflect the call stack depth if the @code{/c} modifier is
7474 specified. The @code{/l}, @code{/i}, and @code{/c} modifiers can be
7478 (@value{GDBP}) @b{list 1, 10}
7489 (@value{GDBP}) @b{record function-call-history /ilc}
7490 1 bar inst 1,4 at foo.c:6,8
7491 2 foo inst 5,10 at foo.c:2,3
7492 3 bar inst 11,13 at foo.c:9,10
7495 By default, ten lines are printed. This can be changed using the
7496 @code{set record function-call-history-size} command. Functions are
7497 printed in execution order. There are several ways to specify what
7501 @item record function-call-history @var{func}
7502 Prints ten functions starting from function number @var{func}.
7504 @item record function-call-history @var{func}, +/-@var{n}
7505 Prints @var{n} functions around function number @var{func}. If
7506 @var{n} is preceded with @code{+}, prints @var{n} functions after
7507 function number @var{func}. If @var{n} is preceded with @code{-},
7508 prints @var{n} functions before function number @var{func}.
7510 @item record function-call-history
7511 Prints ten more functions after the last ten-line print.
7513 @item record function-call-history -
7514 Prints ten more functions before the last ten-line print.
7516 @item record function-call-history @var{begin}, @var{end}
7517 Prints functions beginning with function number @var{begin} until
7518 function number @var{end}. The function number @var{end} is included.
7521 This command may not be available for all recording methods.
7523 @item set record function-call-history-size @var{size}
7524 @itemx set record function-call-history-size unlimited
7525 Define how many lines to print in the
7526 @code{record function-call-history} command. The default value is 10.
7527 A size of @code{unlimited} means unlimited lines.
7529 @item show record function-call-history-size
7530 Show how many lines to print in the
7531 @code{record function-call-history} command.
7536 @chapter Examining the Stack
7538 When your program has stopped, the first thing you need to know is where it
7539 stopped and how it got there.
7542 Each time your program performs a function call, information about the call
7544 That information includes the location of the call in your program,
7545 the arguments of the call,
7546 and the local variables of the function being called.
7547 The information is saved in a block of data called a @dfn{stack frame}.
7548 The stack frames are allocated in a region of memory called the @dfn{call
7551 When your program stops, the @value{GDBN} commands for examining the
7552 stack allow you to see all of this information.
7554 @cindex selected frame
7555 One of the stack frames is @dfn{selected} by @value{GDBN} and many
7556 @value{GDBN} commands refer implicitly to the selected frame. In
7557 particular, whenever you ask @value{GDBN} for the value of a variable in
7558 your program, the value is found in the selected frame. There are
7559 special @value{GDBN} commands to select whichever frame you are
7560 interested in. @xref{Selection, ,Selecting a Frame}.
7562 When your program stops, @value{GDBN} automatically selects the
7563 currently executing frame and describes it briefly, similar to the
7564 @code{frame} command (@pxref{Frame Info, ,Information about a Frame}).
7567 * Frames:: Stack frames
7568 * Backtrace:: Backtraces
7569 * Selection:: Selecting a frame
7570 * Frame Info:: Information on a frame
7571 * Frame Apply:: Applying a command to several frames
7572 * Frame Filter Management:: Managing frame filters
7577 @section Stack Frames
7579 @cindex frame, definition
7581 The call stack is divided up into contiguous pieces called @dfn{stack
7582 frames}, or @dfn{frames} for short; each frame is the data associated
7583 with one call to one function. The frame contains the arguments given
7584 to the function, the function's local variables, and the address at
7585 which the function is executing.
7587 @cindex initial frame
7588 @cindex outermost frame
7589 @cindex innermost frame
7590 When your program is started, the stack has only one frame, that of the
7591 function @code{main}. This is called the @dfn{initial} frame or the
7592 @dfn{outermost} frame. Each time a function is called, a new frame is
7593 made. Each time a function returns, the frame for that function invocation
7594 is eliminated. If a function is recursive, there can be many frames for
7595 the same function. The frame for the function in which execution is
7596 actually occurring is called the @dfn{innermost} frame. This is the most
7597 recently created of all the stack frames that still exist.
7599 @cindex frame pointer
7600 Inside your program, stack frames are identified by their addresses. A
7601 stack frame consists of many bytes, each of which has its own address; each
7602 kind of computer has a convention for choosing one byte whose
7603 address serves as the address of the frame. Usually this address is kept
7604 in a register called the @dfn{frame pointer register}
7605 (@pxref{Registers, $fp}) while execution is going on in that frame.
7608 @cindex frame number
7609 @value{GDBN} labels each existing stack frame with a @dfn{level}, a
7610 number that is zero for the innermost frame, one for the frame that
7611 called it, and so on upward. These level numbers give you a way of
7612 designating stack frames in @value{GDBN} commands. The terms
7613 @dfn{frame number} and @dfn{frame level} can be used interchangeably to
7614 describe this number.
7616 @c The -fomit-frame-pointer below perennially causes hbox overflow
7617 @c underflow problems.
7618 @cindex frameless execution
7619 Some compilers provide a way to compile functions so that they operate
7620 without stack frames. (For example, the @value{NGCC} option
7622 @samp{-fomit-frame-pointer}
7624 generates functions without a frame.)
7625 This is occasionally done with heavily used library functions to save
7626 the frame setup time. @value{GDBN} has limited facilities for dealing
7627 with these function invocations. If the innermost function invocation
7628 has no stack frame, @value{GDBN} nevertheless regards it as though
7629 it had a separate frame, which is numbered zero as usual, allowing
7630 correct tracing of the function call chain. However, @value{GDBN} has
7631 no provision for frameless functions elsewhere in the stack.
7637 @cindex call stack traces
7638 A backtrace is a summary of how your program got where it is. It shows one
7639 line per frame, for many frames, starting with the currently executing
7640 frame (frame zero), followed by its caller (frame one), and on up the
7643 @anchor{backtrace-command}
7645 @kindex bt @r{(@code{backtrace})}
7646 To print a backtrace of the entire stack, use the @code{backtrace}
7647 command, or its alias @code{bt}. This command will print one line per
7648 frame for frames in the stack. By default, all stack frames are
7649 printed. You can stop the backtrace at any time by typing the system
7650 interrupt character, normally @kbd{Ctrl-c}.
7653 @item backtrace [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
7654 @itemx bt [@var{option}]@dots{} [@var{qualifier}]@dots{} [@var{count}]
7655 Print the backtrace of the entire stack.
7657 The optional @var{count} can be one of the following:
7662 Print only the innermost @var{n} frames, where @var{n} is a positive
7667 Print only the outermost @var{n} frames, where @var{n} is a positive
7675 Print the values of the local variables also. This can be combined
7676 with the optional @var{count} to limit the number of frames shown.
7679 Do not run Python frame filters on this backtrace. @xref{Frame
7680 Filter API}, for more information. Additionally use @ref{disable
7681 frame-filter all} to turn off all frame filters. This is only
7682 relevant when @value{GDBN} has been configured with @code{Python}
7686 A Python frame filter might decide to ``elide'' some frames. Normally
7687 such elided frames are still printed, but they are indented relative
7688 to the filtered frames that cause them to be elided. The @code{-hide}
7689 option causes elided frames to not be printed at all.
7692 The @code{backtrace} command also supports a number of options that
7693 allow overriding relevant global print settings as set by @code{set
7694 backtrace} and @code{set print} subcommands:
7697 @item -past-main [@code{on}|@code{off}]
7698 Set whether backtraces should continue past @code{main}. Related setting:
7699 @ref{set backtrace past-main}.
7701 @item -past-entry [@code{on}|@code{off}]
7702 Set whether backtraces should continue past the entry point of a program.
7703 Related setting: @ref{set backtrace past-entry}.
7705 @item -entry-values @code{no}|@code{only}|@code{preferred}|@code{if-needed}|@code{both}|@code{compact}|@code{default}
7706 Set printing of function arguments at function entry.
7707 Related setting: @ref{set print entry-values}.
7709 @item -frame-arguments @code{all}|@code{scalars}|@code{none}
7710 Set printing of non-scalar frame arguments.
7711 Related setting: @ref{set print frame-arguments}.
7713 @item -raw-frame-arguments [@code{on}|@code{off}]
7714 Set whether to print frame arguments in raw form.
7715 Related setting: @ref{set print raw-frame-arguments}.
7718 The optional @var{qualifier} is maintained for backward compatibility.
7719 It can be one of the following:
7723 Equivalent to the @code{-full} option.
7726 Equivalent to the @code{-no-filters} option.
7729 Equivalent to the @code{-hide} option.
7736 The names @code{where} and @code{info stack} (abbreviated @code{info s})
7737 are additional aliases for @code{backtrace}.
7739 @cindex multiple threads, backtrace
7740 In a multi-threaded program, @value{GDBN} by default shows the
7741 backtrace only for the current thread. To display the backtrace for
7742 several or all of the threads, use the command @code{thread apply}
7743 (@pxref{Threads, thread apply}). For example, if you type @kbd{thread
7744 apply all backtrace}, @value{GDBN} will display the backtrace for all
7745 the threads; this is handy when you debug a core dump of a
7746 multi-threaded program.
7748 Each line in the backtrace shows the frame number and the function name.
7749 The program counter value is also shown---unless you use @code{set
7750 print address off}. The backtrace also shows the source file name and
7751 line number, as well as the arguments to the function. The program
7752 counter value is omitted if it is at the beginning of the code for that
7755 Here is an example of a backtrace. It was made with the command
7756 @samp{bt 3}, so it shows the innermost three frames.
7760 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7762 #1 0x6e38 in expand_macro (sym=0x2b600, data=...) at macro.c:242
7763 #2 0x6840 in expand_token (obs=0x0, t=177664, td=0xf7fffb08)
7765 (More stack frames follow...)
7770 The display for frame zero does not begin with a program counter
7771 value, indicating that your program has stopped at the beginning of the
7772 code for line @code{993} of @code{builtin.c}.
7775 The value of parameter @code{data} in frame 1 has been replaced by
7776 @code{@dots{}}. By default, @value{GDBN} prints the value of a parameter
7777 only if it is a scalar (integer, pointer, enumeration, etc). See command
7778 @kbd{set print frame-arguments} in @ref{Print Settings} for more details
7779 on how to configure the way function parameter values are printed.
7781 @cindex optimized out, in backtrace
7782 @cindex function call arguments, optimized out
7783 If your program was compiled with optimizations, some compilers will
7784 optimize away arguments passed to functions if those arguments are
7785 never used after the call. Such optimizations generate code that
7786 passes arguments through registers, but doesn't store those arguments
7787 in the stack frame. @value{GDBN} has no way of displaying such
7788 arguments in stack frames other than the innermost one. Here's what
7789 such a backtrace might look like:
7793 #0 m4_traceon (obs=0x24eb0, argc=1, argv=0x2b8c8)
7795 #1 0x6e38 in expand_macro (sym=<optimized out>) at macro.c:242
7796 #2 0x6840 in expand_token (obs=0x0, t=<optimized out>, td=0xf7fffb08)
7798 (More stack frames follow...)
7803 The values of arguments that were not saved in their stack frames are
7804 shown as @samp{<optimized out>}.
7806 If you need to display the values of such optimized-out arguments,
7807 either deduce that from other variables whose values depend on the one
7808 you are interested in, or recompile without optimizations.
7810 @cindex backtrace beyond @code{main} function
7811 @cindex program entry point
7812 @cindex startup code, and backtrace
7813 Most programs have a standard user entry point---a place where system
7814 libraries and startup code transition into user code. For C this is
7815 @code{main}@footnote{
7816 Note that embedded programs (the so-called ``free-standing''
7817 environment) are not required to have a @code{main} function as the
7818 entry point. They could even have multiple entry points.}.
7819 When @value{GDBN} finds the entry function in a backtrace
7820 it will terminate the backtrace, to avoid tracing into highly
7821 system-specific (and generally uninteresting) code.
7823 If you need to examine the startup code, or limit the number of levels
7824 in a backtrace, you can change this behavior:
7827 @item set backtrace past-main
7828 @itemx set backtrace past-main on
7829 @anchor{set backtrace past-main}
7830 @kindex set backtrace
7831 Backtraces will continue past the user entry point.
7833 @item set backtrace past-main off
7834 Backtraces will stop when they encounter the user entry point. This is the
7837 @item show backtrace past-main
7838 @kindex show backtrace
7839 Display the current user entry point backtrace policy.
7841 @item set backtrace past-entry
7842 @itemx set backtrace past-entry on
7843 @anchor{set backtrace past-entry}
7844 Backtraces will continue past the internal entry point of an application.
7845 This entry point is encoded by the linker when the application is built,
7846 and is likely before the user entry point @code{main} (or equivalent) is called.
7848 @item set backtrace past-entry off
7849 Backtraces will stop when they encounter the internal entry point of an
7850 application. This is the default.
7852 @item show backtrace past-entry
7853 Display the current internal entry point backtrace policy.
7855 @item set backtrace limit @var{n}
7856 @itemx set backtrace limit 0
7857 @itemx set backtrace limit unlimited
7858 @anchor{set backtrace limit}
7859 @cindex backtrace limit
7860 Limit the backtrace to @var{n} levels. A value of @code{unlimited}
7861 or zero means unlimited levels.
7863 @item show backtrace limit
7864 Display the current limit on backtrace levels.
7867 You can control how file names are displayed.
7870 @item set filename-display
7871 @itemx set filename-display relative
7872 @cindex filename-display
7873 Display file names relative to the compilation directory. This is the default.
7875 @item set filename-display basename
7876 Display only basename of a filename.
7878 @item set filename-display absolute
7879 Display an absolute filename.
7881 @item show filename-display
7882 Show the current way to display filenames.
7886 @section Selecting a Frame
7888 Most commands for examining the stack and other data in your program work on
7889 whichever stack frame is selected at the moment. Here are the commands for
7890 selecting a stack frame; all of them finish by printing a brief description
7891 of the stack frame just selected.
7894 @kindex frame@r{, selecting}
7895 @kindex f @r{(@code{frame})}
7896 @item frame @r{[} @var{frame-selection-spec} @r{]}
7897 @item f @r{[} @var{frame-selection-spec} @r{]}
7898 The @command{frame} command allows different stack frames to be
7899 selected. The @var{frame-selection-spec} can be any of the following:
7904 @item level @var{num}
7905 Select frame level @var{num}. Recall that frame zero is the innermost
7906 (currently executing) frame, frame one is the frame that called the
7907 innermost one, and so on. The highest level frame is usually the one
7910 As this is the most common method of navigating the frame stack, the
7911 string @command{level} can be omitted. For example, the following two
7912 commands are equivalent:
7915 (@value{GDBP}) frame 3
7916 (@value{GDBP}) frame level 3
7919 @kindex frame address
7920 @item address @var{stack-address}
7921 Select the frame with stack address @var{stack-address}. The
7922 @var{stack-address} for a frame can be seen in the output of
7923 @command{info frame}, for example:
7927 Stack level 1, frame at 0x7fffffffda30:
7928 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
7929 tail call frame, caller of frame at 0x7fffffffda30
7930 source language c++.
7931 Arglist at unknown address.
7932 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
7935 The @var{stack-address} for this frame is @code{0x7fffffffda30} as
7936 indicated by the line:
7939 Stack level 1, frame at 0x7fffffffda30:
7942 @kindex frame function
7943 @item function @var{function-name}
7944 Select the stack frame for function @var{function-name}. If there are
7945 multiple stack frames for function @var{function-name} then the inner
7946 most stack frame is selected.
7949 @item view @var{stack-address} @r{[} @var{pc-addr} @r{]}
7950 View a frame that is not part of @value{GDBN}'s backtrace. The frame
7951 viewed has stack address @var{stack-addr}, and optionally, a program
7952 counter address of @var{pc-addr}.
7954 This is useful mainly if the chaining of stack frames has been
7955 damaged by a bug, making it impossible for @value{GDBN} to assign
7956 numbers properly to all frames. In addition, this can be useful
7957 when your program has multiple stacks and switches between them.
7959 When viewing a frame outside the current backtrace using
7960 @command{frame view} then you can always return to the original
7961 stack using one of the previous stack frame selection instructions,
7962 for example @command{frame level 0}.
7968 Move @var{n} frames up the stack; @var{n} defaults to 1. For positive
7969 numbers @var{n}, this advances toward the outermost frame, to higher
7970 frame numbers, to frames that have existed longer.
7973 @kindex do @r{(@code{down})}
7975 Move @var{n} frames down the stack; @var{n} defaults to 1. For
7976 positive numbers @var{n}, this advances toward the innermost frame, to
7977 lower frame numbers, to frames that were created more recently.
7978 You may abbreviate @code{down} as @code{do}.
7981 All of these commands end by printing two lines of output describing the
7982 frame. The first line shows the frame number, the function name, the
7983 arguments, and the source file and line number of execution in that
7984 frame. The second line shows the text of that source line.
7992 #1 0x22f0 in main (argc=1, argv=0xf7fffbf4, env=0xf7fffbfc)
7994 10 read_input_file (argv[i]);
7998 After such a printout, the @code{list} command with no arguments
7999 prints ten lines centered on the point of execution in the frame.
8000 You can also edit the program at the point of execution with your favorite
8001 editing program by typing @code{edit}.
8002 @xref{List, ,Printing Source Lines},
8006 @kindex select-frame
8007 @item select-frame @r{[} @var{frame-selection-spec} @r{]}
8008 The @code{select-frame} command is a variant of @code{frame} that does
8009 not display the new frame after selecting it. This command is
8010 intended primarily for use in @value{GDBN} command scripts, where the
8011 output might be unnecessary and distracting. The
8012 @var{frame-selection-spec} is as for the @command{frame} command
8013 described in @ref{Selection, ,Selecting a Frame}.
8015 @kindex down-silently
8017 @item up-silently @var{n}
8018 @itemx down-silently @var{n}
8019 These two commands are variants of @code{up} and @code{down},
8020 respectively; they differ in that they do their work silently, without
8021 causing display of the new frame. They are intended primarily for use
8022 in @value{GDBN} command scripts, where the output might be unnecessary and
8027 @section Information About a Frame
8029 There are several other commands to print information about the selected
8035 When used without any argument, this command does not change which
8036 frame is selected, but prints a brief description of the currently
8037 selected stack frame. It can be abbreviated @code{f}. With an
8038 argument, this command is used to select a stack frame.
8039 @xref{Selection, ,Selecting a Frame}.
8042 @kindex info f @r{(@code{info frame})}
8045 This command prints a verbose description of the selected stack frame,
8050 the address of the frame
8052 the address of the next frame down (called by this frame)
8054 the address of the next frame up (caller of this frame)
8056 the language in which the source code corresponding to this frame is written
8058 the address of the frame's arguments
8060 the address of the frame's local variables
8062 the program counter saved in it (the address of execution in the caller frame)
8064 which registers were saved in the frame
8067 @noindent The verbose description is useful when
8068 something has gone wrong that has made the stack format fail to fit
8069 the usual conventions.
8071 @item info frame @r{[} @var{frame-selection-spec} @r{]}
8072 @itemx info f @r{[} @var{frame-selection-spec} @r{]}
8073 Print a verbose description of the frame selected by
8074 @var{frame-selection-spec}. The @var{frame-selection-spec} is the
8075 same as for the @command{frame} command (@pxref{Selection, ,Selecting
8076 a Frame}). The selected frame remains unchanged by this command.
8079 @item info args [-q]
8080 Print the arguments of the selected frame, each on a separate line.
8082 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
8083 printing header information and messages explaining why no argument
8086 @item info args [-q] [-t @var{type_regexp}] [@var{regexp}]
8087 Like @kbd{info args}, but only print the arguments selected
8088 with the provided regexp(s).
8090 If @var{regexp} is provided, print only the arguments whose names
8091 match the regular expression @var{regexp}.
8093 If @var{type_regexp} is provided, print only the arguments whose
8094 types, as printed by the @code{whatis} command, match
8095 the regular expression @var{type_regexp}.
8096 If @var{type_regexp} contains space(s), it should be enclosed in
8097 quote characters. If needed, use backslash to escape the meaning
8098 of special characters or quotes.
8100 If both @var{regexp} and @var{type_regexp} are provided, an argument
8101 is printed only if its name matches @var{regexp} and its type matches
8104 @item info locals [-q]
8106 Print the local variables of the selected frame, each on a separate
8107 line. These are all variables (declared either static or automatic)
8108 accessible at the point of execution of the selected frame.
8110 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
8111 printing header information and messages explaining why no local variables
8114 @item info locals [-q] [-t @var{type_regexp}] [@var{regexp}]
8115 Like @kbd{info locals}, but only print the local variables selected
8116 with the provided regexp(s).
8118 If @var{regexp} is provided, print only the local variables whose names
8119 match the regular expression @var{regexp}.
8121 If @var{type_regexp} is provided, print only the local variables whose
8122 types, as printed by the @code{whatis} command, match
8123 the regular expression @var{type_regexp}.
8124 If @var{type_regexp} contains space(s), it should be enclosed in
8125 quote characters. If needed, use backslash to escape the meaning
8126 of special characters or quotes.
8128 If both @var{regexp} and @var{type_regexp} are provided, a local variable
8129 is printed only if its name matches @var{regexp} and its type matches
8132 The command @kbd{info locals -q -t @var{type_regexp}} can usefully be
8133 combined with the commands @kbd{frame apply} and @kbd{thread apply}.
8134 For example, your program might use Resource Acquisition Is
8135 Initialization types (RAII) such as @code{lock_something_t}: each
8136 local variable of type @code{lock_something_t} automatically places a
8137 lock that is destroyed when the variable goes out of scope. You can
8138 then list all acquired locks in your program by doing
8140 thread apply all -s frame apply all -s info locals -q -t lock_something_t
8143 or the equivalent shorter form
8145 tfaas i lo -q -t lock_something_t
8151 @section Applying a Command to Several Frames.
8152 @anchor{frame apply}
8154 @cindex apply command to several frames
8156 @item frame apply [all | @var{count} | @var{-count} | level @var{level}@dots{}] [@var{option}]@dots{} @var{command}
8157 The @code{frame apply} command allows you to apply the named
8158 @var{command} to one or more frames.
8162 Specify @code{all} to apply @var{command} to all frames.
8165 Use @var{count} to apply @var{command} to the innermost @var{count}
8166 frames, where @var{count} is a positive number.
8169 Use @var{-count} to apply @var{command} to the outermost @var{count}
8170 frames, where @var{count} is a positive number.
8173 Use @code{level} to apply @var{command} to the set of frames identified
8174 by the @var{level} list. @var{level} is a frame level or a range of frame
8175 levels as @var{level1}-@var{level2}. The frame level is the number shown
8176 in the first field of the @samp{backtrace} command output.
8177 E.g., @samp{2-4 6-8 3} indicates to apply @var{command} for the frames
8178 at levels 2, 3, 4, 6, 7, 8, and then again on frame at level 3.
8182 Note that the frames on which @code{frame apply} applies a command are
8183 also influenced by the @code{set backtrace} settings such as @code{set
8184 backtrace past-main} and @code{set backtrace limit N}.
8185 @xref{Backtrace,,Backtraces}.
8187 The @code{frame apply} command also supports a number of options that
8188 allow overriding relevant @code{set backtrace} settings:
8191 @item -past-main [@code{on}|@code{off}]
8192 Whether backtraces should continue past @code{main}.
8193 Related setting: @ref{set backtrace past-main}.
8195 @item -past-entry [@code{on}|@code{off}]
8196 Whether backtraces should continue past the entry point of a program.
8197 Related setting: @ref{set backtrace past-entry}.
8200 By default, @value{GDBN} displays some frame information before the
8201 output produced by @var{command}, and an error raised during the
8202 execution of a @var{command} will abort @code{frame apply}. The
8203 following options can be used to fine-tune these behaviors:
8207 The flag @code{-c}, which stands for @samp{continue}, causes any
8208 errors in @var{command} to be displayed, and the execution of
8209 @code{frame apply} then continues.
8211 The flag @code{-s}, which stands for @samp{silent}, causes any errors
8212 or empty output produced by a @var{command} to be silently ignored.
8213 That is, the execution continues, but the frame information and errors
8216 The flag @code{-q} (@samp{quiet}) disables printing the frame
8220 The following example shows how the flags @code{-c} and @code{-s} are
8221 working when applying the command @code{p j} to all frames, where
8222 variable @code{j} can only be successfully printed in the outermost
8223 @code{#1 main} frame.
8227 (gdb) frame apply all p j
8228 #0 some_function (i=5) at fun.c:4
8229 No symbol "j" in current context.
8230 (gdb) frame apply all -c p j
8231 #0 some_function (i=5) at fun.c:4
8232 No symbol "j" in current context.
8233 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8235 (gdb) frame apply all -s p j
8236 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8242 By default, @samp{frame apply}, prints the frame location
8243 information before the command output:
8247 (gdb) frame apply all p $sp
8248 #0 some_function (i=5) at fun.c:4
8249 $4 = (void *) 0xffffd1e0
8250 #1 0x565555fb in main (argc=1, argv=0xffffd2c4) at fun.c:11
8251 $5 = (void *) 0xffffd1f0
8256 If the flag @code{-q} is given, no frame information is printed:
8259 (gdb) frame apply all -q p $sp
8260 $12 = (void *) 0xffffd1e0
8261 $13 = (void *) 0xffffd1f0
8271 @cindex apply a command to all frames (ignoring errors and empty output)
8272 @item faas @var{command}
8273 Shortcut for @code{frame apply all -s @var{command}}.
8274 Applies @var{command} on all frames, ignoring errors and empty output.
8276 It can for example be used to print a local variable or a function
8277 argument without knowing the frame where this variable or argument
8280 (@value{GDBP}) faas p some_local_var_i_do_not_remember_where_it_is
8283 The @code{faas} command accepts the same options as the @code{frame
8284 apply} command. @xref{frame apply}.
8286 Note that the command @code{tfaas @var{command}} applies @var{command}
8287 on all frames of all threads. See @xref{Threads,,Threads}.
8291 @node Frame Filter Management
8292 @section Management of Frame Filters.
8293 @cindex managing frame filters
8295 Frame filters are Python based utilities to manage and decorate the
8296 output of frames. @xref{Frame Filter API}, for further information.
8298 Managing frame filters is performed by several commands available
8299 within @value{GDBN}, detailed here.
8302 @kindex info frame-filter
8303 @item info frame-filter
8304 Print a list of installed frame filters from all dictionaries, showing
8305 their name, priority and enabled status.
8307 @kindex disable frame-filter
8308 @anchor{disable frame-filter all}
8309 @item disable frame-filter @var{filter-dictionary} @var{filter-name}
8310 Disable a frame filter in the dictionary matching
8311 @var{filter-dictionary} and @var{filter-name}. The
8312 @var{filter-dictionary} may be @code{all}, @code{global},
8313 @code{progspace}, or the name of the object file where the frame filter
8314 dictionary resides. When @code{all} is specified, all frame filters
8315 across all dictionaries are disabled. The @var{filter-name} is the name
8316 of the frame filter and is used when @code{all} is not the option for
8317 @var{filter-dictionary}. A disabled frame-filter is not deleted, it
8318 may be enabled again later.
8320 @kindex enable frame-filter
8321 @item enable frame-filter @var{filter-dictionary} @var{filter-name}
8322 Enable a frame filter in the dictionary matching
8323 @var{filter-dictionary} and @var{filter-name}. The
8324 @var{filter-dictionary} may be @code{all}, @code{global},
8325 @code{progspace} or the name of the object file where the frame filter
8326 dictionary resides. When @code{all} is specified, all frame filters across
8327 all dictionaries are enabled. The @var{filter-name} is the name of the frame
8328 filter and is used when @code{all} is not the option for
8329 @var{filter-dictionary}.
8334 (gdb) info frame-filter
8336 global frame-filters:
8337 Priority Enabled Name
8338 1000 No PrimaryFunctionFilter
8341 progspace /build/test frame-filters:
8342 Priority Enabled Name
8343 100 Yes ProgspaceFilter
8345 objfile /build/test frame-filters:
8346 Priority Enabled Name
8347 999 Yes BuildProgra Filter
8349 (gdb) disable frame-filter /build/test BuildProgramFilter
8350 (gdb) info frame-filter
8352 global frame-filters:
8353 Priority Enabled Name
8354 1000 No PrimaryFunctionFilter
8357 progspace /build/test frame-filters:
8358 Priority Enabled Name
8359 100 Yes ProgspaceFilter
8361 objfile /build/test frame-filters:
8362 Priority Enabled Name
8363 999 No BuildProgramFilter
8365 (gdb) enable frame-filter global PrimaryFunctionFilter
8366 (gdb) info frame-filter
8368 global frame-filters:
8369 Priority Enabled Name
8370 1000 Yes PrimaryFunctionFilter
8373 progspace /build/test frame-filters:
8374 Priority Enabled Name
8375 100 Yes ProgspaceFilter
8377 objfile /build/test frame-filters:
8378 Priority Enabled Name
8379 999 No BuildProgramFilter
8382 @kindex set frame-filter priority
8383 @item set frame-filter priority @var{filter-dictionary} @var{filter-name} @var{priority}
8384 Set the @var{priority} of a frame filter in the dictionary matching
8385 @var{filter-dictionary}, and the frame filter name matching
8386 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8387 @code{progspace} or the name of the object file where the frame filter
8388 dictionary resides. The @var{priority} is an integer.
8390 @kindex show frame-filter priority
8391 @item show frame-filter priority @var{filter-dictionary} @var{filter-name}
8392 Show the @var{priority} of a frame filter in the dictionary matching
8393 @var{filter-dictionary}, and the frame filter name matching
8394 @var{filter-name}. The @var{filter-dictionary} may be @code{global},
8395 @code{progspace} or the name of the object file where the frame filter
8401 (gdb) info frame-filter
8403 global frame-filters:
8404 Priority Enabled Name
8405 1000 Yes PrimaryFunctionFilter
8408 progspace /build/test frame-filters:
8409 Priority Enabled Name
8410 100 Yes ProgspaceFilter
8412 objfile /build/test frame-filters:
8413 Priority Enabled Name
8414 999 No BuildProgramFilter
8416 (gdb) set frame-filter priority global Reverse 50
8417 (gdb) info frame-filter
8419 global frame-filters:
8420 Priority Enabled Name
8421 1000 Yes PrimaryFunctionFilter
8424 progspace /build/test frame-filters:
8425 Priority Enabled Name
8426 100 Yes ProgspaceFilter
8428 objfile /build/test frame-filters:
8429 Priority Enabled Name
8430 999 No BuildProgramFilter
8435 @chapter Examining Source Files
8437 @value{GDBN} can print parts of your program's source, since the debugging
8438 information recorded in the program tells @value{GDBN} what source files were
8439 used to build it. When your program stops, @value{GDBN} spontaneously prints
8440 the line where it stopped. Likewise, when you select a stack frame
8441 (@pxref{Selection, ,Selecting a Frame}), @value{GDBN} prints the line where
8442 execution in that frame has stopped. You can print other portions of
8443 source files by explicit command.
8445 If you use @value{GDBN} through its @sc{gnu} Emacs interface, you may
8446 prefer to use Emacs facilities to view source; see @ref{Emacs, ,Using
8447 @value{GDBN} under @sc{gnu} Emacs}.
8450 * List:: Printing source lines
8451 * Specify Location:: How to specify code locations
8452 * Edit:: Editing source files
8453 * Search:: Searching source files
8454 * Source Path:: Specifying source directories
8455 * Machine Code:: Source and machine code
8459 @section Printing Source Lines
8462 @kindex l @r{(@code{list})}
8463 To print lines from a source file, use the @code{list} command
8464 (abbreviated @code{l}). By default, ten lines are printed.
8465 There are several ways to specify what part of the file you want to
8466 print; see @ref{Specify Location}, for the full list.
8468 Here are the forms of the @code{list} command most commonly used:
8471 @item list @var{linenum}
8472 Print lines centered around line number @var{linenum} in the
8473 current source file.
8475 @item list @var{function}
8476 Print lines centered around the beginning of function
8480 Print more lines. If the last lines printed were printed with a
8481 @code{list} command, this prints lines following the last lines
8482 printed; however, if the last line printed was a solitary line printed
8483 as part of displaying a stack frame (@pxref{Stack, ,Examining the
8484 Stack}), this prints lines centered around that line.
8487 Print lines just before the lines last printed.
8490 @cindex @code{list}, how many lines to display
8491 By default, @value{GDBN} prints ten source lines with any of these forms of
8492 the @code{list} command. You can change this using @code{set listsize}:
8495 @kindex set listsize
8496 @item set listsize @var{count}
8497 @itemx set listsize unlimited
8498 Make the @code{list} command display @var{count} source lines (unless
8499 the @code{list} argument explicitly specifies some other number).
8500 Setting @var{count} to @code{unlimited} or 0 means there's no limit.
8502 @kindex show listsize
8504 Display the number of lines that @code{list} prints.
8507 Repeating a @code{list} command with @key{RET} discards the argument,
8508 so it is equivalent to typing just @code{list}. This is more useful
8509 than listing the same lines again. An exception is made for an
8510 argument of @samp{-}; that argument is preserved in repetition so that
8511 each repetition moves up in the source file.
8513 In general, the @code{list} command expects you to supply zero, one or two
8514 @dfn{locations}. Locations specify source lines; there are several ways
8515 of writing them (@pxref{Specify Location}), but the effect is always
8516 to specify some source line.
8518 Here is a complete description of the possible arguments for @code{list}:
8521 @item list @var{location}
8522 Print lines centered around the line specified by @var{location}.
8524 @item list @var{first},@var{last}
8525 Print lines from @var{first} to @var{last}. Both arguments are
8526 locations. When a @code{list} command has two locations, and the
8527 source file of the second location is omitted, this refers to
8528 the same source file as the first location.
8530 @item list ,@var{last}
8531 Print lines ending with @var{last}.
8533 @item list @var{first},
8534 Print lines starting with @var{first}.
8537 Print lines just after the lines last printed.
8540 Print lines just before the lines last printed.
8543 As described in the preceding table.
8546 @node Specify Location
8547 @section Specifying a Location
8548 @cindex specifying location
8550 @cindex source location
8553 * Linespec Locations:: Linespec locations
8554 * Explicit Locations:: Explicit locations
8555 * Address Locations:: Address locations
8558 Several @value{GDBN} commands accept arguments that specify a location
8559 of your program's code. Since @value{GDBN} is a source-level
8560 debugger, a location usually specifies some line in the source code.
8561 Locations may be specified using three different formats:
8562 linespec locations, explicit locations, or address locations.
8564 @node Linespec Locations
8565 @subsection Linespec Locations
8566 @cindex linespec locations
8568 A @dfn{linespec} is a colon-separated list of source location parameters such
8569 as file name, function name, etc. Here are all the different ways of
8570 specifying a linespec:
8574 Specifies the line number @var{linenum} of the current source file.
8577 @itemx +@var{offset}
8578 Specifies the line @var{offset} lines before or after the @dfn{current
8579 line}. For the @code{list} command, the current line is the last one
8580 printed; for the breakpoint commands, this is the line at which
8581 execution stopped in the currently selected @dfn{stack frame}
8582 (@pxref{Frames, ,Frames}, for a description of stack frames.) When
8583 used as the second of the two linespecs in a @code{list} command,
8584 this specifies the line @var{offset} lines up or down from the first
8587 @item @var{filename}:@var{linenum}
8588 Specifies the line @var{linenum} in the source file @var{filename}.
8589 If @var{filename} is a relative file name, then it will match any
8590 source file name with the same trailing components. For example, if
8591 @var{filename} is @samp{gcc/expr.c}, then it will match source file
8592 name of @file{/build/trunk/gcc/expr.c}, but not
8593 @file{/build/trunk/libcpp/expr.c} or @file{/build/trunk/gcc/x-expr.c}.
8595 @item @var{function}
8596 Specifies the line that begins the body of the function @var{function}.
8597 For example, in C, this is the line with the open brace.
8599 By default, in C@t{++} and Ada, @var{function} is interpreted as
8600 specifying all functions named @var{function} in all scopes. For
8601 C@t{++}, this means in all namespaces and classes. For Ada, this
8602 means in all packages.
8604 For example, assuming a program with C@t{++} symbols named
8605 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8606 func}} and @w{@kbd{break B::func}} set a breakpoint on both symbols.
8608 Commands that accept a linespec let you override this with the
8609 @code{-qualified} option. For example, @w{@kbd{break -qualified
8610 func}} sets a breakpoint on a free-function named @code{func} ignoring
8611 any C@t{++} class methods and namespace functions called @code{func}.
8613 @xref{Explicit Locations}.
8615 @item @var{function}:@var{label}
8616 Specifies the line where @var{label} appears in @var{function}.
8618 @item @var{filename}:@var{function}
8619 Specifies the line that begins the body of the function @var{function}
8620 in the file @var{filename}. You only need the file name with a
8621 function name to avoid ambiguity when there are identically named
8622 functions in different source files.
8625 Specifies the line at which the label named @var{label} appears
8626 in the function corresponding to the currently selected stack frame.
8627 If there is no current selected stack frame (for instance, if the inferior
8628 is not running), then @value{GDBN} will not search for a label.
8630 @cindex breakpoint at static probe point
8631 @item -pstap|-probe-stap @r{[}@var{objfile}:@r{[}@var{provider}:@r{]}@r{]}@var{name}
8632 The @sc{gnu}/Linux tool @code{SystemTap} provides a way for
8633 applications to embed static probes. @xref{Static Probe Points}, for more
8634 information on finding and using static probes. This form of linespec
8635 specifies the location of such a static probe.
8637 If @var{objfile} is given, only probes coming from that shared library
8638 or executable matching @var{objfile} as a regular expression are considered.
8639 If @var{provider} is given, then only probes from that provider are considered.
8640 If several probes match the spec, @value{GDBN} will insert a breakpoint at
8641 each one of those probes.
8644 @node Explicit Locations
8645 @subsection Explicit Locations
8646 @cindex explicit locations
8648 @dfn{Explicit locations} allow the user to directly specify the source
8649 location's parameters using option-value pairs.
8651 Explicit locations are useful when several functions, labels, or
8652 file names have the same name (base name for files) in the program's
8653 sources. In these cases, explicit locations point to the source
8654 line you meant more accurately and unambiguously. Also, using
8655 explicit locations might be faster in large programs.
8657 For example, the linespec @samp{foo:bar} may refer to a function @code{bar}
8658 defined in the file named @file{foo} or the label @code{bar} in a function
8659 named @code{foo}. @value{GDBN} must search either the file system or
8660 the symbol table to know.
8662 The list of valid explicit location options is summarized in the
8666 @item -source @var{filename}
8667 The value specifies the source file name. To differentiate between
8668 files with the same base name, prepend as many directories as is necessary
8669 to uniquely identify the desired file, e.g., @file{foo/bar/baz.c}. Otherwise
8670 @value{GDBN} will use the first file it finds with the given base
8671 name. This option requires the use of either @code{-function} or @code{-line}.
8673 @item -function @var{function}
8674 The value specifies the name of a function. Operations
8675 on function locations unmodified by other options (such as @code{-label}
8676 or @code{-line}) refer to the line that begins the body of the function.
8677 In C, for example, this is the line with the open brace.
8679 By default, in C@t{++} and Ada, @var{function} is interpreted as
8680 specifying all functions named @var{function} in all scopes. For
8681 C@t{++}, this means in all namespaces and classes. For Ada, this
8682 means in all packages.
8684 For example, assuming a program with C@t{++} symbols named
8685 @code{A::B::func} and @code{B::func}, both commands @w{@kbd{break
8686 -function func}} and @w{@kbd{break -function B::func}} set a
8687 breakpoint on both symbols.
8689 You can use the @kbd{-qualified} flag to override this (see below).
8693 This flag makes @value{GDBN} interpret a function name specified with
8694 @kbd{-function} as a complete fully-qualified name.
8696 For example, assuming a C@t{++} program with symbols named
8697 @code{A::B::func} and @code{B::func}, the @w{@kbd{break -qualified
8698 -function B::func}} command sets a breakpoint on @code{B::func}, only.
8700 (Note: the @kbd{-qualified} option can precede a linespec as well
8701 (@pxref{Linespec Locations}), so the particular example above could be
8702 simplified as @w{@kbd{break -qualified B::func}}.)
8704 @item -label @var{label}
8705 The value specifies the name of a label. When the function
8706 name is not specified, the label is searched in the function of the currently
8707 selected stack frame.
8709 @item -line @var{number}
8710 The value specifies a line offset for the location. The offset may either
8711 be absolute (@code{-line 3}) or relative (@code{-line +3}), depending on
8712 the command. When specified without any other options, the line offset is
8713 relative to the current line.
8716 Explicit location options may be abbreviated by omitting any non-unique
8717 trailing characters from the option name, e.g., @w{@kbd{break -s main.c -li 3}}.
8719 @node Address Locations
8720 @subsection Address Locations
8721 @cindex address locations
8723 @dfn{Address locations} indicate a specific program address. They have
8724 the generalized form *@var{address}.
8726 For line-oriented commands, such as @code{list} and @code{edit}, this
8727 specifies a source line that contains @var{address}. For @code{break} and
8728 other breakpoint-oriented commands, this can be used to set breakpoints in
8729 parts of your program which do not have debugging information or
8732 Here @var{address} may be any expression valid in the current working
8733 language (@pxref{Languages, working language}) that specifies a code
8734 address. In addition, as a convenience, @value{GDBN} extends the
8735 semantics of expressions used in locations to cover several situations
8736 that frequently occur during debugging. Here are the various forms
8740 @item @var{expression}
8741 Any expression valid in the current working language.
8743 @item @var{funcaddr}
8744 An address of a function or procedure derived from its name. In C,
8745 C@t{++}, Objective-C, Fortran, minimal, and assembly, this is
8746 simply the function's name @var{function} (and actually a special case
8747 of a valid expression). In Pascal and Modula-2, this is
8748 @code{&@var{function}}. In Ada, this is @code{@var{function}'Address}
8749 (although the Pascal form also works).
8751 This form specifies the address of the function's first instruction,
8752 before the stack frame and arguments have been set up.
8754 @item '@var{filename}':@var{funcaddr}
8755 Like @var{funcaddr} above, but also specifies the name of the source
8756 file explicitly. This is useful if the name of the function does not
8757 specify the function unambiguously, e.g., if there are several
8758 functions with identical names in different source files.
8762 @section Editing Source Files
8763 @cindex editing source files
8766 @kindex e @r{(@code{edit})}
8767 To edit the lines in a source file, use the @code{edit} command.
8768 The editing program of your choice
8769 is invoked with the current line set to
8770 the active line in the program.
8771 Alternatively, there are several ways to specify what part of the file you
8772 want to print if you want to see other parts of the program:
8775 @item edit @var{location}
8776 Edit the source file specified by @code{location}. Editing starts at
8777 that @var{location}, e.g., at the specified source line of the
8778 specified file. @xref{Specify Location}, for all the possible forms
8779 of the @var{location} argument; here are the forms of the @code{edit}
8780 command most commonly used:
8783 @item edit @var{number}
8784 Edit the current source file with @var{number} as the active line number.
8786 @item edit @var{function}
8787 Edit the file containing @var{function} at the beginning of its definition.
8792 @subsection Choosing your Editor
8793 You can customize @value{GDBN} to use any editor you want
8795 The only restriction is that your editor (say @code{ex}), recognizes the
8796 following command-line syntax:
8798 ex +@var{number} file
8800 The optional numeric value +@var{number} specifies the number of the line in
8801 the file where to start editing.}.
8802 By default, it is @file{@value{EDITOR}}, but you can change this
8803 by setting the environment variable @code{EDITOR} before using
8804 @value{GDBN}. For example, to configure @value{GDBN} to use the
8805 @code{vi} editor, you could use these commands with the @code{sh} shell:
8811 or in the @code{csh} shell,
8813 setenv EDITOR /usr/bin/vi
8818 @section Searching Source Files
8819 @cindex searching source files
8821 There are two commands for searching through the current source file for a
8826 @kindex forward-search
8827 @kindex fo @r{(@code{forward-search})}
8828 @item forward-search @var{regexp}
8829 @itemx search @var{regexp}
8830 The command @samp{forward-search @var{regexp}} checks each line,
8831 starting with the one following the last line listed, for a match for
8832 @var{regexp}. It lists the line that is found. You can use the
8833 synonym @samp{search @var{regexp}} or abbreviate the command name as
8836 @kindex reverse-search
8837 @item reverse-search @var{regexp}
8838 The command @samp{reverse-search @var{regexp}} checks each line, starting
8839 with the one before the last line listed and going backward, for a match
8840 for @var{regexp}. It lists the line that is found. You can abbreviate
8841 this command as @code{rev}.
8845 @section Specifying Source Directories
8848 @cindex directories for source files
8849 Executable programs sometimes do not record the directories of the source
8850 files from which they were compiled, just the names. Even when they do,
8851 the directories could be moved between the compilation and your debugging
8852 session. @value{GDBN} has a list of directories to search for source files;
8853 this is called the @dfn{source path}. Each time @value{GDBN} wants a source file,
8854 it tries all the directories in the list, in the order they are present
8855 in the list, until it finds a file with the desired name.
8857 For example, suppose an executable references the file
8858 @file{/usr/src/foo-1.0/lib/foo.c}, and our source path is
8859 @file{/mnt/cross}. The file is first looked up literally; if this
8860 fails, @file{/mnt/cross/usr/src/foo-1.0/lib/foo.c} is tried; if this
8861 fails, @file{/mnt/cross/foo.c} is opened; if this fails, an error
8862 message is printed. @value{GDBN} does not look up the parts of the
8863 source file name, such as @file{/mnt/cross/src/foo-1.0/lib/foo.c}.
8864 Likewise, the subdirectories of the source path are not searched: if
8865 the source path is @file{/mnt/cross}, and the binary refers to
8866 @file{foo.c}, @value{GDBN} would not find it under
8867 @file{/mnt/cross/usr/src/foo-1.0/lib}.
8869 Plain file names, relative file names with leading directories, file
8870 names containing dots, etc.@: are all treated as described above; for
8871 instance, if the source path is @file{/mnt/cross}, and the source file
8872 is recorded as @file{../lib/foo.c}, @value{GDBN} would first try
8873 @file{../lib/foo.c}, then @file{/mnt/cross/../lib/foo.c}, and after
8874 that---@file{/mnt/cross/foo.c}.
8876 Note that the executable search path is @emph{not} used to locate the
8879 Whenever you reset or rearrange the source path, @value{GDBN} clears out
8880 any information it has cached about where source files are found and where
8881 each line is in the file.
8885 When you start @value{GDBN}, its source path includes only @samp{cdir}
8886 and @samp{cwd}, in that order.
8887 To add other directories, use the @code{directory} command.
8889 The search path is used to find both program source files and @value{GDBN}
8890 script files (read using the @samp{-command} option and @samp{source} command).
8892 In addition to the source path, @value{GDBN} provides a set of commands
8893 that manage a list of source path substitution rules. A @dfn{substitution
8894 rule} specifies how to rewrite source directories stored in the program's
8895 debug information in case the sources were moved to a different
8896 directory between compilation and debugging. A rule is made of
8897 two strings, the first specifying what needs to be rewritten in
8898 the path, and the second specifying how it should be rewritten.
8899 In @ref{set substitute-path}, we name these two parts @var{from} and
8900 @var{to} respectively. @value{GDBN} does a simple string replacement
8901 of @var{from} with @var{to} at the start of the directory part of the
8902 source file name, and uses that result instead of the original file
8903 name to look up the sources.
8905 Using the previous example, suppose the @file{foo-1.0} tree has been
8906 moved from @file{/usr/src} to @file{/mnt/cross}, then you can tell
8907 @value{GDBN} to replace @file{/usr/src} in all source path names with
8908 @file{/mnt/cross}. The first lookup will then be
8909 @file{/mnt/cross/foo-1.0/lib/foo.c} in place of the original location
8910 of @file{/usr/src/foo-1.0/lib/foo.c}. To define a source path
8911 substitution rule, use the @code{set substitute-path} command
8912 (@pxref{set substitute-path}).
8914 To avoid unexpected substitution results, a rule is applied only if the
8915 @var{from} part of the directory name ends at a directory separator.
8916 For instance, a rule substituting @file{/usr/source} into
8917 @file{/mnt/cross} will be applied to @file{/usr/source/foo-1.0} but
8918 not to @file{/usr/sourceware/foo-2.0}. And because the substitution
8919 is applied only at the beginning of the directory name, this rule will
8920 not be applied to @file{/root/usr/source/baz.c} either.
8922 In many cases, you can achieve the same result using the @code{directory}
8923 command. However, @code{set substitute-path} can be more efficient in
8924 the case where the sources are organized in a complex tree with multiple
8925 subdirectories. With the @code{directory} command, you need to add each
8926 subdirectory of your project. If you moved the entire tree while
8927 preserving its internal organization, then @code{set substitute-path}
8928 allows you to direct the debugger to all the sources with one single
8931 @code{set substitute-path} is also more than just a shortcut command.
8932 The source path is only used if the file at the original location no
8933 longer exists. On the other hand, @code{set substitute-path} modifies
8934 the debugger behavior to look at the rewritten location instead. So, if
8935 for any reason a source file that is not relevant to your executable is
8936 located at the original location, a substitution rule is the only
8937 method available to point @value{GDBN} at the new location.
8939 @cindex @samp{--with-relocated-sources}
8940 @cindex default source path substitution
8941 You can configure a default source path substitution rule by
8942 configuring @value{GDBN} with the
8943 @samp{--with-relocated-sources=@var{dir}} option. The @var{dir}
8944 should be the name of a directory under @value{GDBN}'s configured
8945 prefix (set with @samp{--prefix} or @samp{--exec-prefix}), and
8946 directory names in debug information under @var{dir} will be adjusted
8947 automatically if the installed @value{GDBN} is moved to a new
8948 location. This is useful if @value{GDBN}, libraries or executables
8949 with debug information and corresponding source code are being moved
8953 @item directory @var{dirname} @dots{}
8954 @item dir @var{dirname} @dots{}
8955 Add directory @var{dirname} to the front of the source path. Several
8956 directory names may be given to this command, separated by @samp{:}
8957 (@samp{;} on MS-DOS and MS-Windows, where @samp{:} usually appears as
8958 part of absolute file names) or
8959 whitespace. You may specify a directory that is already in the source
8960 path; this moves it forward, so @value{GDBN} searches it sooner.
8964 @vindex $cdir@r{, convenience variable}
8965 @vindex $cwd@r{, convenience variable}
8966 @cindex compilation directory
8967 @cindex current directory
8968 @cindex working directory
8969 @cindex directory, current
8970 @cindex directory, compilation
8971 You can use the string @samp{$cdir} to refer to the compilation
8972 directory (if one is recorded), and @samp{$cwd} to refer to the current
8973 working directory. @samp{$cwd} is not the same as @samp{.}---the former
8974 tracks the current working directory as it changes during your @value{GDBN}
8975 session, while the latter is immediately expanded to the current
8976 directory at the time you add an entry to the source path.
8979 Reset the source path to its default value (@samp{$cdir:$cwd} on Unix systems). This requires confirmation.
8981 @c RET-repeat for @code{directory} is explicitly disabled, but since
8982 @c repeating it would be a no-op we do not say that. (thanks to RMS)
8984 @item set directories @var{path-list}
8985 @kindex set directories
8986 Set the source path to @var{path-list}.
8987 @samp{$cdir:$cwd} are added if missing.
8989 @item show directories
8990 @kindex show directories
8991 Print the source path: show which directories it contains.
8993 @anchor{set substitute-path}
8994 @item set substitute-path @var{from} @var{to}
8995 @kindex set substitute-path
8996 Define a source path substitution rule, and add it at the end of the
8997 current list of existing substitution rules. If a rule with the same
8998 @var{from} was already defined, then the old rule is also deleted.
9000 For example, if the file @file{/foo/bar/baz.c} was moved to
9001 @file{/mnt/cross/baz.c}, then the command
9004 (@value{GDBP}) set substitute-path /foo/bar /mnt/cross
9008 will tell @value{GDBN} to replace @samp{/foo/bar} with
9009 @samp{/mnt/cross}, which will allow @value{GDBN} to find the file
9010 @file{baz.c} even though it was moved.
9012 In the case when more than one substitution rule have been defined,
9013 the rules are evaluated one by one in the order where they have been
9014 defined. The first one matching, if any, is selected to perform
9017 For instance, if we had entered the following commands:
9020 (@value{GDBP}) set substitute-path /usr/src/include /mnt/include
9021 (@value{GDBP}) set substitute-path /usr/src /mnt/src
9025 @value{GDBN} would then rewrite @file{/usr/src/include/defs.h} into
9026 @file{/mnt/include/defs.h} by using the first rule. However, it would
9027 use the second rule to rewrite @file{/usr/src/lib/foo.c} into
9028 @file{/mnt/src/lib/foo.c}.
9031 @item unset substitute-path [path]
9032 @kindex unset substitute-path
9033 If a path is specified, search the current list of substitution rules
9034 for a rule that would rewrite that path. Delete that rule if found.
9035 A warning is emitted by the debugger if no rule could be found.
9037 If no path is specified, then all substitution rules are deleted.
9039 @item show substitute-path [path]
9040 @kindex show substitute-path
9041 If a path is specified, then print the source path substitution rule
9042 which would rewrite that path, if any.
9044 If no path is specified, then print all existing source path substitution
9049 If your source path is cluttered with directories that are no longer of
9050 interest, @value{GDBN} may sometimes cause confusion by finding the wrong
9051 versions of source. You can correct the situation as follows:
9055 Use @code{directory} with no argument to reset the source path to its default value.
9058 Use @code{directory} with suitable arguments to reinstall the
9059 directories you want in the source path. You can add all the
9060 directories in one command.
9064 @section Source and Machine Code
9065 @cindex source line and its code address
9067 You can use the command @code{info line} to map source lines to program
9068 addresses (and vice versa), and the command @code{disassemble} to display
9069 a range of addresses as machine instructions. You can use the command
9070 @code{set disassemble-next-line} to set whether to disassemble next
9071 source line when execution stops. When run under @sc{gnu} Emacs
9072 mode, the @code{info line} command causes the arrow to point to the
9073 line specified. Also, @code{info line} prints addresses in symbolic form as
9079 @itemx info line @var{location}
9080 Print the starting and ending addresses of the compiled code for
9081 source line @var{location}. You can specify source lines in any of
9082 the ways documented in @ref{Specify Location}. With no @var{location}
9083 information about the current source line is printed.
9086 For example, we can use @code{info line} to discover the location of
9087 the object code for the first line of function
9088 @code{m4_changequote}:
9091 (@value{GDBP}) info line m4_changequote
9092 Line 895 of "builtin.c" starts at pc 0x634c <m4_changequote> and \
9093 ends at 0x6350 <m4_changequote+4>.
9097 @cindex code address and its source line
9098 We can also inquire (using @code{*@var{addr}} as the form for
9099 @var{location}) what source line covers a particular address:
9101 (@value{GDBP}) info line *0x63ff
9102 Line 926 of "builtin.c" starts at pc 0x63e4 <m4_changequote+152> and \
9103 ends at 0x6404 <m4_changequote+184>.
9106 @cindex @code{$_} and @code{info line}
9107 @cindex @code{x} command, default address
9108 @kindex x@r{(examine), and} info line
9109 After @code{info line}, the default address for the @code{x} command
9110 is changed to the starting address of the line, so that @samp{x/i} is
9111 sufficient to begin examining the machine code (@pxref{Memory,
9112 ,Examining Memory}). Also, this address is saved as the value of the
9113 convenience variable @code{$_} (@pxref{Convenience Vars, ,Convenience
9116 @cindex info line, repeated calls
9117 After @code{info line}, using @code{info line} again without
9118 specifying a location will display information about the next source
9123 @cindex assembly instructions
9124 @cindex instructions, assembly
9125 @cindex machine instructions
9126 @cindex listing machine instructions
9128 @itemx disassemble /m
9129 @itemx disassemble /s
9130 @itemx disassemble /r
9131 This specialized command dumps a range of memory as machine
9132 instructions. It can also print mixed source+disassembly by specifying
9133 the @code{/m} or @code{/s} modifier and print the raw instructions in hex
9134 as well as in symbolic form by specifying the @code{/r} modifier.
9135 The default memory range is the function surrounding the
9136 program counter of the selected frame. A single argument to this
9137 command is a program counter value; @value{GDBN} dumps the function
9138 surrounding this value. When two arguments are given, they should
9139 be separated by a comma, possibly surrounded by whitespace. The
9140 arguments specify a range of addresses to dump, in one of two forms:
9143 @item @var{start},@var{end}
9144 the addresses from @var{start} (inclusive) to @var{end} (exclusive)
9145 @item @var{start},+@var{length}
9146 the addresses from @var{start} (inclusive) to
9147 @code{@var{start}+@var{length}} (exclusive).
9151 When 2 arguments are specified, the name of the function is also
9152 printed (since there could be several functions in the given range).
9154 The argument(s) can be any expression yielding a numeric value, such as
9155 @samp{0x32c4}, @samp{&main+10} or @samp{$pc - 8}.
9157 If the range of memory being disassembled contains current program counter,
9158 the instruction at that location is shown with a @code{=>} marker.
9161 The following example shows the disassembly of a range of addresses of
9162 HP PA-RISC 2.0 code:
9165 (@value{GDBP}) disas 0x32c4, 0x32e4
9166 Dump of assembler code from 0x32c4 to 0x32e4:
9167 0x32c4 <main+204>: addil 0,dp
9168 0x32c8 <main+208>: ldw 0x22c(sr0,r1),r26
9169 0x32cc <main+212>: ldil 0x3000,r31
9170 0x32d0 <main+216>: ble 0x3f8(sr4,r31)
9171 0x32d4 <main+220>: ldo 0(r31),rp
9172 0x32d8 <main+224>: addil -0x800,dp
9173 0x32dc <main+228>: ldo 0x588(r1),r26
9174 0x32e0 <main+232>: ldil 0x3000,r31
9175 End of assembler dump.
9178 Here is an example showing mixed source+assembly for Intel x86
9179 with @code{/m} or @code{/s}, when the program is stopped just after
9180 function prologue in a non-optimized function with no inline code.
9183 (@value{GDBP}) disas /m main
9184 Dump of assembler code for function main:
9186 0x08048330 <+0>: push %ebp
9187 0x08048331 <+1>: mov %esp,%ebp
9188 0x08048333 <+3>: sub $0x8,%esp
9189 0x08048336 <+6>: and $0xfffffff0,%esp
9190 0x08048339 <+9>: sub $0x10,%esp
9192 6 printf ("Hello.\n");
9193 => 0x0804833c <+12>: movl $0x8048440,(%esp)
9194 0x08048343 <+19>: call 0x8048284 <puts@@plt>
9198 0x08048348 <+24>: mov $0x0,%eax
9199 0x0804834d <+29>: leave
9200 0x0804834e <+30>: ret
9202 End of assembler dump.
9205 The @code{/m} option is deprecated as its output is not useful when
9206 there is either inlined code or re-ordered code.
9207 The @code{/s} option is the preferred choice.
9208 Here is an example for AMD x86-64 showing the difference between
9209 @code{/m} output and @code{/s} output.
9210 This example has one inline function defined in a header file,
9211 and the code is compiled with @samp{-O2} optimization.
9212 Note how the @code{/m} output is missing the disassembly of
9213 several instructions that are present in the @code{/s} output.
9243 (@value{GDBP}) disas /m main
9244 Dump of assembler code for function main:
9248 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9249 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9253 0x000000000040041d <+29>: xor %eax,%eax
9254 0x000000000040041f <+31>: retq
9255 0x0000000000400420 <+32>: add %eax,%eax
9256 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9258 End of assembler dump.
9259 (@value{GDBP}) disas /s main
9260 Dump of assembler code for function main:
9264 0x0000000000400400 <+0>: mov 0x200c2e(%rip),%eax # 0x601034 <y>
9268 0x0000000000400406 <+6>: test %eax,%eax
9269 0x0000000000400408 <+8>: js 0x400420 <main+32>
9274 0x000000000040040a <+10>: lea 0xa(%rax),%edx
9275 0x000000000040040d <+13>: test %eax,%eax
9276 0x000000000040040f <+15>: mov $0x1,%eax
9277 0x0000000000400414 <+20>: cmovne %edx,%eax
9281 0x0000000000400417 <+23>: mov %eax,0x200c13(%rip) # 0x601030 <x>
9285 0x000000000040041d <+29>: xor %eax,%eax
9286 0x000000000040041f <+31>: retq
9290 0x0000000000400420 <+32>: add %eax,%eax
9291 0x0000000000400422 <+34>: jmp 0x400417 <main+23>
9292 End of assembler dump.
9295 Here is another example showing raw instructions in hex for AMD x86-64,
9298 (gdb) disas /r 0x400281,+10
9299 Dump of assembler code from 0x400281 to 0x40028b:
9300 0x0000000000400281: 38 36 cmp %dh,(%rsi)
9301 0x0000000000400283: 2d 36 34 2e 73 sub $0x732e3436,%eax
9302 0x0000000000400288: 6f outsl %ds:(%rsi),(%dx)
9303 0x0000000000400289: 2e 32 00 xor %cs:(%rax),%al
9304 End of assembler dump.
9307 Addresses cannot be specified as a location (@pxref{Specify Location}).
9308 So, for example, if you want to disassemble function @code{bar}
9309 in file @file{foo.c}, you must type @samp{disassemble 'foo.c'::bar}
9310 and not @samp{disassemble foo.c:bar}.
9312 Some architectures have more than one commonly-used set of instruction
9313 mnemonics or other syntax.
9315 For programs that were dynamically linked and use shared libraries,
9316 instructions that call functions or branch to locations in the shared
9317 libraries might show a seemingly bogus location---it's actually a
9318 location of the relocation table. On some architectures, @value{GDBN}
9319 might be able to resolve these to actual function names.
9322 @kindex set disassembler-options
9323 @cindex disassembler options
9324 @item set disassembler-options @var{option1}[,@var{option2}@dots{}]
9325 This command controls the passing of target specific information to
9326 the disassembler. For a list of valid options, please refer to the
9327 @code{-M}/@code{--disassembler-options} section of the @samp{objdump}
9328 manual and/or the output of @kbd{objdump --help}
9329 (@pxref{objdump,,objdump,binutils,The GNU Binary Utilities}).
9330 The default value is the empty string.
9332 If it is necessary to specify more than one disassembler option, then
9333 multiple options can be placed together into a comma separated list.
9334 Currently this command is only supported on targets ARM, MIPS, PowerPC
9337 @kindex show disassembler-options
9338 @item show disassembler-options
9339 Show the current setting of the disassembler options.
9343 @kindex set disassembly-flavor
9344 @cindex Intel disassembly flavor
9345 @cindex AT&T disassembly flavor
9346 @item set disassembly-flavor @var{instruction-set}
9347 Select the instruction set to use when disassembling the
9348 program via the @code{disassemble} or @code{x/i} commands.
9350 Currently this command is only defined for the Intel x86 family. You
9351 can set @var{instruction-set} to either @code{intel} or @code{att}.
9352 The default is @code{att}, the AT&T flavor used by default by Unix
9353 assemblers for x86-based targets.
9355 @kindex show disassembly-flavor
9356 @item show disassembly-flavor
9357 Show the current setting of the disassembly flavor.
9361 @kindex set disassemble-next-line
9362 @kindex show disassemble-next-line
9363 @item set disassemble-next-line
9364 @itemx show disassemble-next-line
9365 Control whether or not @value{GDBN} will disassemble the next source
9366 line or instruction when execution stops. If ON, @value{GDBN} will
9367 display disassembly of the next source line when execution of the
9368 program being debugged stops. This is @emph{in addition} to
9369 displaying the source line itself, which @value{GDBN} always does if
9370 possible. If the next source line cannot be displayed for some reason
9371 (e.g., if @value{GDBN} cannot find the source file, or there's no line
9372 info in the debug info), @value{GDBN} will display disassembly of the
9373 next @emph{instruction} instead of showing the next source line. If
9374 AUTO, @value{GDBN} will display disassembly of next instruction only
9375 if the source line cannot be displayed. This setting causes
9376 @value{GDBN} to display some feedback when you step through a function
9377 with no line info or whose source file is unavailable. The default is
9378 OFF, which means never display the disassembly of the next line or
9384 @chapter Examining Data
9386 @cindex printing data
9387 @cindex examining data
9390 The usual way to examine data in your program is with the @code{print}
9391 command (abbreviated @code{p}), or its synonym @code{inspect}. It
9392 evaluates and prints the value of an expression of the language your
9393 program is written in (@pxref{Languages, ,Using @value{GDBN} with
9394 Different Languages}). It may also print the expression using a
9395 Python-based pretty-printer (@pxref{Pretty Printing}).
9398 @item print [[@var{options}] --] @var{expr}
9399 @itemx print [[@var{options}] --] /@var{f} @var{expr}
9400 @var{expr} is an expression (in the source language). By default the
9401 value of @var{expr} is printed in a format appropriate to its data type;
9402 you can choose a different format by specifying @samp{/@var{f}}, where
9403 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
9406 @anchor{print options}
9407 The @code{print} command supports a number of options that allow
9408 overriding relevant global print settings as set by @code{set print}
9412 @item -address [@code{on}|@code{off}]
9413 Set printing of addresses.
9414 Related setting: @ref{set print address}.
9416 @item -array [@code{on}|@code{off}]
9417 Pretty formatting of arrays.
9418 Related setting: @ref{set print array}.
9420 @item -array-indexes [@code{on}|@code{off}]
9421 Set printing of array indexes.
9422 Related setting: @ref{set print array-indexes}.
9424 @item -elements @var{number-of-elements}|@code{unlimited}
9425 Set limit on string chars or array elements to print. The value
9426 @code{unlimited} causes there to be no limit. Related setting:
9427 @ref{set print elements}.
9429 @item -max-depth @var{depth}|@code{unlimited}
9430 Set the threshold after which nested structures are replaced with
9431 ellipsis. Related setting: @ref{set print max-depth}.
9433 @item -null-stop [@code{on}|@code{off}]
9434 Set printing of char arrays to stop at first null char. Related
9435 setting: @ref{set print null-stop}.
9437 @item -object [@code{on}|@code{off}]
9438 Set printing C@t{++} virtual function tables. Related setting:
9439 @ref{set print object}.
9441 @item -pretty [@code{on}|@code{off}]
9442 Set pretty formatting of structures. Related setting: @ref{set print
9445 @item -repeats @var{number-of-repeats}|@code{unlimited}
9446 Set threshold for repeated print elements. @code{unlimited} causes
9447 all elements to be individually printed. Related setting: @ref{set
9450 @item -static-members [@code{on}|@code{off}]
9451 Set printing C@t{++} static members. Related setting: @ref{set print
9454 @item -symbol [@code{on}|@code{off}]
9455 Set printing of symbol names when printing pointers. Related setting:
9456 @ref{set print symbol}.
9458 @item -union [@code{on}|@code{off}]
9459 Set printing of unions interior to structures. Related setting:
9460 @ref{set print union}.
9462 @item -vtbl [@code{on}|@code{off}]
9463 Set printing of C++ virtual function tables. Related setting:
9464 @ref{set print vtbl}.
9467 Because the @code{print} command accepts arbitrary expressions which
9468 may look like options (including abbreviations), if you specify any
9469 command option, then you must use a double dash (@code{--}) to mark
9470 the end of option processing.
9472 For example, this prints the value of the @code{-r} expression:
9475 (@value{GDBP}) print -r
9478 While this repeats the last value in the value history (see below)
9479 with the @code{-raw} option in effect:
9482 (@value{GDBP}) print -r --
9485 Here is an example including both on option and an expression:
9489 (@value{GDBP}) print -pretty -- *myptr
9501 @item print [@var{options}]
9502 @itemx print [@var{options}] /@var{f}
9503 @cindex reprint the last value
9504 If you omit @var{expr}, @value{GDBN} displays the last value again (from the
9505 @dfn{value history}; @pxref{Value History, ,Value History}). This allows you to
9506 conveniently inspect the same value in an alternative format.
9509 A more low-level way of examining data is with the @code{x} command.
9510 It examines data in memory at a specified address and prints it in a
9511 specified format. @xref{Memory, ,Examining Memory}.
9513 If you are interested in information about types, or about how the
9514 fields of a struct or a class are declared, use the @code{ptype @var{exp}}
9515 command rather than @code{print}. @xref{Symbols, ,Examining the Symbol
9518 @cindex exploring hierarchical data structures
9520 Another way of examining values of expressions and type information is
9521 through the Python extension command @code{explore} (available only if
9522 the @value{GDBN} build is configured with @code{--with-python}). It
9523 offers an interactive way to start at the highest level (or, the most
9524 abstract level) of the data type of an expression (or, the data type
9525 itself) and explore all the way down to leaf scalar values/fields
9526 embedded in the higher level data types.
9529 @item explore @var{arg}
9530 @var{arg} is either an expression (in the source language), or a type
9531 visible in the current context of the program being debugged.
9534 The working of the @code{explore} command can be illustrated with an
9535 example. If a data type @code{struct ComplexStruct} is defined in your
9545 struct ComplexStruct
9547 struct SimpleStruct *ss_p;
9553 followed by variable declarations as
9556 struct SimpleStruct ss = @{ 10, 1.11 @};
9557 struct ComplexStruct cs = @{ &ss, @{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 @} @};
9561 then, the value of the variable @code{cs} can be explored using the
9562 @code{explore} command as follows.
9566 The value of `cs' is a struct/class of type `struct ComplexStruct' with
9567 the following fields:
9569 ss_p = <Enter 0 to explore this field of type `struct SimpleStruct *'>
9570 arr = <Enter 1 to explore this field of type `int [10]'>
9572 Enter the field number of choice:
9576 Since the fields of @code{cs} are not scalar values, you are being
9577 prompted to chose the field you want to explore. Let's say you choose
9578 the field @code{ss_p} by entering @code{0}. Then, since this field is a
9579 pointer, you will be asked if it is pointing to a single value. From
9580 the declaration of @code{cs} above, it is indeed pointing to a single
9581 value, hence you enter @code{y}. If you enter @code{n}, then you will
9582 be asked if it were pointing to an array of values, in which case this
9583 field will be explored as if it were an array.
9586 `cs.ss_p' is a pointer to a value of type `struct SimpleStruct'
9587 Continue exploring it as a pointer to a single value [y/n]: y
9588 The value of `*(cs.ss_p)' is a struct/class of type `struct
9589 SimpleStruct' with the following fields:
9591 i = 10 .. (Value of type `int')
9592 d = 1.1100000000000001 .. (Value of type `double')
9594 Press enter to return to parent value:
9598 If the field @code{arr} of @code{cs} was chosen for exploration by
9599 entering @code{1} earlier, then since it is as array, you will be
9600 prompted to enter the index of the element in the array that you want
9604 `cs.arr' is an array of `int'.
9605 Enter the index of the element you want to explore in `cs.arr': 5
9607 `(cs.arr)[5]' is a scalar value of type `int'.
9611 Press enter to return to parent value:
9614 In general, at any stage of exploration, you can go deeper towards the
9615 leaf values by responding to the prompts appropriately, or hit the
9616 return key to return to the enclosing data structure (the @i{higher}
9617 level data structure).
9619 Similar to exploring values, you can use the @code{explore} command to
9620 explore types. Instead of specifying a value (which is typically a
9621 variable name or an expression valid in the current context of the
9622 program being debugged), you specify a type name. If you consider the
9623 same example as above, your can explore the type
9624 @code{struct ComplexStruct} by passing the argument
9625 @code{struct ComplexStruct} to the @code{explore} command.
9628 (gdb) explore struct ComplexStruct
9632 By responding to the prompts appropriately in the subsequent interactive
9633 session, you can explore the type @code{struct ComplexStruct} in a
9634 manner similar to how the value @code{cs} was explored in the above
9637 The @code{explore} command also has two sub-commands,
9638 @code{explore value} and @code{explore type}. The former sub-command is
9639 a way to explicitly specify that value exploration of the argument is
9640 being invoked, while the latter is a way to explicitly specify that type
9641 exploration of the argument is being invoked.
9644 @item explore value @var{expr}
9645 @cindex explore value
9646 This sub-command of @code{explore} explores the value of the
9647 expression @var{expr} (if @var{expr} is an expression valid in the
9648 current context of the program being debugged). The behavior of this
9649 command is identical to that of the behavior of the @code{explore}
9650 command being passed the argument @var{expr}.
9652 @item explore type @var{arg}
9653 @cindex explore type
9654 This sub-command of @code{explore} explores the type of @var{arg} (if
9655 @var{arg} is a type visible in the current context of program being
9656 debugged), or the type of the value/expression @var{arg} (if @var{arg}
9657 is an expression valid in the current context of the program being
9658 debugged). If @var{arg} is a type, then the behavior of this command is
9659 identical to that of the @code{explore} command being passed the
9660 argument @var{arg}. If @var{arg} is an expression, then the behavior of
9661 this command will be identical to that of the @code{explore} command
9662 being passed the type of @var{arg} as the argument.
9666 * Expressions:: Expressions
9667 * Ambiguous Expressions:: Ambiguous Expressions
9668 * Variables:: Program variables
9669 * Arrays:: Artificial arrays
9670 * Output Formats:: Output formats
9671 * Memory:: Examining memory
9672 * Auto Display:: Automatic display
9673 * Print Settings:: Print settings
9674 * Pretty Printing:: Python pretty printing
9675 * Value History:: Value history
9676 * Convenience Vars:: Convenience variables
9677 * Convenience Funs:: Convenience functions
9678 * Registers:: Registers
9679 * Floating Point Hardware:: Floating point hardware
9680 * Vector Unit:: Vector Unit
9681 * OS Information:: Auxiliary data provided by operating system
9682 * Memory Region Attributes:: Memory region attributes
9683 * Dump/Restore Files:: Copy between memory and a file
9684 * Core File Generation:: Cause a program dump its core
9685 * Character Sets:: Debugging programs that use a different
9686 character set than GDB does
9687 * Caching Target Data:: Data caching for targets
9688 * Searching Memory:: Searching memory for a sequence of bytes
9689 * Value Sizes:: Managing memory allocated for values
9693 @section Expressions
9696 @code{print} and many other @value{GDBN} commands accept an expression and
9697 compute its value. Any kind of constant, variable or operator defined
9698 by the programming language you are using is valid in an expression in
9699 @value{GDBN}. This includes conditional expressions, function calls,
9700 casts, and string constants. It also includes preprocessor macros, if
9701 you compiled your program to include this information; see
9704 @cindex arrays in expressions
9705 @value{GDBN} supports array constants in expressions input by
9706 the user. The syntax is @{@var{element}, @var{element}@dots{}@}. For example,
9707 you can use the command @code{print @{1, 2, 3@}} to create an array
9708 of three integers. If you pass an array to a function or assign it
9709 to a program variable, @value{GDBN} copies the array to memory that
9710 is @code{malloc}ed in the target program.
9712 Because C is so widespread, most of the expressions shown in examples in
9713 this manual are in C. @xref{Languages, , Using @value{GDBN} with Different
9714 Languages}, for information on how to use expressions in other
9717 In this section, we discuss operators that you can use in @value{GDBN}
9718 expressions regardless of your programming language.
9720 @cindex casts, in expressions
9721 Casts are supported in all languages, not just in C, because it is so
9722 useful to cast a number into a pointer in order to examine a structure
9723 at that address in memory.
9724 @c FIXME: casts supported---Mod2 true?
9726 @value{GDBN} supports these operators, in addition to those common
9727 to programming languages:
9731 @samp{@@} is a binary operator for treating parts of memory as arrays.
9732 @xref{Arrays, ,Artificial Arrays}, for more information.
9735 @samp{::} allows you to specify a variable in terms of the file or
9736 function where it is defined. @xref{Variables, ,Program Variables}.
9738 @cindex @{@var{type}@}
9739 @cindex type casting memory
9740 @cindex memory, viewing as typed object
9741 @cindex casts, to view memory
9742 @item @{@var{type}@} @var{addr}
9743 Refers to an object of type @var{type} stored at address @var{addr} in
9744 memory. The address @var{addr} may be any expression whose value is
9745 an integer or pointer (but parentheses are required around binary
9746 operators, just as in a cast). This construct is allowed regardless
9747 of what kind of data is normally supposed to reside at @var{addr}.
9750 @node Ambiguous Expressions
9751 @section Ambiguous Expressions
9752 @cindex ambiguous expressions
9754 Expressions can sometimes contain some ambiguous elements. For instance,
9755 some programming languages (notably Ada, C@t{++} and Objective-C) permit
9756 a single function name to be defined several times, for application in
9757 different contexts. This is called @dfn{overloading}. Another example
9758 involving Ada is generics. A @dfn{generic package} is similar to C@t{++}
9759 templates and is typically instantiated several times, resulting in
9760 the same function name being defined in different contexts.
9762 In some cases and depending on the language, it is possible to adjust
9763 the expression to remove the ambiguity. For instance in C@t{++}, you
9764 can specify the signature of the function you want to break on, as in
9765 @kbd{break @var{function}(@var{types})}. In Ada, using the fully
9766 qualified name of your function often makes the expression unambiguous
9769 When an ambiguity that needs to be resolved is detected, the debugger
9770 has the capability to display a menu of numbered choices for each
9771 possibility, and then waits for the selection with the prompt @samp{>}.
9772 The first option is always @samp{[0] cancel}, and typing @kbd{0 @key{RET}}
9773 aborts the current command. If the command in which the expression was
9774 used allows more than one choice to be selected, the next option in the
9775 menu is @samp{[1] all}, and typing @kbd{1 @key{RET}} selects all possible
9778 For example, the following session excerpt shows an attempt to set a
9779 breakpoint at the overloaded symbol @code{String::after}.
9780 We choose three particular definitions of that function name:
9782 @c FIXME! This is likely to change to show arg type lists, at least
9785 (@value{GDBP}) b String::after
9788 [2] file:String.cc; line number:867
9789 [3] file:String.cc; line number:860
9790 [4] file:String.cc; line number:875
9791 [5] file:String.cc; line number:853
9792 [6] file:String.cc; line number:846
9793 [7] file:String.cc; line number:735
9795 Breakpoint 1 at 0xb26c: file String.cc, line 867.
9796 Breakpoint 2 at 0xb344: file String.cc, line 875.
9797 Breakpoint 3 at 0xafcc: file String.cc, line 846.
9798 Multiple breakpoints were set.
9799 Use the "delete" command to delete unwanted
9806 @kindex set multiple-symbols
9807 @item set multiple-symbols @var{mode}
9808 @cindex multiple-symbols menu
9810 This option allows you to adjust the debugger behavior when an expression
9813 By default, @var{mode} is set to @code{all}. If the command with which
9814 the expression is used allows more than one choice, then @value{GDBN}
9815 automatically selects all possible choices. For instance, inserting
9816 a breakpoint on a function using an ambiguous name results in a breakpoint
9817 inserted on each possible match. However, if a unique choice must be made,
9818 then @value{GDBN} uses the menu to help you disambiguate the expression.
9819 For instance, printing the address of an overloaded function will result
9820 in the use of the menu.
9822 When @var{mode} is set to @code{ask}, the debugger always uses the menu
9823 when an ambiguity is detected.
9825 Finally, when @var{mode} is set to @code{cancel}, the debugger reports
9826 an error due to the ambiguity and the command is aborted.
9828 @kindex show multiple-symbols
9829 @item show multiple-symbols
9830 Show the current value of the @code{multiple-symbols} setting.
9834 @section Program Variables
9836 The most common kind of expression to use is the name of a variable
9839 Variables in expressions are understood in the selected stack frame
9840 (@pxref{Selection, ,Selecting a Frame}); they must be either:
9844 global (or file-static)
9851 visible according to the scope rules of the
9852 programming language from the point of execution in that frame
9855 @noindent This means that in the function
9870 you can examine and use the variable @code{a} whenever your program is
9871 executing within the function @code{foo}, but you can only use or
9872 examine the variable @code{b} while your program is executing inside
9873 the block where @code{b} is declared.
9875 @cindex variable name conflict
9876 There is an exception: you can refer to a variable or function whose
9877 scope is a single source file even if the current execution point is not
9878 in this file. But it is possible to have more than one such variable or
9879 function with the same name (in different source files). If that
9880 happens, referring to that name has unpredictable effects. If you wish,
9881 you can specify a static variable in a particular function or file by
9882 using the colon-colon (@code{::}) notation:
9884 @cindex colon-colon, context for variables/functions
9886 @c info cannot cope with a :: index entry, but why deprive hard copy readers?
9887 @cindex @code{::}, context for variables/functions
9890 @var{file}::@var{variable}
9891 @var{function}::@var{variable}
9895 Here @var{file} or @var{function} is the name of the context for the
9896 static @var{variable}. In the case of file names, you can use quotes to
9897 make sure @value{GDBN} parses the file name as a single word---for example,
9898 to print a global value of @code{x} defined in @file{f2.c}:
9901 (@value{GDBP}) p 'f2.c'::x
9904 The @code{::} notation is normally used for referring to
9905 static variables, since you typically disambiguate uses of local variables
9906 in functions by selecting the appropriate frame and using the
9907 simple name of the variable. However, you may also use this notation
9908 to refer to local variables in frames enclosing the selected frame:
9917 process (a); /* Stop here */
9928 For example, if there is a breakpoint at the commented line,
9929 here is what you might see
9930 when the program stops after executing the call @code{bar(0)}:
9935 (@value{GDBP}) p bar::a
9938 #2 0x080483d0 in foo (a=5) at foobar.c:12
9941 (@value{GDBP}) p bar::a
9945 @cindex C@t{++} scope resolution
9946 These uses of @samp{::} are very rarely in conflict with the very
9947 similar use of the same notation in C@t{++}. When they are in
9948 conflict, the C@t{++} meaning takes precedence; however, this can be
9949 overridden by quoting the file or function name with single quotes.
9951 For example, suppose the program is stopped in a method of a class
9952 that has a field named @code{includefile}, and there is also an
9953 include file named @file{includefile} that defines a variable,
9957 (@value{GDBP}) p includefile
9959 (@value{GDBP}) p includefile::some_global
9960 A syntax error in expression, near `'.
9961 (@value{GDBP}) p 'includefile'::some_global
9965 @cindex wrong values
9966 @cindex variable values, wrong
9967 @cindex function entry/exit, wrong values of variables
9968 @cindex optimized code, wrong values of variables
9970 @emph{Warning:} Occasionally, a local variable may appear to have the
9971 wrong value at certain points in a function---just after entry to a new
9972 scope, and just before exit.
9974 You may see this problem when you are stepping by machine instructions.
9975 This is because, on most machines, it takes more than one instruction to
9976 set up a stack frame (including local variable definitions); if you are
9977 stepping by machine instructions, variables may appear to have the wrong
9978 values until the stack frame is completely built. On exit, it usually
9979 also takes more than one machine instruction to destroy a stack frame;
9980 after you begin stepping through that group of instructions, local
9981 variable definitions may be gone.
9983 This may also happen when the compiler does significant optimizations.
9984 To be sure of always seeing accurate values, turn off all optimization
9987 @cindex ``No symbol "foo" in current context''
9988 Another possible effect of compiler optimizations is to optimize
9989 unused variables out of existence, or assign variables to registers (as
9990 opposed to memory addresses). Depending on the support for such cases
9991 offered by the debug info format used by the compiler, @value{GDBN}
9992 might not be able to display values for such local variables. If that
9993 happens, @value{GDBN} will print a message like this:
9996 No symbol "foo" in current context.
9999 To solve such problems, either recompile without optimizations, or use a
10000 different debug info format, if the compiler supports several such
10001 formats. @xref{Compilation}, for more information on choosing compiler
10002 options. @xref{C, ,C and C@t{++}}, for more information about debug
10003 info formats that are best suited to C@t{++} programs.
10005 If you ask to print an object whose contents are unknown to
10006 @value{GDBN}, e.g., because its data type is not completely specified
10007 by the debug information, @value{GDBN} will say @samp{<incomplete
10008 type>}. @xref{Symbols, incomplete type}, for more about this.
10010 @cindex no debug info variables
10011 If you try to examine or use the value of a (global) variable for
10012 which @value{GDBN} has no type information, e.g., because the program
10013 includes no debug information, @value{GDBN} displays an error message.
10014 @xref{Symbols, unknown type}, for more about unknown types. If you
10015 cast the variable to its declared type, @value{GDBN} gets the
10016 variable's value using the cast-to type as the variable's type. For
10017 example, in a C program:
10020 (@value{GDBP}) p var
10021 'var' has unknown type; cast it to its declared type
10022 (@value{GDBP}) p (float) var
10026 If you append @kbd{@@entry} string to a function parameter name you get its
10027 value at the time the function got called. If the value is not available an
10028 error message is printed. Entry values are available only with some compilers.
10029 Entry values are normally also printed at the function parameter list according
10030 to @ref{set print entry-values}.
10033 Breakpoint 1, d (i=30) at gdb.base/entry-value.c:29
10039 (gdb) print i@@entry
10043 Strings are identified as arrays of @code{char} values without specified
10044 signedness. Arrays of either @code{signed char} or @code{unsigned char} get
10045 printed as arrays of 1 byte sized integers. @code{-fsigned-char} or
10046 @code{-funsigned-char} @value{NGCC} options have no effect as @value{GDBN}
10047 defines literal string type @code{"char"} as @code{char} without a sign.
10052 signed char var1[] = "A";
10055 You get during debugging
10060 $2 = @{65 'A', 0 '\0'@}
10064 @section Artificial Arrays
10066 @cindex artificial array
10068 @kindex @@@r{, referencing memory as an array}
10069 It is often useful to print out several successive objects of the
10070 same type in memory; a section of an array, or an array of
10071 dynamically determined size for which only a pointer exists in the
10074 You can do this by referring to a contiguous span of memory as an
10075 @dfn{artificial array}, using the binary operator @samp{@@}. The left
10076 operand of @samp{@@} should be the first element of the desired array
10077 and be an individual object. The right operand should be the desired length
10078 of the array. The result is an array value whose elements are all of
10079 the type of the left argument. The first element is actually the left
10080 argument; the second element comes from bytes of memory immediately
10081 following those that hold the first element, and so on. Here is an
10082 example. If a program says
10085 int *array = (int *) malloc (len * sizeof (int));
10089 you can print the contents of @code{array} with
10095 The left operand of @samp{@@} must reside in memory. Array values made
10096 with @samp{@@} in this way behave just like other arrays in terms of
10097 subscripting, and are coerced to pointers when used in expressions.
10098 Artificial arrays most often appear in expressions via the value history
10099 (@pxref{Value History, ,Value History}), after printing one out.
10101 Another way to create an artificial array is to use a cast.
10102 This re-interprets a value as if it were an array.
10103 The value need not be in memory:
10105 (@value{GDBP}) p/x (short[2])0x12345678
10106 $1 = @{0x1234, 0x5678@}
10109 As a convenience, if you leave the array length out (as in
10110 @samp{(@var{type}[])@var{value}}) @value{GDBN} calculates the size to fill
10111 the value (as @samp{sizeof(@var{value})/sizeof(@var{type})}:
10113 (@value{GDBP}) p/x (short[])0x12345678
10114 $2 = @{0x1234, 0x5678@}
10117 Sometimes the artificial array mechanism is not quite enough; in
10118 moderately complex data structures, the elements of interest may not
10119 actually be adjacent---for example, if you are interested in the values
10120 of pointers in an array. One useful work-around in this situation is
10121 to use a convenience variable (@pxref{Convenience Vars, ,Convenience
10122 Variables}) as a counter in an expression that prints the first
10123 interesting value, and then repeat that expression via @key{RET}. For
10124 instance, suppose you have an array @code{dtab} of pointers to
10125 structures, and you are interested in the values of a field @code{fv}
10126 in each structure. Here is an example of what you might type:
10136 @node Output Formats
10137 @section Output Formats
10139 @cindex formatted output
10140 @cindex output formats
10141 By default, @value{GDBN} prints a value according to its data type. Sometimes
10142 this is not what you want. For example, you might want to print a number
10143 in hex, or a pointer in decimal. Or you might want to view data in memory
10144 at a certain address as a character string or as an instruction. To do
10145 these things, specify an @dfn{output format} when you print a value.
10147 The simplest use of output formats is to say how to print a value
10148 already computed. This is done by starting the arguments of the
10149 @code{print} command with a slash and a format letter. The format
10150 letters supported are:
10154 Regard the bits of the value as an integer, and print the integer in
10158 Print as integer in signed decimal.
10161 Print as integer in unsigned decimal.
10164 Print as integer in octal.
10167 Print as integer in binary. The letter @samp{t} stands for ``two''.
10168 @footnote{@samp{b} cannot be used because these format letters are also
10169 used with the @code{x} command, where @samp{b} stands for ``byte'';
10170 see @ref{Memory,,Examining Memory}.}
10173 @cindex unknown address, locating
10174 @cindex locate address
10175 Print as an address, both absolute in hexadecimal and as an offset from
10176 the nearest preceding symbol. You can use this format used to discover
10177 where (in what function) an unknown address is located:
10180 (@value{GDBP}) p/a 0x54320
10181 $3 = 0x54320 <_initialize_vx+396>
10185 The command @code{info symbol 0x54320} yields similar results.
10186 @xref{Symbols, info symbol}.
10189 Regard as an integer and print it as a character constant. This
10190 prints both the numerical value and its character representation. The
10191 character representation is replaced with the octal escape @samp{\nnn}
10192 for characters outside the 7-bit @sc{ascii} range.
10194 Without this format, @value{GDBN} displays @code{char},
10195 @w{@code{unsigned char}}, and @w{@code{signed char}} data as character
10196 constants. Single-byte members of vectors are displayed as integer
10200 Regard the bits of the value as a floating point number and print
10201 using typical floating point syntax.
10204 @cindex printing strings
10205 @cindex printing byte arrays
10206 Regard as a string, if possible. With this format, pointers to single-byte
10207 data are displayed as null-terminated strings and arrays of single-byte data
10208 are displayed as fixed-length strings. Other values are displayed in their
10211 Without this format, @value{GDBN} displays pointers to and arrays of
10212 @code{char}, @w{@code{unsigned char}}, and @w{@code{signed char}} as
10213 strings. Single-byte members of a vector are displayed as an integer
10217 Like @samp{x} formatting, the value is treated as an integer and
10218 printed as hexadecimal, but leading zeros are printed to pad the value
10219 to the size of the integer type.
10222 @cindex raw printing
10223 Print using the @samp{raw} formatting. By default, @value{GDBN} will
10224 use a Python-based pretty-printer, if one is available (@pxref{Pretty
10225 Printing}). This typically results in a higher-level display of the
10226 value's contents. The @samp{r} format bypasses any Python
10227 pretty-printer which might exist.
10230 For example, to print the program counter in hex (@pxref{Registers}), type
10237 Note that no space is required before the slash; this is because command
10238 names in @value{GDBN} cannot contain a slash.
10240 To reprint the last value in the value history with a different format,
10241 you can use the @code{print} command with just a format and no
10242 expression. For example, @samp{p/x} reprints the last value in hex.
10245 @section Examining Memory
10247 You can use the command @code{x} (for ``examine'') to examine memory in
10248 any of several formats, independently of your program's data types.
10250 @cindex examining memory
10252 @kindex x @r{(examine memory)}
10253 @item x/@var{nfu} @var{addr}
10254 @itemx x @var{addr}
10256 Use the @code{x} command to examine memory.
10259 @var{n}, @var{f}, and @var{u} are all optional parameters that specify how
10260 much memory to display and how to format it; @var{addr} is an
10261 expression giving the address where you want to start displaying memory.
10262 If you use defaults for @var{nfu}, you need not type the slash @samp{/}.
10263 Several commands set convenient defaults for @var{addr}.
10266 @item @var{n}, the repeat count
10267 The repeat count is a decimal integer; the default is 1. It specifies
10268 how much memory (counting by units @var{u}) to display. If a negative
10269 number is specified, memory is examined backward from @var{addr}.
10270 @c This really is **decimal**; unaffected by 'set radix' as of GDB
10273 @item @var{f}, the display format
10274 The display format is one of the formats used by @code{print}
10275 (@samp{x}, @samp{d}, @samp{u}, @samp{o}, @samp{t}, @samp{a}, @samp{c},
10276 @samp{f}, @samp{s}), and in addition @samp{i} (for machine instructions).
10277 The default is @samp{x} (hexadecimal) initially. The default changes
10278 each time you use either @code{x} or @code{print}.
10280 @item @var{u}, the unit size
10281 The unit size is any of
10287 Halfwords (two bytes).
10289 Words (four bytes). This is the initial default.
10291 Giant words (eight bytes).
10294 Each time you specify a unit size with @code{x}, that size becomes the
10295 default unit the next time you use @code{x}. For the @samp{i} format,
10296 the unit size is ignored and is normally not written. For the @samp{s} format,
10297 the unit size defaults to @samp{b}, unless it is explicitly given.
10298 Use @kbd{x /hs} to display 16-bit char strings and @kbd{x /ws} to display
10299 32-bit strings. The next use of @kbd{x /s} will again display 8-bit strings.
10300 Note that the results depend on the programming language of the
10301 current compilation unit. If the language is C, the @samp{s}
10302 modifier will use the UTF-16 encoding while @samp{w} will use
10303 UTF-32. The encoding is set by the programming language and cannot
10306 @item @var{addr}, starting display address
10307 @var{addr} is the address where you want @value{GDBN} to begin displaying
10308 memory. The expression need not have a pointer value (though it may);
10309 it is always interpreted as an integer address of a byte of memory.
10310 @xref{Expressions, ,Expressions}, for more information on expressions. The default for
10311 @var{addr} is usually just after the last address examined---but several
10312 other commands also set the default address: @code{info breakpoints} (to
10313 the address of the last breakpoint listed), @code{info line} (to the
10314 starting address of a line), and @code{print} (if you use it to display
10315 a value from memory).
10318 For example, @samp{x/3uh 0x54320} is a request to display three halfwords
10319 (@code{h}) of memory, formatted as unsigned decimal integers (@samp{u}),
10320 starting at address @code{0x54320}. @samp{x/4xw $sp} prints the four
10321 words (@samp{w}) of memory above the stack pointer (here, @samp{$sp};
10322 @pxref{Registers, ,Registers}) in hexadecimal (@samp{x}).
10324 You can also specify a negative repeat count to examine memory backward
10325 from the given address. For example, @samp{x/-3uh 0x54320} prints three
10326 halfwords (@code{h}) at @code{0x54314}, @code{0x54328}, and @code{0x5431c}.
10328 Since the letters indicating unit sizes are all distinct from the
10329 letters specifying output formats, you do not have to remember whether
10330 unit size or format comes first; either order works. The output
10331 specifications @samp{4xw} and @samp{4wx} mean exactly the same thing.
10332 (However, the count @var{n} must come first; @samp{wx4} does not work.)
10334 Even though the unit size @var{u} is ignored for the formats @samp{s}
10335 and @samp{i}, you might still want to use a count @var{n}; for example,
10336 @samp{3i} specifies that you want to see three machine instructions,
10337 including any operands. For convenience, especially when used with
10338 the @code{display} command, the @samp{i} format also prints branch delay
10339 slot instructions, if any, beyond the count specified, which immediately
10340 follow the last instruction that is within the count. The command
10341 @code{disassemble} gives an alternative way of inspecting machine
10342 instructions; see @ref{Machine Code,,Source and Machine Code}.
10344 If a negative repeat count is specified for the formats @samp{s} or @samp{i},
10345 the command displays null-terminated strings or instructions before the given
10346 address as many as the absolute value of the given number. For the @samp{i}
10347 format, we use line number information in the debug info to accurately locate
10348 instruction boundaries while disassembling backward. If line info is not
10349 available, the command stops examining memory with an error message.
10351 All the defaults for the arguments to @code{x} are designed to make it
10352 easy to continue scanning memory with minimal specifications each time
10353 you use @code{x}. For example, after you have inspected three machine
10354 instructions with @samp{x/3i @var{addr}}, you can inspect the next seven
10355 with just @samp{x/7}. If you use @key{RET} to repeat the @code{x} command,
10356 the repeat count @var{n} is used again; the other arguments default as
10357 for successive uses of @code{x}.
10359 When examining machine instructions, the instruction at current program
10360 counter is shown with a @code{=>} marker. For example:
10363 (@value{GDBP}) x/5i $pc-6
10364 0x804837f <main+11>: mov %esp,%ebp
10365 0x8048381 <main+13>: push %ecx
10366 0x8048382 <main+14>: sub $0x4,%esp
10367 => 0x8048385 <main+17>: movl $0x8048460,(%esp)
10368 0x804838c <main+24>: call 0x80482d4 <puts@@plt>
10371 @cindex @code{$_}, @code{$__}, and value history
10372 The addresses and contents printed by the @code{x} command are not saved
10373 in the value history because there is often too much of them and they
10374 would get in the way. Instead, @value{GDBN} makes these values available for
10375 subsequent use in expressions as values of the convenience variables
10376 @code{$_} and @code{$__}. After an @code{x} command, the last address
10377 examined is available for use in expressions in the convenience variable
10378 @code{$_}. The contents of that address, as examined, are available in
10379 the convenience variable @code{$__}.
10381 If the @code{x} command has a repeat count, the address and contents saved
10382 are from the last memory unit printed; this is not the same as the last
10383 address printed if several units were printed on the last line of output.
10385 @anchor{addressable memory unit}
10386 @cindex addressable memory unit
10387 Most targets have an addressable memory unit size of 8 bits. This means
10388 that to each memory address are associated 8 bits of data. Some
10389 targets, however, have other addressable memory unit sizes.
10390 Within @value{GDBN} and this document, the term
10391 @dfn{addressable memory unit} (or @dfn{memory unit} for short) is used
10392 when explicitly referring to a chunk of data of that size. The word
10393 @dfn{byte} is used to refer to a chunk of data of 8 bits, regardless of
10394 the addressable memory unit size of the target. For most systems,
10395 addressable memory unit is a synonym of byte.
10397 @cindex remote memory comparison
10398 @cindex target memory comparison
10399 @cindex verify remote memory image
10400 @cindex verify target memory image
10401 When you are debugging a program running on a remote target machine
10402 (@pxref{Remote Debugging}), you may wish to verify the program's image
10403 in the remote machine's memory against the executable file you
10404 downloaded to the target. Or, on any target, you may want to check
10405 whether the program has corrupted its own read-only sections. The
10406 @code{compare-sections} command is provided for such situations.
10409 @kindex compare-sections
10410 @item compare-sections @r{[}@var{section-name}@r{|}@code{-r}@r{]}
10411 Compare the data of a loadable section @var{section-name} in the
10412 executable file of the program being debugged with the same section in
10413 the target machine's memory, and report any mismatches. With no
10414 arguments, compares all loadable sections. With an argument of
10415 @code{-r}, compares all loadable read-only sections.
10417 Note: for remote targets, this command can be accelerated if the
10418 target supports computing the CRC checksum of a block of memory
10419 (@pxref{qCRC packet}).
10423 @section Automatic Display
10424 @cindex automatic display
10425 @cindex display of expressions
10427 If you find that you want to print the value of an expression frequently
10428 (to see how it changes), you might want to add it to the @dfn{automatic
10429 display list} so that @value{GDBN} prints its value each time your program stops.
10430 Each expression added to the list is given a number to identify it;
10431 to remove an expression from the list, you specify that number.
10432 The automatic display looks like this:
10436 3: bar[5] = (struct hack *) 0x3804
10440 This display shows item numbers, expressions and their current values. As with
10441 displays you request manually using @code{x} or @code{print}, you can
10442 specify the output format you prefer; in fact, @code{display} decides
10443 whether to use @code{print} or @code{x} depending your format
10444 specification---it uses @code{x} if you specify either the @samp{i}
10445 or @samp{s} format, or a unit size; otherwise it uses @code{print}.
10449 @item display @var{expr}
10450 Add the expression @var{expr} to the list of expressions to display
10451 each time your program stops. @xref{Expressions, ,Expressions}.
10453 @code{display} does not repeat if you press @key{RET} again after using it.
10455 @item display/@var{fmt} @var{expr}
10456 For @var{fmt} specifying only a display format and not a size or
10457 count, add the expression @var{expr} to the auto-display list but
10458 arrange to display it each time in the specified format @var{fmt}.
10459 @xref{Output Formats,,Output Formats}.
10461 @item display/@var{fmt} @var{addr}
10462 For @var{fmt} @samp{i} or @samp{s}, or including a unit-size or a
10463 number of units, add the expression @var{addr} as a memory address to
10464 be examined each time your program stops. Examining means in effect
10465 doing @samp{x/@var{fmt} @var{addr}}. @xref{Memory, ,Examining Memory}.
10468 For example, @samp{display/i $pc} can be helpful, to see the machine
10469 instruction about to be executed each time execution stops (@samp{$pc}
10470 is a common name for the program counter; @pxref{Registers, ,Registers}).
10473 @kindex delete display
10475 @item undisplay @var{dnums}@dots{}
10476 @itemx delete display @var{dnums}@dots{}
10477 Remove items from the list of expressions to display. Specify the
10478 numbers of the displays that you want affected with the command
10479 argument @var{dnums}. It can be a single display number, one of the
10480 numbers shown in the first field of the @samp{info display} display;
10481 or it could be a range of display numbers, as in @code{2-4}.
10483 @code{undisplay} does not repeat if you press @key{RET} after using it.
10484 (Otherwise you would just get the error @samp{No display number @dots{}}.)
10486 @kindex disable display
10487 @item disable display @var{dnums}@dots{}
10488 Disable the display of item numbers @var{dnums}. A disabled display
10489 item is not printed automatically, but is not forgotten. It may be
10490 enabled again later. Specify the numbers of the displays that you
10491 want affected with the command argument @var{dnums}. It can be a
10492 single display number, one of the numbers shown in the first field of
10493 the @samp{info display} display; or it could be a range of display
10494 numbers, as in @code{2-4}.
10496 @kindex enable display
10497 @item enable display @var{dnums}@dots{}
10498 Enable display of item numbers @var{dnums}. It becomes effective once
10499 again in auto display of its expression, until you specify otherwise.
10500 Specify the numbers of the displays that you want affected with the
10501 command argument @var{dnums}. It can be a single display number, one
10502 of the numbers shown in the first field of the @samp{info display}
10503 display; or it could be a range of display numbers, as in @code{2-4}.
10506 Display the current values of the expressions on the list, just as is
10507 done when your program stops.
10509 @kindex info display
10511 Print the list of expressions previously set up to display
10512 automatically, each one with its item number, but without showing the
10513 values. This includes disabled expressions, which are marked as such.
10514 It also includes expressions which would not be displayed right now
10515 because they refer to automatic variables not currently available.
10518 @cindex display disabled out of scope
10519 If a display expression refers to local variables, then it does not make
10520 sense outside the lexical context for which it was set up. Such an
10521 expression is disabled when execution enters a context where one of its
10522 variables is not defined. For example, if you give the command
10523 @code{display last_char} while inside a function with an argument
10524 @code{last_char}, @value{GDBN} displays this argument while your program
10525 continues to stop inside that function. When it stops elsewhere---where
10526 there is no variable @code{last_char}---the display is disabled
10527 automatically. The next time your program stops where @code{last_char}
10528 is meaningful, you can enable the display expression once again.
10530 @node Print Settings
10531 @section Print Settings
10533 @cindex format options
10534 @cindex print settings
10535 @value{GDBN} provides the following ways to control how arrays, structures,
10536 and symbols are printed.
10539 These settings are useful for debugging programs in any language:
10543 @anchor{set print address}
10544 @item set print address
10545 @itemx set print address on
10546 @cindex print/don't print memory addresses
10547 @value{GDBN} prints memory addresses showing the location of stack
10548 traces, structure values, pointer values, breakpoints, and so forth,
10549 even when it also displays the contents of those addresses. The default
10550 is @code{on}. For example, this is what a stack frame display looks like with
10551 @code{set print address on}:
10556 #0 set_quotes (lq=0x34c78 "<<", rq=0x34c88 ">>")
10558 530 if (lquote != def_lquote)
10562 @item set print address off
10563 Do not print addresses when displaying their contents. For example,
10564 this is the same stack frame displayed with @code{set print address off}:
10568 (@value{GDBP}) set print addr off
10570 #0 set_quotes (lq="<<", rq=">>") at input.c:530
10571 530 if (lquote != def_lquote)
10575 You can use @samp{set print address off} to eliminate all machine
10576 dependent displays from the @value{GDBN} interface. For example, with
10577 @code{print address off}, you should get the same text for backtraces on
10578 all machines---whether or not they involve pointer arguments.
10581 @item show print address
10582 Show whether or not addresses are to be printed.
10585 When @value{GDBN} prints a symbolic address, it normally prints the
10586 closest earlier symbol plus an offset. If that symbol does not uniquely
10587 identify the address (for example, it is a name whose scope is a single
10588 source file), you may need to clarify. One way to do this is with
10589 @code{info line}, for example @samp{info line *0x4537}. Alternately,
10590 you can set @value{GDBN} to print the source file and line number when
10591 it prints a symbolic address:
10594 @item set print symbol-filename on
10595 @cindex source file and line of a symbol
10596 @cindex symbol, source file and line
10597 Tell @value{GDBN} to print the source file name and line number of a
10598 symbol in the symbolic form of an address.
10600 @item set print symbol-filename off
10601 Do not print source file name and line number of a symbol. This is the
10604 @item show print symbol-filename
10605 Show whether or not @value{GDBN} will print the source file name and
10606 line number of a symbol in the symbolic form of an address.
10609 Another situation where it is helpful to show symbol filenames and line
10610 numbers is when disassembling code; @value{GDBN} shows you the line
10611 number and source file that corresponds to each instruction.
10613 Also, you may wish to see the symbolic form only if the address being
10614 printed is reasonably close to the closest earlier symbol:
10617 @item set print max-symbolic-offset @var{max-offset}
10618 @itemx set print max-symbolic-offset unlimited
10619 @cindex maximum value for offset of closest symbol
10620 Tell @value{GDBN} to only display the symbolic form of an address if the
10621 offset between the closest earlier symbol and the address is less than
10622 @var{max-offset}. The default is @code{unlimited}, which tells @value{GDBN}
10623 to always print the symbolic form of an address if any symbol precedes
10624 it. Zero is equivalent to @code{unlimited}.
10626 @item show print max-symbolic-offset
10627 Ask how large the maximum offset is that @value{GDBN} prints in a
10631 @cindex wild pointer, interpreting
10632 @cindex pointer, finding referent
10633 If you have a pointer and you are not sure where it points, try
10634 @samp{set print symbol-filename on}. Then you can determine the name
10635 and source file location of the variable where it points, using
10636 @samp{p/a @var{pointer}}. This interprets the address in symbolic form.
10637 For example, here @value{GDBN} shows that a variable @code{ptt} points
10638 at another variable @code{t}, defined in @file{hi2.c}:
10641 (@value{GDBP}) set print symbol-filename on
10642 (@value{GDBP}) p/a ptt
10643 $4 = 0xe008 <t in hi2.c>
10647 @emph{Warning:} For pointers that point to a local variable, @samp{p/a}
10648 does not show the symbol name and filename of the referent, even with
10649 the appropriate @code{set print} options turned on.
10652 You can also enable @samp{/a}-like formatting all the time using
10653 @samp{set print symbol on}:
10655 @anchor{set print symbol}
10657 @item set print symbol on
10658 Tell @value{GDBN} to print the symbol corresponding to an address, if
10661 @item set print symbol off
10662 Tell @value{GDBN} not to print the symbol corresponding to an
10663 address. In this mode, @value{GDBN} will still print the symbol
10664 corresponding to pointers to functions. This is the default.
10666 @item show print symbol
10667 Show whether @value{GDBN} will display the symbol corresponding to an
10671 Other settings control how different kinds of objects are printed:
10674 @anchor{set print array}
10675 @item set print array
10676 @itemx set print array on
10677 @cindex pretty print arrays
10678 Pretty print arrays. This format is more convenient to read,
10679 but uses more space. The default is off.
10681 @item set print array off
10682 Return to compressed format for arrays.
10684 @item show print array
10685 Show whether compressed or pretty format is selected for displaying
10688 @cindex print array indexes
10689 @anchor{set print array-indexes}
10690 @item set print array-indexes
10691 @itemx set print array-indexes on
10692 Print the index of each element when displaying arrays. May be more
10693 convenient to locate a given element in the array or quickly find the
10694 index of a given element in that printed array. The default is off.
10696 @item set print array-indexes off
10697 Stop printing element indexes when displaying arrays.
10699 @item show print array-indexes
10700 Show whether the index of each element is printed when displaying
10703 @anchor{set print elements}
10704 @item set print elements @var{number-of-elements}
10705 @itemx set print elements unlimited
10706 @cindex number of array elements to print
10707 @cindex limit on number of printed array elements
10708 Set a limit on how many elements of an array @value{GDBN} will print.
10709 If @value{GDBN} is printing a large array, it stops printing after it has
10710 printed the number of elements set by the @code{set print elements} command.
10711 This limit also applies to the display of strings.
10712 When @value{GDBN} starts, this limit is set to 200.
10713 Setting @var{number-of-elements} to @code{unlimited} or zero means
10714 that the number of elements to print is unlimited.
10716 @item show print elements
10717 Display the number of elements of a large array that @value{GDBN} will print.
10718 If the number is 0, then the printing is unlimited.
10720 @anchor{set print frame-arguments}
10721 @item set print frame-arguments @var{value}
10722 @kindex set print frame-arguments
10723 @cindex printing frame argument values
10724 @cindex print all frame argument values
10725 @cindex print frame argument values for scalars only
10726 @cindex do not print frame argument values
10727 This command allows to control how the values of arguments are printed
10728 when the debugger prints a frame (@pxref{Frames}). The possible
10733 The values of all arguments are printed.
10736 Print the value of an argument only if it is a scalar. The value of more
10737 complex arguments such as arrays, structures, unions, etc, is replaced
10738 by @code{@dots{}}. This is the default. Here is an example where
10739 only scalar arguments are shown:
10742 #1 0x08048361 in call_me (i=3, s=@dots{}, ss=0xbf8d508c, u=@dots{}, e=green)
10747 None of the argument values are printed. Instead, the value of each argument
10748 is replaced by @code{@dots{}}. In this case, the example above now becomes:
10751 #1 0x08048361 in call_me (i=@dots{}, s=@dots{}, ss=@dots{}, u=@dots{}, e=@dots{})
10756 By default, only scalar arguments are printed. This command can be used
10757 to configure the debugger to print the value of all arguments, regardless
10758 of their type. However, it is often advantageous to not print the value
10759 of more complex parameters. For instance, it reduces the amount of
10760 information printed in each frame, making the backtrace more readable.
10761 Also, it improves performance when displaying Ada frames, because
10762 the computation of large arguments can sometimes be CPU-intensive,
10763 especially in large applications. Setting @code{print frame-arguments}
10764 to @code{scalars} (the default) or @code{none} avoids this computation,
10765 thus speeding up the display of each Ada frame.
10767 @item show print frame-arguments
10768 Show how the value of arguments should be displayed when printing a frame.
10770 @anchor{set print raw-frame-arguments}
10771 @item set print raw-frame-arguments on
10772 Print frame arguments in raw, non pretty-printed, form.
10774 @item set print raw-frame-arguments off
10775 Print frame arguments in pretty-printed form, if there is a pretty-printer
10776 for the value (@pxref{Pretty Printing}),
10777 otherwise print the value in raw form.
10778 This is the default.
10780 @item show print raw-frame-arguments
10781 Show whether to print frame arguments in raw form.
10783 @anchor{set print entry-values}
10784 @item set print entry-values @var{value}
10785 @kindex set print entry-values
10786 Set printing of frame argument values at function entry. In some cases
10787 @value{GDBN} can determine the value of function argument which was passed by
10788 the function caller, even if the value was modified inside the called function
10789 and therefore is different. With optimized code, the current value could be
10790 unavailable, but the entry value may still be known.
10792 The default value is @code{default} (see below for its description). Older
10793 @value{GDBN} behaved as with the setting @code{no}. Compilers not supporting
10794 this feature will behave in the @code{default} setting the same way as with the
10797 This functionality is currently supported only by DWARF 2 debugging format and
10798 the compiler has to produce @samp{DW_TAG_call_site} tags. With
10799 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
10802 The @var{value} parameter can be one of the following:
10806 Print only actual parameter values, never print values from function entry
10810 #0 different (val=6)
10811 #0 lost (val=<optimized out>)
10813 #0 invalid (val=<optimized out>)
10817 Print only parameter values from function entry point. The actual parameter
10818 values are never printed.
10820 #0 equal (val@@entry=5)
10821 #0 different (val@@entry=5)
10822 #0 lost (val@@entry=5)
10823 #0 born (val@@entry=<optimized out>)
10824 #0 invalid (val@@entry=<optimized out>)
10828 Print only parameter values from function entry point. If value from function
10829 entry point is not known while the actual value is known, print the actual
10830 value for such parameter.
10832 #0 equal (val@@entry=5)
10833 #0 different (val@@entry=5)
10834 #0 lost (val@@entry=5)
10836 #0 invalid (val@@entry=<optimized out>)
10840 Print actual parameter values. If actual parameter value is not known while
10841 value from function entry point is known, print the entry point value for such
10845 #0 different (val=6)
10846 #0 lost (val@@entry=5)
10848 #0 invalid (val=<optimized out>)
10852 Always print both the actual parameter value and its value from function entry
10853 point, even if values of one or both are not available due to compiler
10856 #0 equal (val=5, val@@entry=5)
10857 #0 different (val=6, val@@entry=5)
10858 #0 lost (val=<optimized out>, val@@entry=5)
10859 #0 born (val=10, val@@entry=<optimized out>)
10860 #0 invalid (val=<optimized out>, val@@entry=<optimized out>)
10864 Print the actual parameter value if it is known and also its value from
10865 function entry point if it is known. If neither is known, print for the actual
10866 value @code{<optimized out>}. If not in MI mode (@pxref{GDB/MI}) and if both
10867 values are known and identical, print the shortened
10868 @code{param=param@@entry=VALUE} notation.
10870 #0 equal (val=val@@entry=5)
10871 #0 different (val=6, val@@entry=5)
10872 #0 lost (val@@entry=5)
10874 #0 invalid (val=<optimized out>)
10878 Always print the actual parameter value. Print also its value from function
10879 entry point, but only if it is known. If not in MI mode (@pxref{GDB/MI}) and
10880 if both values are known and identical, print the shortened
10881 @code{param=param@@entry=VALUE} notation.
10883 #0 equal (val=val@@entry=5)
10884 #0 different (val=6, val@@entry=5)
10885 #0 lost (val=<optimized out>, val@@entry=5)
10887 #0 invalid (val=<optimized out>)
10891 For analysis messages on possible failures of frame argument values at function
10892 entry resolution see @ref{set debug entry-values}.
10894 @item show print entry-values
10895 Show the method being used for printing of frame argument values at function
10898 @anchor{set print repeats}
10899 @item set print repeats @var{number-of-repeats}
10900 @itemx set print repeats unlimited
10901 @cindex repeated array elements
10902 Set the threshold for suppressing display of repeated array
10903 elements. When the number of consecutive identical elements of an
10904 array exceeds the threshold, @value{GDBN} prints the string
10905 @code{"<repeats @var{n} times>"}, where @var{n} is the number of
10906 identical repetitions, instead of displaying the identical elements
10907 themselves. Setting the threshold to @code{unlimited} or zero will
10908 cause all elements to be individually printed. The default threshold
10911 @item show print repeats
10912 Display the current threshold for printing repeated identical
10915 @anchor{set print max-depth}
10916 @item set print max-depth @var{depth}
10917 @item set print max-depth unlimited
10918 @cindex printing nested structures
10919 Set the threshold after which nested structures are replaced with
10920 ellipsis, this can make visualising deeply nested structures easier.
10922 For example, given this C code
10925 typedef struct s1 @{ int a; @} s1;
10926 typedef struct s2 @{ s1 b; @} s2;
10927 typedef struct s3 @{ s2 c; @} s3;
10928 typedef struct s4 @{ s3 d; @} s4;
10930 s4 var = @{ @{ @{ @{ 3 @} @} @} @};
10933 The following table shows how different values of @var{depth} will
10934 effect how @code{var} is printed by @value{GDBN}:
10936 @multitable @columnfractions .3 .7
10937 @headitem @var{depth} setting @tab Result of @samp{p var}
10939 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
10941 @tab @code{$1 = @{...@}}
10943 @tab @code{$1 = @{d = @{...@}@}}
10945 @tab @code{$1 = @{d = @{c = @{...@}@}@}}
10947 @tab @code{$1 = @{d = @{c = @{b = @{...@}@}@}@}}
10949 @tab @code{$1 = @{d = @{c = @{b = @{a = 3@}@}@}@}}
10952 To see the contents of structures that have been hidden the user can
10953 either increase the print max-depth, or they can print the elements of
10954 the structure that are visible, for example
10957 (gdb) set print max-depth 2
10959 $1 = @{d = @{c = @{...@}@}@}
10961 $2 = @{c = @{b = @{...@}@}@}
10963 $3 = @{b = @{a = 3@}@}
10966 The pattern used to replace nested structures varies based on
10967 language, for most languages @code{@{...@}} is used, but Fortran uses
10970 @item show print max-depth
10971 Display the current threshold after which nested structures are
10972 replaces with ellipsis.
10974 @anchor{set print null-stop}
10975 @item set print null-stop
10976 @cindex @sc{null} elements in arrays
10977 Cause @value{GDBN} to stop printing the characters of an array when the first
10978 @sc{null} is encountered. This is useful when large arrays actually
10979 contain only short strings.
10980 The default is off.
10982 @item show print null-stop
10983 Show whether @value{GDBN} stops printing an array on the first
10984 @sc{null} character.
10986 @anchor{set print pretty}
10987 @item set print pretty on
10988 @cindex print structures in indented form
10989 @cindex indentation in structure display
10990 Cause @value{GDBN} to print structures in an indented format with one member
10991 per line, like this:
11006 @item set print pretty off
11007 Cause @value{GDBN} to print structures in a compact format, like this:
11011 $1 = @{next = 0x0, flags = @{sweet = 1, sour = 1@}, \
11012 meat = 0x54 "Pork"@}
11017 This is the default format.
11019 @item show print pretty
11020 Show which format @value{GDBN} is using to print structures.
11022 @item set print sevenbit-strings on
11023 @cindex eight-bit characters in strings
11024 @cindex octal escapes in strings
11025 Print using only seven-bit characters; if this option is set,
11026 @value{GDBN} displays any eight-bit characters (in strings or
11027 character values) using the notation @code{\}@var{nnn}. This setting is
11028 best if you are working in English (@sc{ascii}) and you use the
11029 high-order bit of characters as a marker or ``meta'' bit.
11031 @item set print sevenbit-strings off
11032 Print full eight-bit characters. This allows the use of more
11033 international character sets, and is the default.
11035 @item show print sevenbit-strings
11036 Show whether or not @value{GDBN} is printing only seven-bit characters.
11038 @anchor{set print union}
11039 @item set print union on
11040 @cindex unions in structures, printing
11041 Tell @value{GDBN} to print unions which are contained in structures
11042 and other unions. This is the default setting.
11044 @item set print union off
11045 Tell @value{GDBN} not to print unions which are contained in
11046 structures and other unions. @value{GDBN} will print @code{"@{...@}"}
11049 @item show print union
11050 Ask @value{GDBN} whether or not it will print unions which are contained in
11051 structures and other unions.
11053 For example, given the declarations
11056 typedef enum @{Tree, Bug@} Species;
11057 typedef enum @{Big_tree, Acorn, Seedling@} Tree_forms;
11058 typedef enum @{Caterpillar, Cocoon, Butterfly@}
11069 struct thing foo = @{Tree, @{Acorn@}@};
11073 with @code{set print union on} in effect @samp{p foo} would print
11076 $1 = @{it = Tree, form = @{tree = Acorn, bug = Cocoon@}@}
11080 and with @code{set print union off} in effect it would print
11083 $1 = @{it = Tree, form = @{...@}@}
11087 @code{set print union} affects programs written in C-like languages
11093 These settings are of interest when debugging C@t{++} programs:
11096 @cindex demangling C@t{++} names
11097 @item set print demangle
11098 @itemx set print demangle on
11099 Print C@t{++} names in their source form rather than in the encoded
11100 (``mangled'') form passed to the assembler and linker for type-safe
11101 linkage. The default is on.
11103 @item show print demangle
11104 Show whether C@t{++} names are printed in mangled or demangled form.
11106 @item set print asm-demangle
11107 @itemx set print asm-demangle on
11108 Print C@t{++} names in their source form rather than their mangled form, even
11109 in assembler code printouts such as instruction disassemblies.
11110 The default is off.
11112 @item show print asm-demangle
11113 Show whether C@t{++} names in assembly listings are printed in mangled
11116 @cindex C@t{++} symbol decoding style
11117 @cindex symbol decoding style, C@t{++}
11118 @kindex set demangle-style
11119 @item set demangle-style @var{style}
11120 Choose among several encoding schemes used by different compilers to represent
11121 C@t{++} names. If you omit @var{style}, you will see a list of possible
11122 formats. The default value is @var{auto}, which lets @value{GDBN} choose a
11123 decoding style by inspecting your program.
11125 @item show demangle-style
11126 Display the encoding style currently in use for decoding C@t{++} symbols.
11128 @anchor{set print object}
11129 @item set print object
11130 @itemx set print object on
11131 @cindex derived type of an object, printing
11132 @cindex display derived types
11133 When displaying a pointer to an object, identify the @emph{actual}
11134 (derived) type of the object rather than the @emph{declared} type, using
11135 the virtual function table. Note that the virtual function table is
11136 required---this feature can only work for objects that have run-time
11137 type identification; a single virtual method in the object's declared
11138 type is sufficient. Note that this setting is also taken into account when
11139 working with variable objects via MI (@pxref{GDB/MI}).
11141 @item set print object off
11142 Display only the declared type of objects, without reference to the
11143 virtual function table. This is the default setting.
11145 @item show print object
11146 Show whether actual, or declared, object types are displayed.
11148 @anchor{set print static-members}
11149 @item set print static-members
11150 @itemx set print static-members on
11151 @cindex static members of C@t{++} objects
11152 Print static members when displaying a C@t{++} object. The default is on.
11154 @item set print static-members off
11155 Do not print static members when displaying a C@t{++} object.
11157 @item show print static-members
11158 Show whether C@t{++} static members are printed or not.
11160 @item set print pascal_static-members
11161 @itemx set print pascal_static-members on
11162 @cindex static members of Pascal objects
11163 @cindex Pascal objects, static members display
11164 Print static members when displaying a Pascal object. The default is on.
11166 @item set print pascal_static-members off
11167 Do not print static members when displaying a Pascal object.
11169 @item show print pascal_static-members
11170 Show whether Pascal static members are printed or not.
11172 @c These don't work with HP ANSI C++ yet.
11173 @anchor{set print vtbl}
11174 @item set print vtbl
11175 @itemx set print vtbl on
11176 @cindex pretty print C@t{++} virtual function tables
11177 @cindex virtual functions (C@t{++}) display
11178 @cindex VTBL display
11179 Pretty print C@t{++} virtual function tables. The default is off.
11180 (The @code{vtbl} commands do not work on programs compiled with the HP
11181 ANSI C@t{++} compiler (@code{aCC}).)
11183 @item set print vtbl off
11184 Do not pretty print C@t{++} virtual function tables.
11186 @item show print vtbl
11187 Show whether C@t{++} virtual function tables are pretty printed, or not.
11190 @node Pretty Printing
11191 @section Pretty Printing
11193 @value{GDBN} provides a mechanism to allow pretty-printing of values using
11194 Python code. It greatly simplifies the display of complex objects. This
11195 mechanism works for both MI and the CLI.
11198 * Pretty-Printer Introduction:: Introduction to pretty-printers
11199 * Pretty-Printer Example:: An example pretty-printer
11200 * Pretty-Printer Commands:: Pretty-printer commands
11203 @node Pretty-Printer Introduction
11204 @subsection Pretty-Printer Introduction
11206 When @value{GDBN} prints a value, it first sees if there is a pretty-printer
11207 registered for the value. If there is then @value{GDBN} invokes the
11208 pretty-printer to print the value. Otherwise the value is printed normally.
11210 Pretty-printers are normally named. This makes them easy to manage.
11211 The @samp{info pretty-printer} command will list all the installed
11212 pretty-printers with their names.
11213 If a pretty-printer can handle multiple data types, then its
11214 @dfn{subprinters} are the printers for the individual data types.
11215 Each such subprinter has its own name.
11216 The format of the name is @var{printer-name};@var{subprinter-name}.
11218 Pretty-printers are installed by @dfn{registering} them with @value{GDBN}.
11219 Typically they are automatically loaded and registered when the corresponding
11220 debug information is loaded, thus making them available without having to
11221 do anything special.
11223 There are three places where a pretty-printer can be registered.
11227 Pretty-printers registered globally are available when debugging
11231 Pretty-printers registered with a program space are available only
11232 when debugging that program.
11233 @xref{Progspaces In Python}, for more details on program spaces in Python.
11236 Pretty-printers registered with an objfile are loaded and unloaded
11237 with the corresponding objfile (e.g., shared library).
11238 @xref{Objfiles In Python}, for more details on objfiles in Python.
11241 @xref{Selecting Pretty-Printers}, for further information on how
11242 pretty-printers are selected,
11244 @xref{Writing a Pretty-Printer}, for implementing pretty printers
11247 @node Pretty-Printer Example
11248 @subsection Pretty-Printer Example
11250 Here is how a C@t{++} @code{std::string} looks without a pretty-printer:
11253 (@value{GDBP}) print s
11255 static npos = 4294967295,
11257 <std::allocator<char>> = @{
11258 <__gnu_cxx::new_allocator<char>> = @{
11259 <No data fields>@}, <No data fields>
11261 members of std::basic_string<char, std::char_traits<char>,
11262 std::allocator<char> >::_Alloc_hider:
11263 _M_p = 0x804a014 "abcd"
11268 With a pretty-printer for @code{std::string} only the contents are printed:
11271 (@value{GDBP}) print s
11275 @node Pretty-Printer Commands
11276 @subsection Pretty-Printer Commands
11277 @cindex pretty-printer commands
11280 @kindex info pretty-printer
11281 @item info pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11282 Print the list of installed pretty-printers.
11283 This includes disabled pretty-printers, which are marked as such.
11285 @var{object-regexp} is a regular expression matching the objects
11286 whose pretty-printers to list.
11287 Objects can be @code{global}, the program space's file
11288 (@pxref{Progspaces In Python}),
11289 and the object files within that program space (@pxref{Objfiles In Python}).
11290 @xref{Selecting Pretty-Printers}, for details on how @value{GDBN}
11291 looks up a printer from these three objects.
11293 @var{name-regexp} is a regular expression matching the name of the printers
11296 @kindex disable pretty-printer
11297 @item disable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11298 Disable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11299 A disabled pretty-printer is not forgotten, it may be enabled again later.
11301 @kindex enable pretty-printer
11302 @item enable pretty-printer [@var{object-regexp} [@var{name-regexp}]]
11303 Enable pretty-printers matching @var{object-regexp} and @var{name-regexp}.
11308 Suppose we have three pretty-printers installed: one from library1.so
11309 named @code{foo} that prints objects of type @code{foo}, and
11310 another from library2.so named @code{bar} that prints two types of objects,
11311 @code{bar1} and @code{bar2}.
11314 (gdb) info pretty-printer
11321 (gdb) info pretty-printer library2
11326 (gdb) disable pretty-printer library1
11328 2 of 3 printers enabled
11329 (gdb) info pretty-printer
11336 (gdb) disable pretty-printer library2 bar;bar1
11338 1 of 3 printers enabled
11339 (gdb) info pretty-printer library2
11346 (gdb) disable pretty-printer library2 bar
11348 0 of 3 printers enabled
11349 (gdb) info pretty-printer library2
11358 Note that for @code{bar} the entire printer can be disabled,
11359 as can each individual subprinter.
11361 @node Value History
11362 @section Value History
11364 @cindex value history
11365 @cindex history of values printed by @value{GDBN}
11366 Values printed by the @code{print} command are saved in the @value{GDBN}
11367 @dfn{value history}. This allows you to refer to them in other expressions.
11368 Values are kept until the symbol table is re-read or discarded
11369 (for example with the @code{file} or @code{symbol-file} commands).
11370 When the symbol table changes, the value history is discarded,
11371 since the values may contain pointers back to the types defined in the
11376 @cindex history number
11377 The values printed are given @dfn{history numbers} by which you can
11378 refer to them. These are successive integers starting with one.
11379 @code{print} shows you the history number assigned to a value by
11380 printing @samp{$@var{num} = } before the value; here @var{num} is the
11383 To refer to any previous value, use @samp{$} followed by the value's
11384 history number. The way @code{print} labels its output is designed to
11385 remind you of this. Just @code{$} refers to the most recent value in
11386 the history, and @code{$$} refers to the value before that.
11387 @code{$$@var{n}} refers to the @var{n}th value from the end; @code{$$2}
11388 is the value just prior to @code{$$}, @code{$$1} is equivalent to
11389 @code{$$}, and @code{$$0} is equivalent to @code{$}.
11391 For example, suppose you have just printed a pointer to a structure and
11392 want to see the contents of the structure. It suffices to type
11398 If you have a chain of structures where the component @code{next} points
11399 to the next one, you can print the contents of the next one with this:
11406 You can print successive links in the chain by repeating this
11407 command---which you can do by just typing @key{RET}.
11409 Note that the history records values, not expressions. If the value of
11410 @code{x} is 4 and you type these commands:
11418 then the value recorded in the value history by the @code{print} command
11419 remains 4 even though the value of @code{x} has changed.
11422 @kindex show values
11424 Print the last ten values in the value history, with their item numbers.
11425 This is like @samp{p@ $$9} repeated ten times, except that @code{show
11426 values} does not change the history.
11428 @item show values @var{n}
11429 Print ten history values centered on history item number @var{n}.
11431 @item show values +
11432 Print ten history values just after the values last printed. If no more
11433 values are available, @code{show values +} produces no display.
11436 Pressing @key{RET} to repeat @code{show values @var{n}} has exactly the
11437 same effect as @samp{show values +}.
11439 @node Convenience Vars
11440 @section Convenience Variables
11442 @cindex convenience variables
11443 @cindex user-defined variables
11444 @value{GDBN} provides @dfn{convenience variables} that you can use within
11445 @value{GDBN} to hold on to a value and refer to it later. These variables
11446 exist entirely within @value{GDBN}; they are not part of your program, and
11447 setting a convenience variable has no direct effect on further execution
11448 of your program. That is why you can use them freely.
11450 Convenience variables are prefixed with @samp{$}. Any name preceded by
11451 @samp{$} can be used for a convenience variable, unless it is one of
11452 the predefined machine-specific register names (@pxref{Registers, ,Registers}).
11453 (Value history references, in contrast, are @emph{numbers} preceded
11454 by @samp{$}. @xref{Value History, ,Value History}.)
11456 You can save a value in a convenience variable with an assignment
11457 expression, just as you would set a variable in your program.
11461 set $foo = *object_ptr
11465 would save in @code{$foo} the value contained in the object pointed to by
11468 Using a convenience variable for the first time creates it, but its
11469 value is @code{void} until you assign a new value. You can alter the
11470 value with another assignment at any time.
11472 Convenience variables have no fixed types. You can assign a convenience
11473 variable any type of value, including structures and arrays, even if
11474 that variable already has a value of a different type. The convenience
11475 variable, when used as an expression, has the type of its current value.
11478 @kindex show convenience
11479 @cindex show all user variables and functions
11480 @item show convenience
11481 Print a list of convenience variables used so far, and their values,
11482 as well as a list of the convenience functions.
11483 Abbreviated @code{show conv}.
11485 @kindex init-if-undefined
11486 @cindex convenience variables, initializing
11487 @item init-if-undefined $@var{variable} = @var{expression}
11488 Set a convenience variable if it has not already been set. This is useful
11489 for user-defined commands that keep some state. It is similar, in concept,
11490 to using local static variables with initializers in C (except that
11491 convenience variables are global). It can also be used to allow users to
11492 override default values used in a command script.
11494 If the variable is already defined then the expression is not evaluated so
11495 any side-effects do not occur.
11498 One of the ways to use a convenience variable is as a counter to be
11499 incremented or a pointer to be advanced. For example, to print
11500 a field from successive elements of an array of structures:
11504 print bar[$i++]->contents
11508 Repeat that command by typing @key{RET}.
11510 Some convenience variables are created automatically by @value{GDBN} and given
11511 values likely to be useful.
11514 @vindex $_@r{, convenience variable}
11516 The variable @code{$_} is automatically set by the @code{x} command to
11517 the last address examined (@pxref{Memory, ,Examining Memory}). Other
11518 commands which provide a default address for @code{x} to examine also
11519 set @code{$_} to that address; these commands include @code{info line}
11520 and @code{info breakpoint}. The type of @code{$_} is @code{void *}
11521 except when set by the @code{x} command, in which case it is a pointer
11522 to the type of @code{$__}.
11524 @vindex $__@r{, convenience variable}
11526 The variable @code{$__} is automatically set by the @code{x} command
11527 to the value found in the last address examined. Its type is chosen
11528 to match the format in which the data was printed.
11531 @vindex $_exitcode@r{, convenience variable}
11532 When the program being debugged terminates normally, @value{GDBN}
11533 automatically sets this variable to the exit code of the program, and
11534 resets @code{$_exitsignal} to @code{void}.
11537 @vindex $_exitsignal@r{, convenience variable}
11538 When the program being debugged dies due to an uncaught signal,
11539 @value{GDBN} automatically sets this variable to that signal's number,
11540 and resets @code{$_exitcode} to @code{void}.
11542 To distinguish between whether the program being debugged has exited
11543 (i.e., @code{$_exitcode} is not @code{void}) or signalled (i.e.,
11544 @code{$_exitsignal} is not @code{void}), the convenience function
11545 @code{$_isvoid} can be used (@pxref{Convenience Funs,, Convenience
11546 Functions}). For example, considering the following source code:
11549 #include <signal.h>
11552 main (int argc, char *argv[])
11559 A valid way of telling whether the program being debugged has exited
11560 or signalled would be:
11563 (@value{GDBP}) define has_exited_or_signalled
11564 Type commands for definition of ``has_exited_or_signalled''.
11565 End with a line saying just ``end''.
11566 >if $_isvoid ($_exitsignal)
11567 >echo The program has exited\n
11569 >echo The program has signalled\n
11575 Program terminated with signal SIGALRM, Alarm clock.
11576 The program no longer exists.
11577 (@value{GDBP}) has_exited_or_signalled
11578 The program has signalled
11581 As can be seen, @value{GDBN} correctly informs that the program being
11582 debugged has signalled, since it calls @code{raise} and raises a
11583 @code{SIGALRM} signal. If the program being debugged had not called
11584 @code{raise}, then @value{GDBN} would report a normal exit:
11587 (@value{GDBP}) has_exited_or_signalled
11588 The program has exited
11592 The variable @code{$_exception} is set to the exception object being
11593 thrown at an exception-related catchpoint. @xref{Set Catchpoints}.
11596 @itemx $_probe_arg0@dots{}$_probe_arg11
11597 Arguments to a static probe. @xref{Static Probe Points}.
11600 @vindex $_sdata@r{, inspect, convenience variable}
11601 The variable @code{$_sdata} contains extra collected static tracepoint
11602 data. @xref{Tracepoint Actions,,Tracepoint Action Lists}. Note that
11603 @code{$_sdata} could be empty, if not inspecting a trace buffer, or
11604 if extra static tracepoint data has not been collected.
11607 @vindex $_siginfo@r{, convenience variable}
11608 The variable @code{$_siginfo} contains extra signal information
11609 (@pxref{extra signal information}). Note that @code{$_siginfo}
11610 could be empty, if the application has not yet received any signals.
11611 For example, it will be empty before you execute the @code{run} command.
11614 @vindex $_tlb@r{, convenience variable}
11615 The variable @code{$_tlb} is automatically set when debugging
11616 applications running on MS-Windows in native mode or connected to
11617 gdbserver that supports the @code{qGetTIBAddr} request.
11618 @xref{General Query Packets}.
11619 This variable contains the address of the thread information block.
11622 The number of the current inferior. @xref{Inferiors and
11623 Programs, ,Debugging Multiple Inferiors and Programs}.
11626 The thread number of the current thread. @xref{thread numbers}.
11629 The global number of the current thread. @xref{global thread numbers}.
11633 @vindex $_gdb_major@r{, convenience variable}
11634 @vindex $_gdb_minor@r{, convenience variable}
11635 The major and minor version numbers of the running @value{GDBN}.
11636 Development snapshots and pretest versions have their minor version
11637 incremented by one; thus, @value{GDBN} pretest 9.11.90 will produce
11638 the value 12 for @code{$_gdb_minor}. These variables allow you to
11639 write scripts that work with different versions of @value{GDBN}
11640 without errors caused by features unavailable in some of those
11643 @item $_shell_exitcode
11644 @itemx $_shell_exitsignal
11645 @vindex $_shell_exitcode@r{, convenience variable}
11646 @vindex $_shell_exitsignal@r{, convenience variable}
11647 @cindex shell command, exit code
11648 @cindex shell command, exit signal
11649 @cindex exit status of shell commands
11650 @value{GDBN} commands such as @code{shell} and @code{|} are launching
11651 shell commands. When a launched command terminates, @value{GDBN}
11652 automatically maintains the variables @code{$_shell_exitcode}
11653 and @code{$_shell_exitsignal} according to the exit status of the last
11654 launched command. These variables are set and used similarly to
11655 the variables @code{$_exitcode} and @code{$_exitsignal}.
11659 @node Convenience Funs
11660 @section Convenience Functions
11662 @cindex convenience functions
11663 @value{GDBN} also supplies some @dfn{convenience functions}. These
11664 have a syntax similar to convenience variables. A convenience
11665 function can be used in an expression just like an ordinary function;
11666 however, a convenience function is implemented internally to
11669 These functions do not require @value{GDBN} to be configured with
11670 @code{Python} support, which means that they are always available.
11674 @item $_isvoid (@var{expr})
11675 @findex $_isvoid@r{, convenience function}
11676 Return one if the expression @var{expr} is @code{void}. Otherwise it
11679 A @code{void} expression is an expression where the type of the result
11680 is @code{void}. For example, you can examine a convenience variable
11681 (see @ref{Convenience Vars,, Convenience Variables}) to check whether
11685 (@value{GDBP}) print $_exitcode
11687 (@value{GDBP}) print $_isvoid ($_exitcode)
11690 Starting program: ./a.out
11691 [Inferior 1 (process 29572) exited normally]
11692 (@value{GDBP}) print $_exitcode
11694 (@value{GDBP}) print $_isvoid ($_exitcode)
11698 In the example above, we used @code{$_isvoid} to check whether
11699 @code{$_exitcode} is @code{void} before and after the execution of the
11700 program being debugged. Before the execution there is no exit code to
11701 be examined, therefore @code{$_exitcode} is @code{void}. After the
11702 execution the program being debugged returned zero, therefore
11703 @code{$_exitcode} is zero, which means that it is not @code{void}
11706 The @code{void} expression can also be a call of a function from the
11707 program being debugged. For example, given the following function:
11716 The result of calling it inside @value{GDBN} is @code{void}:
11719 (@value{GDBP}) print foo ()
11721 (@value{GDBP}) print $_isvoid (foo ())
11723 (@value{GDBP}) set $v = foo ()
11724 (@value{GDBP}) print $v
11726 (@value{GDBP}) print $_isvoid ($v)
11732 These functions require @value{GDBN} to be configured with
11733 @code{Python} support.
11737 @item $_memeq(@var{buf1}, @var{buf2}, @var{length})
11738 @findex $_memeq@r{, convenience function}
11739 Returns one if the @var{length} bytes at the addresses given by
11740 @var{buf1} and @var{buf2} are equal.
11741 Otherwise it returns zero.
11743 @item $_regex(@var{str}, @var{regex})
11744 @findex $_regex@r{, convenience function}
11745 Returns one if the string @var{str} matches the regular expression
11746 @var{regex}. Otherwise it returns zero.
11747 The syntax of the regular expression is that specified by @code{Python}'s
11748 regular expression support.
11750 @item $_streq(@var{str1}, @var{str2})
11751 @findex $_streq@r{, convenience function}
11752 Returns one if the strings @var{str1} and @var{str2} are equal.
11753 Otherwise it returns zero.
11755 @item $_strlen(@var{str})
11756 @findex $_strlen@r{, convenience function}
11757 Returns the length of string @var{str}.
11759 @item $_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
11760 @findex $_caller_is@r{, convenience function}
11761 Returns one if the calling function's name is equal to @var{name}.
11762 Otherwise it returns zero.
11764 If the optional argument @var{number_of_frames} is provided,
11765 it is the number of frames up in the stack to look.
11773 at testsuite/gdb.python/py-caller-is.c:21
11774 #1 0x00000000004005a0 in middle_func ()
11775 at testsuite/gdb.python/py-caller-is.c:27
11776 #2 0x00000000004005ab in top_func ()
11777 at testsuite/gdb.python/py-caller-is.c:33
11778 #3 0x00000000004005b6 in main ()
11779 at testsuite/gdb.python/py-caller-is.c:39
11780 (gdb) print $_caller_is ("middle_func")
11782 (gdb) print $_caller_is ("top_func", 2)
11786 @item $_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
11787 @findex $_caller_matches@r{, convenience function}
11788 Returns one if the calling function's name matches the regular expression
11789 @var{regexp}. Otherwise it returns zero.
11791 If the optional argument @var{number_of_frames} is provided,
11792 it is the number of frames up in the stack to look.
11795 @item $_any_caller_is(@var{name}@r{[}, @var{number_of_frames}@r{]})
11796 @findex $_any_caller_is@r{, convenience function}
11797 Returns one if any calling function's name is equal to @var{name}.
11798 Otherwise it returns zero.
11800 If the optional argument @var{number_of_frames} is provided,
11801 it is the number of frames up in the stack to look.
11804 This function differs from @code{$_caller_is} in that this function
11805 checks all stack frames from the immediate caller to the frame specified
11806 by @var{number_of_frames}, whereas @code{$_caller_is} only checks the
11807 frame specified by @var{number_of_frames}.
11809 @item $_any_caller_matches(@var{regexp}@r{[}, @var{number_of_frames}@r{]})
11810 @findex $_any_caller_matches@r{, convenience function}
11811 Returns one if any calling function's name matches the regular expression
11812 @var{regexp}. Otherwise it returns zero.
11814 If the optional argument @var{number_of_frames} is provided,
11815 it is the number of frames up in the stack to look.
11818 This function differs from @code{$_caller_matches} in that this function
11819 checks all stack frames from the immediate caller to the frame specified
11820 by @var{number_of_frames}, whereas @code{$_caller_matches} only checks the
11821 frame specified by @var{number_of_frames}.
11823 @item $_as_string(@var{value})
11824 @findex $_as_string@r{, convenience function}
11825 Return the string representation of @var{value}.
11827 This function is useful to obtain the textual label (enumerator) of an
11828 enumeration value. For example, assuming the variable @var{node} is of
11829 an enumerated type:
11832 (gdb) printf "Visiting node of type %s\n", $_as_string(node)
11833 Visiting node of type NODE_INTEGER
11836 @item $_cimag(@var{value})
11837 @itemx $_creal(@var{value})
11838 @findex $_cimag@r{, convenience function}
11839 @findex $_creal@r{, convenience function}
11840 Return the imaginary (@code{$_cimag}) or real (@code{$_creal}) part of
11841 the complex number @var{value}.
11843 The type of the imaginary or real part depends on the type of the
11844 complex number, e.g., using @code{$_cimag} on a @code{float complex}
11845 will return an imaginary part of type @code{float}.
11849 @value{GDBN} provides the ability to list and get help on
11850 convenience functions.
11853 @item help function
11854 @kindex help function
11855 @cindex show all convenience functions
11856 Print a list of all convenience functions.
11863 You can refer to machine register contents, in expressions, as variables
11864 with names starting with @samp{$}. The names of registers are different
11865 for each machine; use @code{info registers} to see the names used on
11869 @kindex info registers
11870 @item info registers
11871 Print the names and values of all registers except floating-point
11872 and vector registers (in the selected stack frame).
11874 @kindex info all-registers
11875 @cindex floating point registers
11876 @item info all-registers
11877 Print the names and values of all registers, including floating-point
11878 and vector registers (in the selected stack frame).
11880 @item info registers @var{reggroup} @dots{}
11881 Print the name and value of the registers in each of the specified
11882 @var{reggroup}s. The @var{reggoup} can be any of those returned by
11883 @code{maint print reggroups} (@pxref{Maintenance Commands}).
11885 @item info registers @var{regname} @dots{}
11886 Print the @dfn{relativized} value of each specified register @var{regname}.
11887 As discussed in detail below, register values are normally relative to
11888 the selected stack frame. The @var{regname} may be any register name valid on
11889 the machine you are using, with or without the initial @samp{$}.
11892 @anchor{standard registers}
11893 @cindex stack pointer register
11894 @cindex program counter register
11895 @cindex process status register
11896 @cindex frame pointer register
11897 @cindex standard registers
11898 @value{GDBN} has four ``standard'' register names that are available (in
11899 expressions) on most machines---whenever they do not conflict with an
11900 architecture's canonical mnemonics for registers. The register names
11901 @code{$pc} and @code{$sp} are used for the program counter register and
11902 the stack pointer. @code{$fp} is used for a register that contains a
11903 pointer to the current stack frame, and @code{$ps} is used for a
11904 register that contains the processor status. For example,
11905 you could print the program counter in hex with
11912 or print the instruction to be executed next with
11919 or add four to the stack pointer@footnote{This is a way of removing
11920 one word from the stack, on machines where stacks grow downward in
11921 memory (most machines, nowadays). This assumes that the innermost
11922 stack frame is selected; setting @code{$sp} is not allowed when other
11923 stack frames are selected. To pop entire frames off the stack,
11924 regardless of machine architecture, use @code{return};
11925 see @ref{Returning, ,Returning from a Function}.} with
11931 Whenever possible, these four standard register names are available on
11932 your machine even though the machine has different canonical mnemonics,
11933 so long as there is no conflict. The @code{info registers} command
11934 shows the canonical names. For example, on the SPARC, @code{info
11935 registers} displays the processor status register as @code{$psr} but you
11936 can also refer to it as @code{$ps}; and on x86-based machines @code{$ps}
11937 is an alias for the @sc{eflags} register.
11939 @value{GDBN} always considers the contents of an ordinary register as an
11940 integer when the register is examined in this way. Some machines have
11941 special registers which can hold nothing but floating point; these
11942 registers are considered to have floating point values. There is no way
11943 to refer to the contents of an ordinary register as floating point value
11944 (although you can @emph{print} it as a floating point value with
11945 @samp{print/f $@var{regname}}).
11947 Some registers have distinct ``raw'' and ``virtual'' data formats. This
11948 means that the data format in which the register contents are saved by
11949 the operating system is not the same one that your program normally
11950 sees. For example, the registers of the 68881 floating point
11951 coprocessor are always saved in ``extended'' (raw) format, but all C
11952 programs expect to work with ``double'' (virtual) format. In such
11953 cases, @value{GDBN} normally works with the virtual format only (the format
11954 that makes sense for your program), but the @code{info registers} command
11955 prints the data in both formats.
11957 @cindex SSE registers (x86)
11958 @cindex MMX registers (x86)
11959 Some machines have special registers whose contents can be interpreted
11960 in several different ways. For example, modern x86-based machines
11961 have SSE and MMX registers that can hold several values packed
11962 together in several different formats. @value{GDBN} refers to such
11963 registers in @code{struct} notation:
11966 (@value{GDBP}) print $xmm1
11968 v4_float = @{0, 3.43859137e-038, 1.54142831e-044, 1.821688e-044@},
11969 v2_double = @{9.92129282474342e-303, 2.7585945287983262e-313@},
11970 v16_int8 = "\000\000\000\000\3706;\001\v\000\000\000\r\000\000",
11971 v8_int16 = @{0, 0, 14072, 315, 11, 0, 13, 0@},
11972 v4_int32 = @{0, 20657912, 11, 13@},
11973 v2_int64 = @{88725056443645952, 55834574859@},
11974 uint128 = 0x0000000d0000000b013b36f800000000
11979 To set values of such registers, you need to tell @value{GDBN} which
11980 view of the register you wish to change, as if you were assigning
11981 value to a @code{struct} member:
11984 (@value{GDBP}) set $xmm1.uint128 = 0x000000000000000000000000FFFFFFFF
11987 Normally, register values are relative to the selected stack frame
11988 (@pxref{Selection, ,Selecting a Frame}). This means that you get the
11989 value that the register would contain if all stack frames farther in
11990 were exited and their saved registers restored. In order to see the
11991 true contents of hardware registers, you must select the innermost
11992 frame (with @samp{frame 0}).
11994 @cindex caller-saved registers
11995 @cindex call-clobbered registers
11996 @cindex volatile registers
11997 @cindex <not saved> values
11998 Usually ABIs reserve some registers as not needed to be saved by the
11999 callee (a.k.a.: ``caller-saved'', ``call-clobbered'' or ``volatile''
12000 registers). It may therefore not be possible for @value{GDBN} to know
12001 the value a register had before the call (in other words, in the outer
12002 frame), if the register value has since been changed by the callee.
12003 @value{GDBN} tries to deduce where the inner frame saved
12004 (``callee-saved'') registers, from the debug info, unwind info, or the
12005 machine code generated by your compiler. If some register is not
12006 saved, and @value{GDBN} knows the register is ``caller-saved'' (via
12007 its own knowledge of the ABI, or because the debug/unwind info
12008 explicitly says the register's value is undefined), @value{GDBN}
12009 displays @w{@samp{<not saved>}} as the register's value. With targets
12010 that @value{GDBN} has no knowledge of the register saving convention,
12011 if a register was not saved by the callee, then its value and location
12012 in the outer frame are assumed to be the same of the inner frame.
12013 This is usually harmless, because if the register is call-clobbered,
12014 the caller either does not care what is in the register after the
12015 call, or has code to restore the value that it does care about. Note,
12016 however, that if you change such a register in the outer frame, you
12017 may also be affecting the inner frame. Also, the more ``outer'' the
12018 frame is you're looking at, the more likely a call-clobbered
12019 register's value is to be wrong, in the sense that it doesn't actually
12020 represent the value the register had just before the call.
12022 @node Floating Point Hardware
12023 @section Floating Point Hardware
12024 @cindex floating point
12026 Depending on the configuration, @value{GDBN} may be able to give
12027 you more information about the status of the floating point hardware.
12032 Display hardware-dependent information about the floating
12033 point unit. The exact contents and layout vary depending on the
12034 floating point chip. Currently, @samp{info float} is supported on
12035 the ARM and x86 machines.
12039 @section Vector Unit
12040 @cindex vector unit
12042 Depending on the configuration, @value{GDBN} may be able to give you
12043 more information about the status of the vector unit.
12046 @kindex info vector
12048 Display information about the vector unit. The exact contents and
12049 layout vary depending on the hardware.
12052 @node OS Information
12053 @section Operating System Auxiliary Information
12054 @cindex OS information
12056 @value{GDBN} provides interfaces to useful OS facilities that can help
12057 you debug your program.
12059 @cindex auxiliary vector
12060 @cindex vector, auxiliary
12061 Some operating systems supply an @dfn{auxiliary vector} to programs at
12062 startup. This is akin to the arguments and environment that you
12063 specify for a program, but contains a system-dependent variety of
12064 binary values that tell system libraries important details about the
12065 hardware, operating system, and process. Each value's purpose is
12066 identified by an integer tag; the meanings are well-known but system-specific.
12067 Depending on the configuration and operating system facilities,
12068 @value{GDBN} may be able to show you this information. For remote
12069 targets, this functionality may further depend on the remote stub's
12070 support of the @samp{qXfer:auxv:read} packet, see
12071 @ref{qXfer auxiliary vector read}.
12076 Display the auxiliary vector of the inferior, which can be either a
12077 live process or a core dump file. @value{GDBN} prints each tag value
12078 numerically, and also shows names and text descriptions for recognized
12079 tags. Some values in the vector are numbers, some bit masks, and some
12080 pointers to strings or other data. @value{GDBN} displays each value in the
12081 most appropriate form for a recognized tag, and in hexadecimal for
12082 an unrecognized tag.
12085 On some targets, @value{GDBN} can access operating system-specific
12086 information and show it to you. The types of information available
12087 will differ depending on the type of operating system running on the
12088 target. The mechanism used to fetch the data is described in
12089 @ref{Operating System Information}. For remote targets, this
12090 functionality depends on the remote stub's support of the
12091 @samp{qXfer:osdata:read} packet, see @ref{qXfer osdata read}.
12095 @item info os @var{infotype}
12097 Display OS information of the requested type.
12099 On @sc{gnu}/Linux, the following values of @var{infotype} are valid:
12101 @anchor{linux info os infotypes}
12103 @kindex info os cpus
12105 Display the list of all CPUs/cores. For each CPU/core, @value{GDBN} prints
12106 the available fields from /proc/cpuinfo. For each supported architecture
12107 different fields are available. Two common entries are processor which gives
12108 CPU number and bogomips; a system constant that is calculated during
12109 kernel initialization.
12111 @kindex info os files
12113 Display the list of open file descriptors on the target. For each
12114 file descriptor, @value{GDBN} prints the identifier of the process
12115 owning the descriptor, the command of the owning process, the value
12116 of the descriptor, and the target of the descriptor.
12118 @kindex info os modules
12120 Display the list of all loaded kernel modules on the target. For each
12121 module, @value{GDBN} prints the module name, the size of the module in
12122 bytes, the number of times the module is used, the dependencies of the
12123 module, the status of the module, and the address of the loaded module
12126 @kindex info os msg
12128 Display the list of all System V message queues on the target. For each
12129 message queue, @value{GDBN} prints the message queue key, the message
12130 queue identifier, the access permissions, the current number of bytes
12131 on the queue, the current number of messages on the queue, the processes
12132 that last sent and received a message on the queue, the user and group
12133 of the owner and creator of the message queue, the times at which a
12134 message was last sent and received on the queue, and the time at which
12135 the message queue was last changed.
12137 @kindex info os processes
12139 Display the list of processes on the target. For each process,
12140 @value{GDBN} prints the process identifier, the name of the user, the
12141 command corresponding to the process, and the list of processor cores
12142 that the process is currently running on. (To understand what these
12143 properties mean, for this and the following info types, please consult
12144 the general @sc{gnu}/Linux documentation.)
12146 @kindex info os procgroups
12148 Display the list of process groups on the target. For each process,
12149 @value{GDBN} prints the identifier of the process group that it belongs
12150 to, the command corresponding to the process group leader, the process
12151 identifier, and the command line of the process. The list is sorted
12152 first by the process group identifier, then by the process identifier,
12153 so that processes belonging to the same process group are grouped together
12154 and the process group leader is listed first.
12156 @kindex info os semaphores
12158 Display the list of all System V semaphore sets on the target. For each
12159 semaphore set, @value{GDBN} prints the semaphore set key, the semaphore
12160 set identifier, the access permissions, the number of semaphores in the
12161 set, the user and group of the owner and creator of the semaphore set,
12162 and the times at which the semaphore set was operated upon and changed.
12164 @kindex info os shm
12166 Display the list of all System V shared-memory regions on the target.
12167 For each shared-memory region, @value{GDBN} prints the region key,
12168 the shared-memory identifier, the access permissions, the size of the
12169 region, the process that created the region, the process that last
12170 attached to or detached from the region, the current number of live
12171 attaches to the region, and the times at which the region was last
12172 attached to, detach from, and changed.
12174 @kindex info os sockets
12176 Display the list of Internet-domain sockets on the target. For each
12177 socket, @value{GDBN} prints the address and port of the local and
12178 remote endpoints, the current state of the connection, the creator of
12179 the socket, the IP address family of the socket, and the type of the
12182 @kindex info os threads
12184 Display the list of threads running on the target. For each thread,
12185 @value{GDBN} prints the identifier of the process that the thread
12186 belongs to, the command of the process, the thread identifier, and the
12187 processor core that it is currently running on. The main thread of a
12188 process is not listed.
12192 If @var{infotype} is omitted, then list the possible values for
12193 @var{infotype} and the kind of OS information available for each
12194 @var{infotype}. If the target does not return a list of possible
12195 types, this command will report an error.
12198 @node Memory Region Attributes
12199 @section Memory Region Attributes
12200 @cindex memory region attributes
12202 @dfn{Memory region attributes} allow you to describe special handling
12203 required by regions of your target's memory. @value{GDBN} uses
12204 attributes to determine whether to allow certain types of memory
12205 accesses; whether to use specific width accesses; and whether to cache
12206 target memory. By default the description of memory regions is
12207 fetched from the target (if the current target supports this), but the
12208 user can override the fetched regions.
12210 Defined memory regions can be individually enabled and disabled. When a
12211 memory region is disabled, @value{GDBN} uses the default attributes when
12212 accessing memory in that region. Similarly, if no memory regions have
12213 been defined, @value{GDBN} uses the default attributes when accessing
12216 When a memory region is defined, it is given a number to identify it;
12217 to enable, disable, or remove a memory region, you specify that number.
12221 @item mem @var{lower} @var{upper} @var{attributes}@dots{}
12222 Define a memory region bounded by @var{lower} and @var{upper} with
12223 attributes @var{attributes}@dots{}, and add it to the list of regions
12224 monitored by @value{GDBN}. Note that @var{upper} == 0 is a special
12225 case: it is treated as the target's maximum memory address.
12226 (0xffff on 16 bit targets, 0xffffffff on 32 bit targets, etc.)
12229 Discard any user changes to the memory regions and use target-supplied
12230 regions, if available, or no regions if the target does not support.
12233 @item delete mem @var{nums}@dots{}
12234 Remove memory regions @var{nums}@dots{} from the list of regions
12235 monitored by @value{GDBN}.
12237 @kindex disable mem
12238 @item disable mem @var{nums}@dots{}
12239 Disable monitoring of memory regions @var{nums}@dots{}.
12240 A disabled memory region is not forgotten.
12241 It may be enabled again later.
12244 @item enable mem @var{nums}@dots{}
12245 Enable monitoring of memory regions @var{nums}@dots{}.
12249 Print a table of all defined memory regions, with the following columns
12253 @item Memory Region Number
12254 @item Enabled or Disabled.
12255 Enabled memory regions are marked with @samp{y}.
12256 Disabled memory regions are marked with @samp{n}.
12259 The address defining the inclusive lower bound of the memory region.
12262 The address defining the exclusive upper bound of the memory region.
12265 The list of attributes set for this memory region.
12270 @subsection Attributes
12272 @subsubsection Memory Access Mode
12273 The access mode attributes set whether @value{GDBN} may make read or
12274 write accesses to a memory region.
12276 While these attributes prevent @value{GDBN} from performing invalid
12277 memory accesses, they do nothing to prevent the target system, I/O DMA,
12278 etc.@: from accessing memory.
12282 Memory is read only.
12284 Memory is write only.
12286 Memory is read/write. This is the default.
12289 @subsubsection Memory Access Size
12290 The access size attribute tells @value{GDBN} to use specific sized
12291 accesses in the memory region. Often memory mapped device registers
12292 require specific sized accesses. If no access size attribute is
12293 specified, @value{GDBN} may use accesses of any size.
12297 Use 8 bit memory accesses.
12299 Use 16 bit memory accesses.
12301 Use 32 bit memory accesses.
12303 Use 64 bit memory accesses.
12306 @c @subsubsection Hardware/Software Breakpoints
12307 @c The hardware/software breakpoint attributes set whether @value{GDBN}
12308 @c will use hardware or software breakpoints for the internal breakpoints
12309 @c used by the step, next, finish, until, etc. commands.
12313 @c Always use hardware breakpoints
12314 @c @item swbreak (default)
12317 @subsubsection Data Cache
12318 The data cache attributes set whether @value{GDBN} will cache target
12319 memory. While this generally improves performance by reducing debug
12320 protocol overhead, it can lead to incorrect results because @value{GDBN}
12321 does not know about volatile variables or memory mapped device
12326 Enable @value{GDBN} to cache target memory.
12328 Disable @value{GDBN} from caching target memory. This is the default.
12331 @subsection Memory Access Checking
12332 @value{GDBN} can be instructed to refuse accesses to memory that is
12333 not explicitly described. This can be useful if accessing such
12334 regions has undesired effects for a specific target, or to provide
12335 better error checking. The following commands control this behaviour.
12338 @kindex set mem inaccessible-by-default
12339 @item set mem inaccessible-by-default [on|off]
12340 If @code{on} is specified, make @value{GDBN} treat memory not
12341 explicitly described by the memory ranges as non-existent and refuse accesses
12342 to such memory. The checks are only performed if there's at least one
12343 memory range defined. If @code{off} is specified, make @value{GDBN}
12344 treat the memory not explicitly described by the memory ranges as RAM.
12345 The default value is @code{on}.
12346 @kindex show mem inaccessible-by-default
12347 @item show mem inaccessible-by-default
12348 Show the current handling of accesses to unknown memory.
12352 @c @subsubsection Memory Write Verification
12353 @c The memory write verification attributes set whether @value{GDBN}
12354 @c will re-reads data after each write to verify the write was successful.
12358 @c @item noverify (default)
12361 @node Dump/Restore Files
12362 @section Copy Between Memory and a File
12363 @cindex dump/restore files
12364 @cindex append data to a file
12365 @cindex dump data to a file
12366 @cindex restore data from a file
12368 You can use the commands @code{dump}, @code{append}, and
12369 @code{restore} to copy data between target memory and a file. The
12370 @code{dump} and @code{append} commands write data to a file, and the
12371 @code{restore} command reads data from a file back into the inferior's
12372 memory. Files may be in binary, Motorola S-record, Intel hex,
12373 Tektronix Hex, or Verilog Hex format; however, @value{GDBN} can only
12374 append to binary files, and cannot read from Verilog Hex files.
12379 @item dump @r{[}@var{format}@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12380 @itemx dump @r{[}@var{format}@r{]} value @var{filename} @var{expr}
12381 Dump the contents of memory from @var{start_addr} to @var{end_addr},
12382 or the value of @var{expr}, to @var{filename} in the given format.
12384 The @var{format} parameter may be any one of:
12391 Motorola S-record format.
12393 Tektronix Hex format.
12395 Verilog Hex format.
12398 @value{GDBN} uses the same definitions of these formats as the
12399 @sc{gnu} binary utilities, like @samp{objdump} and @samp{objcopy}. If
12400 @var{format} is omitted, @value{GDBN} dumps the data in raw binary
12404 @item append @r{[}binary@r{]} memory @var{filename} @var{start_addr} @var{end_addr}
12405 @itemx append @r{[}binary@r{]} value @var{filename} @var{expr}
12406 Append the contents of memory from @var{start_addr} to @var{end_addr},
12407 or the value of @var{expr}, to the file @var{filename}, in raw binary form.
12408 (@value{GDBN} can only append data to files in raw binary form.)
12411 @item restore @var{filename} @r{[}binary@r{]} @var{bias} @var{start} @var{end}
12412 Restore the contents of file @var{filename} into memory. The
12413 @code{restore} command can automatically recognize any known @sc{bfd}
12414 file format, except for raw binary. To restore a raw binary file you
12415 must specify the optional keyword @code{binary} after the filename.
12417 If @var{bias} is non-zero, its value will be added to the addresses
12418 contained in the file. Binary files always start at address zero, so
12419 they will be restored at address @var{bias}. Other bfd files have
12420 a built-in location; they will be restored at offset @var{bias}
12421 from that location.
12423 If @var{start} and/or @var{end} are non-zero, then only data between
12424 file offset @var{start} and file offset @var{end} will be restored.
12425 These offsets are relative to the addresses in the file, before
12426 the @var{bias} argument is applied.
12430 @node Core File Generation
12431 @section How to Produce a Core File from Your Program
12432 @cindex dump core from inferior
12434 A @dfn{core file} or @dfn{core dump} is a file that records the memory
12435 image of a running process and its process status (register values
12436 etc.). Its primary use is post-mortem debugging of a program that
12437 crashed while it ran outside a debugger. A program that crashes
12438 automatically produces a core file, unless this feature is disabled by
12439 the user. @xref{Files}, for information on invoking @value{GDBN} in
12440 the post-mortem debugging mode.
12442 Occasionally, you may wish to produce a core file of the program you
12443 are debugging in order to preserve a snapshot of its state.
12444 @value{GDBN} has a special command for that.
12448 @kindex generate-core-file
12449 @item generate-core-file [@var{file}]
12450 @itemx gcore [@var{file}]
12451 Produce a core dump of the inferior process. The optional argument
12452 @var{file} specifies the file name where to put the core dump. If not
12453 specified, the file name defaults to @file{core.@var{pid}}, where
12454 @var{pid} is the inferior process ID.
12456 Note that this command is implemented only for some systems (as of
12457 this writing, @sc{gnu}/Linux, FreeBSD, Solaris, and S390).
12459 On @sc{gnu}/Linux, this command can take into account the value of the
12460 file @file{/proc/@var{pid}/coredump_filter} when generating the core
12461 dump (@pxref{set use-coredump-filter}), and by default honors the
12462 @code{VM_DONTDUMP} flag for mappings where it is present in the file
12463 @file{/proc/@var{pid}/smaps} (@pxref{set dump-excluded-mappings}).
12465 @kindex set use-coredump-filter
12466 @anchor{set use-coredump-filter}
12467 @item set use-coredump-filter on
12468 @itemx set use-coredump-filter off
12469 Enable or disable the use of the file
12470 @file{/proc/@var{pid}/coredump_filter} when generating core dump
12471 files. This file is used by the Linux kernel to decide what types of
12472 memory mappings will be dumped or ignored when generating a core dump
12473 file. @var{pid} is the process ID of a currently running process.
12475 To make use of this feature, you have to write in the
12476 @file{/proc/@var{pid}/coredump_filter} file a value, in hexadecimal,
12477 which is a bit mask representing the memory mapping types. If a bit
12478 is set in the bit mask, then the memory mappings of the corresponding
12479 types will be dumped; otherwise, they will be ignored. This
12480 configuration is inherited by child processes. For more information
12481 about the bits that can be set in the
12482 @file{/proc/@var{pid}/coredump_filter} file, please refer to the
12483 manpage of @code{core(5)}.
12485 By default, this option is @code{on}. If this option is turned
12486 @code{off}, @value{GDBN} does not read the @file{coredump_filter} file
12487 and instead uses the same default value as the Linux kernel in order
12488 to decide which pages will be dumped in the core dump file. This
12489 value is currently @code{0x33}, which means that bits @code{0}
12490 (anonymous private mappings), @code{1} (anonymous shared mappings),
12491 @code{4} (ELF headers) and @code{5} (private huge pages) are active.
12492 This will cause these memory mappings to be dumped automatically.
12494 @kindex set dump-excluded-mappings
12495 @anchor{set dump-excluded-mappings}
12496 @item set dump-excluded-mappings on
12497 @itemx set dump-excluded-mappings off
12498 If @code{on} is specified, @value{GDBN} will dump memory mappings
12499 marked with the @code{VM_DONTDUMP} flag. This flag is represented in
12500 the file @file{/proc/@var{pid}/smaps} with the acronym @code{dd}.
12502 The default value is @code{off}.
12505 @node Character Sets
12506 @section Character Sets
12507 @cindex character sets
12509 @cindex translating between character sets
12510 @cindex host character set
12511 @cindex target character set
12513 If the program you are debugging uses a different character set to
12514 represent characters and strings than the one @value{GDBN} uses itself,
12515 @value{GDBN} can automatically translate between the character sets for
12516 you. The character set @value{GDBN} uses we call the @dfn{host
12517 character set}; the one the inferior program uses we call the
12518 @dfn{target character set}.
12520 For example, if you are running @value{GDBN} on a @sc{gnu}/Linux system, which
12521 uses the ISO Latin 1 character set, but you are using @value{GDBN}'s
12522 remote protocol (@pxref{Remote Debugging}) to debug a program
12523 running on an IBM mainframe, which uses the @sc{ebcdic} character set,
12524 then the host character set is Latin-1, and the target character set is
12525 @sc{ebcdic}. If you give @value{GDBN} the command @code{set
12526 target-charset EBCDIC-US}, then @value{GDBN} translates between
12527 @sc{ebcdic} and Latin 1 as you print character or string values, or use
12528 character and string literals in expressions.
12530 @value{GDBN} has no way to automatically recognize which character set
12531 the inferior program uses; you must tell it, using the @code{set
12532 target-charset} command, described below.
12534 Here are the commands for controlling @value{GDBN}'s character set
12538 @item set target-charset @var{charset}
12539 @kindex set target-charset
12540 Set the current target character set to @var{charset}. To display the
12541 list of supported target character sets, type
12542 @kbd{@w{set target-charset @key{TAB}@key{TAB}}}.
12544 @item set host-charset @var{charset}
12545 @kindex set host-charset
12546 Set the current host character set to @var{charset}.
12548 By default, @value{GDBN} uses a host character set appropriate to the
12549 system it is running on; you can override that default using the
12550 @code{set host-charset} command. On some systems, @value{GDBN} cannot
12551 automatically determine the appropriate host character set. In this
12552 case, @value{GDBN} uses @samp{UTF-8}.
12554 @value{GDBN} can only use certain character sets as its host character
12555 set. If you type @kbd{@w{set host-charset @key{TAB}@key{TAB}}},
12556 @value{GDBN} will list the host character sets it supports.
12558 @item set charset @var{charset}
12559 @kindex set charset
12560 Set the current host and target character sets to @var{charset}. As
12561 above, if you type @kbd{@w{set charset @key{TAB}@key{TAB}}},
12562 @value{GDBN} will list the names of the character sets that can be used
12563 for both host and target.
12566 @kindex show charset
12567 Show the names of the current host and target character sets.
12569 @item show host-charset
12570 @kindex show host-charset
12571 Show the name of the current host character set.
12573 @item show target-charset
12574 @kindex show target-charset
12575 Show the name of the current target character set.
12577 @item set target-wide-charset @var{charset}
12578 @kindex set target-wide-charset
12579 Set the current target's wide character set to @var{charset}. This is
12580 the character set used by the target's @code{wchar_t} type. To
12581 display the list of supported wide character sets, type
12582 @kbd{@w{set target-wide-charset @key{TAB}@key{TAB}}}.
12584 @item show target-wide-charset
12585 @kindex show target-wide-charset
12586 Show the name of the current target's wide character set.
12589 Here is an example of @value{GDBN}'s character set support in action.
12590 Assume that the following source code has been placed in the file
12591 @file{charset-test.c}:
12597 = @{72, 101, 108, 108, 111, 44, 32, 119,
12598 111, 114, 108, 100, 33, 10, 0@};
12599 char ibm1047_hello[]
12600 = @{200, 133, 147, 147, 150, 107, 64, 166,
12601 150, 153, 147, 132, 90, 37, 0@};
12605 printf ("Hello, world!\n");
12609 In this program, @code{ascii_hello} and @code{ibm1047_hello} are arrays
12610 containing the string @samp{Hello, world!} followed by a newline,
12611 encoded in the @sc{ascii} and @sc{ibm1047} character sets.
12613 We compile the program, and invoke the debugger on it:
12616 $ gcc -g charset-test.c -o charset-test
12617 $ gdb -nw charset-test
12618 GNU gdb 2001-12-19-cvs
12619 Copyright 2001 Free Software Foundation, Inc.
12624 We can use the @code{show charset} command to see what character sets
12625 @value{GDBN} is currently using to interpret and display characters and
12629 (@value{GDBP}) show charset
12630 The current host and target character set is `ISO-8859-1'.
12634 For the sake of printing this manual, let's use @sc{ascii} as our
12635 initial character set:
12637 (@value{GDBP}) set charset ASCII
12638 (@value{GDBP}) show charset
12639 The current host and target character set is `ASCII'.
12643 Let's assume that @sc{ascii} is indeed the correct character set for our
12644 host system --- in other words, let's assume that if @value{GDBN} prints
12645 characters using the @sc{ascii} character set, our terminal will display
12646 them properly. Since our current target character set is also
12647 @sc{ascii}, the contents of @code{ascii_hello} print legibly:
12650 (@value{GDBP}) print ascii_hello
12651 $1 = 0x401698 "Hello, world!\n"
12652 (@value{GDBP}) print ascii_hello[0]
12657 @value{GDBN} uses the target character set for character and string
12658 literals you use in expressions:
12661 (@value{GDBP}) print '+'
12666 The @sc{ascii} character set uses the number 43 to encode the @samp{+}
12669 @value{GDBN} relies on the user to tell it which character set the
12670 target program uses. If we print @code{ibm1047_hello} while our target
12671 character set is still @sc{ascii}, we get jibberish:
12674 (@value{GDBP}) print ibm1047_hello
12675 $4 = 0x4016a8 "\310\205\223\223\226k@@\246\226\231\223\204Z%"
12676 (@value{GDBP}) print ibm1047_hello[0]
12681 If we invoke the @code{set target-charset} followed by @key{TAB}@key{TAB},
12682 @value{GDBN} tells us the character sets it supports:
12685 (@value{GDBP}) set target-charset
12686 ASCII EBCDIC-US IBM1047 ISO-8859-1
12687 (@value{GDBP}) set target-charset
12690 We can select @sc{ibm1047} as our target character set, and examine the
12691 program's strings again. Now the @sc{ascii} string is wrong, but
12692 @value{GDBN} translates the contents of @code{ibm1047_hello} from the
12693 target character set, @sc{ibm1047}, to the host character set,
12694 @sc{ascii}, and they display correctly:
12697 (@value{GDBP}) set target-charset IBM1047
12698 (@value{GDBP}) show charset
12699 The current host character set is `ASCII'.
12700 The current target character set is `IBM1047'.
12701 (@value{GDBP}) print ascii_hello
12702 $6 = 0x401698 "\110\145%%?\054\040\167?\162%\144\041\012"
12703 (@value{GDBP}) print ascii_hello[0]
12705 (@value{GDBP}) print ibm1047_hello
12706 $8 = 0x4016a8 "Hello, world!\n"
12707 (@value{GDBP}) print ibm1047_hello[0]
12712 As above, @value{GDBN} uses the target character set for character and
12713 string literals you use in expressions:
12716 (@value{GDBP}) print '+'
12721 The @sc{ibm1047} character set uses the number 78 to encode the @samp{+}
12724 @node Caching Target Data
12725 @section Caching Data of Targets
12726 @cindex caching data of targets
12728 @value{GDBN} caches data exchanged between the debugger and a target.
12729 Each cache is associated with the address space of the inferior.
12730 @xref{Inferiors and Programs}, about inferior and address space.
12731 Such caching generally improves performance in remote debugging
12732 (@pxref{Remote Debugging}), because it reduces the overhead of the
12733 remote protocol by bundling memory reads and writes into large chunks.
12734 Unfortunately, simply caching everything would lead to incorrect results,
12735 since @value{GDBN} does not necessarily know anything about volatile
12736 values, memory-mapped I/O addresses, etc. Furthermore, in non-stop mode
12737 (@pxref{Non-Stop Mode}) memory can be changed @emph{while} a gdb command
12739 Therefore, by default, @value{GDBN} only caches data
12740 known to be on the stack@footnote{In non-stop mode, it is moderately
12741 rare for a running thread to modify the stack of a stopped thread
12742 in a way that would interfere with a backtrace, and caching of
12743 stack reads provides a significant speed up of remote backtraces.} or
12744 in the code segment.
12745 Other regions of memory can be explicitly marked as
12746 cacheable; @pxref{Memory Region Attributes}.
12749 @kindex set remotecache
12750 @item set remotecache on
12751 @itemx set remotecache off
12752 This option no longer does anything; it exists for compatibility
12755 @kindex show remotecache
12756 @item show remotecache
12757 Show the current state of the obsolete remotecache flag.
12759 @kindex set stack-cache
12760 @item set stack-cache on
12761 @itemx set stack-cache off
12762 Enable or disable caching of stack accesses. When @code{on}, use
12763 caching. By default, this option is @code{on}.
12765 @kindex show stack-cache
12766 @item show stack-cache
12767 Show the current state of data caching for memory accesses.
12769 @kindex set code-cache
12770 @item set code-cache on
12771 @itemx set code-cache off
12772 Enable or disable caching of code segment accesses. When @code{on},
12773 use caching. By default, this option is @code{on}. This improves
12774 performance of disassembly in remote debugging.
12776 @kindex show code-cache
12777 @item show code-cache
12778 Show the current state of target memory cache for code segment
12781 @kindex info dcache
12782 @item info dcache @r{[}line@r{]}
12783 Print the information about the performance of data cache of the
12784 current inferior's address space. The information displayed
12785 includes the dcache width and depth, and for each cache line, its
12786 number, address, and how many times it was referenced. This
12787 command is useful for debugging the data cache operation.
12789 If a line number is specified, the contents of that line will be
12792 @item set dcache size @var{size}
12793 @cindex dcache size
12794 @kindex set dcache size
12795 Set maximum number of entries in dcache (dcache depth above).
12797 @item set dcache line-size @var{line-size}
12798 @cindex dcache line-size
12799 @kindex set dcache line-size
12800 Set number of bytes each dcache entry caches (dcache width above).
12801 Must be a power of 2.
12803 @item show dcache size
12804 @kindex show dcache size
12805 Show maximum number of dcache entries. @xref{Caching Target Data, info dcache}.
12807 @item show dcache line-size
12808 @kindex show dcache line-size
12809 Show default size of dcache lines.
12813 @node Searching Memory
12814 @section Search Memory
12815 @cindex searching memory
12817 Memory can be searched for a particular sequence of bytes with the
12818 @code{find} command.
12822 @item find @r{[}/@var{sn}@r{]} @var{start_addr}, +@var{len}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
12823 @itemx find @r{[}/@var{sn}@r{]} @var{start_addr}, @var{end_addr}, @var{val1} @r{[}, @var{val2}, @dots{}@r{]}
12824 Search memory for the sequence of bytes specified by @var{val1}, @var{val2},
12825 etc. The search begins at address @var{start_addr} and continues for either
12826 @var{len} bytes or through to @var{end_addr} inclusive.
12829 @var{s} and @var{n} are optional parameters.
12830 They may be specified in either order, apart or together.
12833 @item @var{s}, search query size
12834 The size of each search query value.
12840 halfwords (two bytes)
12844 giant words (eight bytes)
12847 All values are interpreted in the current language.
12848 This means, for example, that if the current source language is C/C@t{++}
12849 then searching for the string ``hello'' includes the trailing '\0'.
12850 The null terminator can be removed from searching by using casts,
12851 e.g.: @samp{@{char[5]@}"hello"}.
12853 If the value size is not specified, it is taken from the
12854 value's type in the current language.
12855 This is useful when one wants to specify the search
12856 pattern as a mixture of types.
12857 Note that this means, for example, that in the case of C-like languages
12858 a search for an untyped 0x42 will search for @samp{(int) 0x42}
12859 which is typically four bytes.
12861 @item @var{n}, maximum number of finds
12862 The maximum number of matches to print. The default is to print all finds.
12865 You can use strings as search values. Quote them with double-quotes
12867 The string value is copied into the search pattern byte by byte,
12868 regardless of the endianness of the target and the size specification.
12870 The address of each match found is printed as well as a count of the
12871 number of matches found.
12873 The address of the last value found is stored in convenience variable
12875 A count of the number of matches is stored in @samp{$numfound}.
12877 For example, if stopped at the @code{printf} in this function:
12883 static char hello[] = "hello-hello";
12884 static struct @{ char c; short s; int i; @}
12885 __attribute__ ((packed)) mixed
12886 = @{ 'c', 0x1234, 0x87654321 @};
12887 printf ("%s\n", hello);
12892 you get during debugging:
12895 (gdb) find &hello[0], +sizeof(hello), "hello"
12896 0x804956d <hello.1620+6>
12898 (gdb) find &hello[0], +sizeof(hello), 'h', 'e', 'l', 'l', 'o'
12899 0x8049567 <hello.1620>
12900 0x804956d <hello.1620+6>
12902 (gdb) find &hello[0], +sizeof(hello), @{char[5]@}"hello"
12903 0x8049567 <hello.1620>
12904 0x804956d <hello.1620+6>
12906 (gdb) find /b1 &hello[0], +sizeof(hello), 'h', 0x65, 'l'
12907 0x8049567 <hello.1620>
12909 (gdb) find &mixed, +sizeof(mixed), (char) 'c', (short) 0x1234, (int) 0x87654321
12910 0x8049560 <mixed.1625>
12912 (gdb) print $numfound
12915 $2 = (void *) 0x8049560
12919 @section Value Sizes
12921 Whenever @value{GDBN} prints a value memory will be allocated within
12922 @value{GDBN} to hold the contents of the value. It is possible in
12923 some languages with dynamic typing systems, that an invalid program
12924 may indicate a value that is incorrectly large, this in turn may cause
12925 @value{GDBN} to try and allocate an overly large ammount of memory.
12928 @kindex set max-value-size
12929 @item set max-value-size @var{bytes}
12930 @itemx set max-value-size unlimited
12931 Set the maximum size of memory that @value{GDBN} will allocate for the
12932 contents of a value to @var{bytes}, trying to display a value that
12933 requires more memory than that will result in an error.
12935 Setting this variable does not effect values that have already been
12936 allocated within @value{GDBN}, only future allocations.
12938 There's a minimum size that @code{max-value-size} can be set to in
12939 order that @value{GDBN} can still operate correctly, this minimum is
12940 currently 16 bytes.
12942 The limit applies to the results of some subexpressions as well as to
12943 complete expressions. For example, an expression denoting a simple
12944 integer component, such as @code{x.y.z}, may fail if the size of
12945 @var{x.y} is dynamic and exceeds @var{bytes}. On the other hand,
12946 @value{GDBN} is sometimes clever; the expression @code{A[i]}, where
12947 @var{A} is an array variable with non-constant size, will generally
12948 succeed regardless of the bounds on @var{A}, as long as the component
12949 size is less than @var{bytes}.
12951 The default value of @code{max-value-size} is currently 64k.
12953 @kindex show max-value-size
12954 @item show max-value-size
12955 Show the maximum size of memory, in bytes, that @value{GDBN} will
12956 allocate for the contents of a value.
12959 @node Optimized Code
12960 @chapter Debugging Optimized Code
12961 @cindex optimized code, debugging
12962 @cindex debugging optimized code
12964 Almost all compilers support optimization. With optimization
12965 disabled, the compiler generates assembly code that corresponds
12966 directly to your source code, in a simplistic way. As the compiler
12967 applies more powerful optimizations, the generated assembly code
12968 diverges from your original source code. With help from debugging
12969 information generated by the compiler, @value{GDBN} can map from
12970 the running program back to constructs from your original source.
12972 @value{GDBN} is more accurate with optimization disabled. If you
12973 can recompile without optimization, it is easier to follow the
12974 progress of your program during debugging. But, there are many cases
12975 where you may need to debug an optimized version.
12977 When you debug a program compiled with @samp{-g -O}, remember that the
12978 optimizer has rearranged your code; the debugger shows you what is
12979 really there. Do not be too surprised when the execution path does not
12980 exactly match your source file! An extreme example: if you define a
12981 variable, but never use it, @value{GDBN} never sees that
12982 variable---because the compiler optimizes it out of existence.
12984 Some things do not work as well with @samp{-g -O} as with just
12985 @samp{-g}, particularly on machines with instruction scheduling. If in
12986 doubt, recompile with @samp{-g} alone, and if this fixes the problem,
12987 please report it to us as a bug (including a test case!).
12988 @xref{Variables}, for more information about debugging optimized code.
12991 * Inline Functions:: How @value{GDBN} presents inlining
12992 * Tail Call Frames:: @value{GDBN} analysis of jumps to functions
12995 @node Inline Functions
12996 @section Inline Functions
12997 @cindex inline functions, debugging
12999 @dfn{Inlining} is an optimization that inserts a copy of the function
13000 body directly at each call site, instead of jumping to a shared
13001 routine. @value{GDBN} displays inlined functions just like
13002 non-inlined functions. They appear in backtraces. You can view their
13003 arguments and local variables, step into them with @code{step}, skip
13004 them with @code{next}, and escape from them with @code{finish}.
13005 You can check whether a function was inlined by using the
13006 @code{info frame} command.
13008 For @value{GDBN} to support inlined functions, the compiler must
13009 record information about inlining in the debug information ---
13010 @value{NGCC} using the @sc{dwarf 2} format does this, and several
13011 other compilers do also. @value{GDBN} only supports inlined functions
13012 when using @sc{dwarf 2}. Versions of @value{NGCC} before 4.1
13013 do not emit two required attributes (@samp{DW_AT_call_file} and
13014 @samp{DW_AT_call_line}); @value{GDBN} does not display inlined
13015 function calls with earlier versions of @value{NGCC}. It instead
13016 displays the arguments and local variables of inlined functions as
13017 local variables in the caller.
13019 The body of an inlined function is directly included at its call site;
13020 unlike a non-inlined function, there are no instructions devoted to
13021 the call. @value{GDBN} still pretends that the call site and the
13022 start of the inlined function are different instructions. Stepping to
13023 the call site shows the call site, and then stepping again shows
13024 the first line of the inlined function, even though no additional
13025 instructions are executed.
13027 This makes source-level debugging much clearer; you can see both the
13028 context of the call and then the effect of the call. Only stepping by
13029 a single instruction using @code{stepi} or @code{nexti} does not do
13030 this; single instruction steps always show the inlined body.
13032 There are some ways that @value{GDBN} does not pretend that inlined
13033 function calls are the same as normal calls:
13037 Setting breakpoints at the call site of an inlined function may not
13038 work, because the call site does not contain any code. @value{GDBN}
13039 may incorrectly move the breakpoint to the next line of the enclosing
13040 function, after the call. This limitation will be removed in a future
13041 version of @value{GDBN}; until then, set a breakpoint on an earlier line
13042 or inside the inlined function instead.
13045 @value{GDBN} cannot locate the return value of inlined calls after
13046 using the @code{finish} command. This is a limitation of compiler-generated
13047 debugging information; after @code{finish}, you can step to the next line
13048 and print a variable where your program stored the return value.
13052 @node Tail Call Frames
13053 @section Tail Call Frames
13054 @cindex tail call frames, debugging
13056 Function @code{B} can call function @code{C} in its very last statement. In
13057 unoptimized compilation the call of @code{C} is immediately followed by return
13058 instruction at the end of @code{B} code. Optimizing compiler may replace the
13059 call and return in function @code{B} into one jump to function @code{C}
13060 instead. Such use of a jump instruction is called @dfn{tail call}.
13062 During execution of function @code{C}, there will be no indication in the
13063 function call stack frames that it was tail-called from @code{B}. If function
13064 @code{A} regularly calls function @code{B} which tail-calls function @code{C},
13065 then @value{GDBN} will see @code{A} as the caller of @code{C}. However, in
13066 some cases @value{GDBN} can determine that @code{C} was tail-called from
13067 @code{B}, and it will then create fictitious call frame for that, with the
13068 return address set up as if @code{B} called @code{C} normally.
13070 This functionality is currently supported only by DWARF 2 debugging format and
13071 the compiler has to produce @samp{DW_TAG_call_site} tags. With
13072 @value{NGCC}, you need to specify @option{-O -g} during compilation, to get
13075 @kbd{info frame} command (@pxref{Frame Info}) will indicate the tail call frame
13076 kind by text @code{tail call frame} such as in this sample @value{GDBN} output:
13080 0x40066b <b(int, double)+11>: jmp 0x400640 <c(int, double)>
13082 Stack level 1, frame at 0x7fffffffda30:
13083 rip = 0x40066d in b (amd64-entry-value.cc:59); saved rip 0x4004c5
13084 tail call frame, caller of frame at 0x7fffffffda30
13085 source language c++.
13086 Arglist at unknown address.
13087 Locals at unknown address, Previous frame's sp is 0x7fffffffda30
13090 The detection of all the possible code path executions can find them ambiguous.
13091 There is no execution history stored (possible @ref{Reverse Execution} is never
13092 used for this purpose) and the last known caller could have reached the known
13093 callee by multiple different jump sequences. In such case @value{GDBN} still
13094 tries to show at least all the unambiguous top tail callers and all the
13095 unambiguous bottom tail calees, if any.
13098 @anchor{set debug entry-values}
13099 @item set debug entry-values
13100 @kindex set debug entry-values
13101 When set to on, enables printing of analysis messages for both frame argument
13102 values at function entry and tail calls. It will show all the possible valid
13103 tail calls code paths it has considered. It will also print the intersection
13104 of them with the final unambiguous (possibly partial or even empty) code path
13107 @item show debug entry-values
13108 @kindex show debug entry-values
13109 Show the current state of analysis messages printing for both frame argument
13110 values at function entry and tail calls.
13113 The analysis messages for tail calls can for example show why the virtual tail
13114 call frame for function @code{c} has not been recognized (due to the indirect
13115 reference by variable @code{x}):
13118 static void __attribute__((noinline, noclone)) c (void);
13119 void (*x) (void) = c;
13120 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
13121 static void __attribute__((noinline, noclone)) c (void) @{ a (); @}
13122 int main (void) @{ x (); return 0; @}
13124 Breakpoint 1, DW_OP_entry_value resolving cannot find
13125 DW_TAG_call_site 0x40039a in main
13127 3 static void __attribute__((noinline, noclone)) a (void) @{ x++; @}
13130 #1 0x000000000040039a in main () at t.c:5
13133 Another possibility is an ambiguous virtual tail call frames resolution:
13137 static void __attribute__((noinline, noclone)) f (void) @{ i++; @}
13138 static void __attribute__((noinline, noclone)) e (void) @{ f (); @}
13139 static void __attribute__((noinline, noclone)) d (void) @{ f (); @}
13140 static void __attribute__((noinline, noclone)) c (void) @{ d (); @}
13141 static void __attribute__((noinline, noclone)) b (void)
13142 @{ if (i) c (); else e (); @}
13143 static void __attribute__((noinline, noclone)) a (void) @{ b (); @}
13144 int main (void) @{ a (); return 0; @}
13146 tailcall: initial: 0x4004d2(a) 0x4004ce(b) 0x4004b2(c) 0x4004a2(d)
13147 tailcall: compare: 0x4004d2(a) 0x4004cc(b) 0x400492(e)
13148 tailcall: reduced: 0x4004d2(a) |
13151 #1 0x00000000004004d2 in a () at t.c:8
13152 #2 0x0000000000400395 in main () at t.c:9
13155 @set CALLSEQ1A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}c@value{ARROW}d@value{ARROW}f}
13156 @set CALLSEQ2A @code{main@value{ARROW}a@value{ARROW}b@value{ARROW}e@value{ARROW}f}
13158 @c Convert CALLSEQ#A to CALLSEQ#B depending on HAVE_MAKEINFO_CLICK.
13159 @ifset HAVE_MAKEINFO_CLICK
13160 @set ARROW @click{}
13161 @set CALLSEQ1B @clicksequence{@value{CALLSEQ1A}}
13162 @set CALLSEQ2B @clicksequence{@value{CALLSEQ2A}}
13164 @ifclear HAVE_MAKEINFO_CLICK
13166 @set CALLSEQ1B @value{CALLSEQ1A}
13167 @set CALLSEQ2B @value{CALLSEQ2A}
13170 Frames #0 and #2 are real, #1 is a virtual tail call frame.
13171 The code can have possible execution paths @value{CALLSEQ1B} or
13172 @value{CALLSEQ2B}, @value{GDBN} cannot find which one from the inferior state.
13174 @code{initial:} state shows some random possible calling sequence @value{GDBN}
13175 has found. It then finds another possible calling sequcen - that one is
13176 prefixed by @code{compare:}. The non-ambiguous intersection of these two is
13177 printed as the @code{reduced:} calling sequence. That one could have many
13178 futher @code{compare:} and @code{reduced:} statements as long as there remain
13179 any non-ambiguous sequence entries.
13181 For the frame of function @code{b} in both cases there are different possible
13182 @code{$pc} values (@code{0x4004cc} or @code{0x4004ce}), therefore this frame is
13183 also ambigous. The only non-ambiguous frame is the one for function @code{a},
13184 therefore this one is displayed to the user while the ambiguous frames are
13187 There can be also reasons why printing of frame argument values at function
13192 static void __attribute__((noinline, noclone)) c (int i) @{ v++; @}
13193 static void __attribute__((noinline, noclone)) a (int i);
13194 static void __attribute__((noinline, noclone)) b (int i) @{ a (i); @}
13195 static void __attribute__((noinline, noclone)) a (int i)
13196 @{ if (i) b (i - 1); else c (0); @}
13197 int main (void) @{ a (5); return 0; @}
13200 #0 c (i=i@@entry=0) at t.c:2
13201 #1 0x0000000000400428 in a (DW_OP_entry_value resolving has found
13202 function "a" at 0x400420 can call itself via tail calls
13203 i=<optimized out>) at t.c:6
13204 #2 0x000000000040036e in main () at t.c:7
13207 @value{GDBN} cannot find out from the inferior state if and how many times did
13208 function @code{a} call itself (via function @code{b}) as these calls would be
13209 tail calls. Such tail calls would modify thue @code{i} variable, therefore
13210 @value{GDBN} cannot be sure the value it knows would be right - @value{GDBN}
13211 prints @code{<optimized out>} instead.
13214 @chapter C Preprocessor Macros
13216 Some languages, such as C and C@t{++}, provide a way to define and invoke
13217 ``preprocessor macros'' which expand into strings of tokens.
13218 @value{GDBN} can evaluate expressions containing macro invocations, show
13219 the result of macro expansion, and show a macro's definition, including
13220 where it was defined.
13222 You may need to compile your program specially to provide @value{GDBN}
13223 with information about preprocessor macros. Most compilers do not
13224 include macros in their debugging information, even when you compile
13225 with the @option{-g} flag. @xref{Compilation}.
13227 A program may define a macro at one point, remove that definition later,
13228 and then provide a different definition after that. Thus, at different
13229 points in the program, a macro may have different definitions, or have
13230 no definition at all. If there is a current stack frame, @value{GDBN}
13231 uses the macros in scope at that frame's source code line. Otherwise,
13232 @value{GDBN} uses the macros in scope at the current listing location;
13235 Whenever @value{GDBN} evaluates an expression, it always expands any
13236 macro invocations present in the expression. @value{GDBN} also provides
13237 the following commands for working with macros explicitly.
13241 @kindex macro expand
13242 @cindex macro expansion, showing the results of preprocessor
13243 @cindex preprocessor macro expansion, showing the results of
13244 @cindex expanding preprocessor macros
13245 @item macro expand @var{expression}
13246 @itemx macro exp @var{expression}
13247 Show the results of expanding all preprocessor macro invocations in
13248 @var{expression}. Since @value{GDBN} simply expands macros, but does
13249 not parse the result, @var{expression} need not be a valid expression;
13250 it can be any string of tokens.
13253 @item macro expand-once @var{expression}
13254 @itemx macro exp1 @var{expression}
13255 @cindex expand macro once
13256 @i{(This command is not yet implemented.)} Show the results of
13257 expanding those preprocessor macro invocations that appear explicitly in
13258 @var{expression}. Macro invocations appearing in that expansion are
13259 left unchanged. This command allows you to see the effect of a
13260 particular macro more clearly, without being confused by further
13261 expansions. Since @value{GDBN} simply expands macros, but does not
13262 parse the result, @var{expression} need not be a valid expression; it
13263 can be any string of tokens.
13266 @cindex macro definition, showing
13267 @cindex definition of a macro, showing
13268 @cindex macros, from debug info
13269 @item info macro [-a|-all] [--] @var{macro}
13270 Show the current definition or all definitions of the named @var{macro},
13271 and describe the source location or compiler command-line where that
13272 definition was established. The optional double dash is to signify the end of
13273 argument processing and the beginning of @var{macro} for non C-like macros where
13274 the macro may begin with a hyphen.
13276 @kindex info macros
13277 @item info macros @var{location}
13278 Show all macro definitions that are in effect at the location specified
13279 by @var{location}, and describe the source location or compiler
13280 command-line where those definitions were established.
13282 @kindex macro define
13283 @cindex user-defined macros
13284 @cindex defining macros interactively
13285 @cindex macros, user-defined
13286 @item macro define @var{macro} @var{replacement-list}
13287 @itemx macro define @var{macro}(@var{arglist}) @var{replacement-list}
13288 Introduce a definition for a preprocessor macro named @var{macro},
13289 invocations of which are replaced by the tokens given in
13290 @var{replacement-list}. The first form of this command defines an
13291 ``object-like'' macro, which takes no arguments; the second form
13292 defines a ``function-like'' macro, which takes the arguments given in
13295 A definition introduced by this command is in scope in every
13296 expression evaluated in @value{GDBN}, until it is removed with the
13297 @code{macro undef} command, described below. The definition overrides
13298 all definitions for @var{macro} present in the program being debugged,
13299 as well as any previous user-supplied definition.
13301 @kindex macro undef
13302 @item macro undef @var{macro}
13303 Remove any user-supplied definition for the macro named @var{macro}.
13304 This command only affects definitions provided with the @code{macro
13305 define} command, described above; it cannot remove definitions present
13306 in the program being debugged.
13310 List all the macros defined using the @code{macro define} command.
13313 @cindex macros, example of debugging with
13314 Here is a transcript showing the above commands in action. First, we
13315 show our source files:
13320 #include "sample.h"
13323 #define ADD(x) (M + x)
13328 printf ("Hello, world!\n");
13330 printf ("We're so creative.\n");
13332 printf ("Goodbye, world!\n");
13339 Now, we compile the program using the @sc{gnu} C compiler,
13340 @value{NGCC}. We pass the @option{-gdwarf-2}@footnote{This is the
13341 minimum. Recent versions of @value{NGCC} support @option{-gdwarf-3}
13342 and @option{-gdwarf-4}; we recommend always choosing the most recent
13343 version of DWARF.} @emph{and} @option{-g3} flags to ensure the compiler
13344 includes information about preprocessor macros in the debugging
13348 $ gcc -gdwarf-2 -g3 sample.c -o sample
13352 Now, we start @value{GDBN} on our sample program:
13356 GNU gdb 2002-05-06-cvs
13357 Copyright 2002 Free Software Foundation, Inc.
13358 GDB is free software, @dots{}
13362 We can expand macros and examine their definitions, even when the
13363 program is not running. @value{GDBN} uses the current listing position
13364 to decide which macro definitions are in scope:
13367 (@value{GDBP}) list main
13370 5 #define ADD(x) (M + x)
13375 10 printf ("Hello, world!\n");
13377 12 printf ("We're so creative.\n");
13378 (@value{GDBP}) info macro ADD
13379 Defined at /home/jimb/gdb/macros/play/sample.c:5
13380 #define ADD(x) (M + x)
13381 (@value{GDBP}) info macro Q
13382 Defined at /home/jimb/gdb/macros/play/sample.h:1
13383 included at /home/jimb/gdb/macros/play/sample.c:2
13385 (@value{GDBP}) macro expand ADD(1)
13386 expands to: (42 + 1)
13387 (@value{GDBP}) macro expand-once ADD(1)
13388 expands to: once (M + 1)
13392 In the example above, note that @code{macro expand-once} expands only
13393 the macro invocation explicit in the original text --- the invocation of
13394 @code{ADD} --- but does not expand the invocation of the macro @code{M},
13395 which was introduced by @code{ADD}.
13397 Once the program is running, @value{GDBN} uses the macro definitions in
13398 force at the source line of the current stack frame:
13401 (@value{GDBP}) break main
13402 Breakpoint 1 at 0x8048370: file sample.c, line 10.
13404 Starting program: /home/jimb/gdb/macros/play/sample
13406 Breakpoint 1, main () at sample.c:10
13407 10 printf ("Hello, world!\n");
13411 At line 10, the definition of the macro @code{N} at line 9 is in force:
13414 (@value{GDBP}) info macro N
13415 Defined at /home/jimb/gdb/macros/play/sample.c:9
13417 (@value{GDBP}) macro expand N Q M
13418 expands to: 28 < 42
13419 (@value{GDBP}) print N Q M
13424 As we step over directives that remove @code{N}'s definition, and then
13425 give it a new definition, @value{GDBN} finds the definition (or lack
13426 thereof) in force at each point:
13429 (@value{GDBP}) next
13431 12 printf ("We're so creative.\n");
13432 (@value{GDBP}) info macro N
13433 The symbol `N' has no definition as a C/C++ preprocessor macro
13434 at /home/jimb/gdb/macros/play/sample.c:12
13435 (@value{GDBP}) next
13437 14 printf ("Goodbye, world!\n");
13438 (@value{GDBP}) info macro N
13439 Defined at /home/jimb/gdb/macros/play/sample.c:13
13441 (@value{GDBP}) macro expand N Q M
13442 expands to: 1729 < 42
13443 (@value{GDBP}) print N Q M
13448 In addition to source files, macros can be defined on the compilation command
13449 line using the @option{-D@var{name}=@var{value}} syntax. For macros defined in
13450 such a way, @value{GDBN} displays the location of their definition as line zero
13451 of the source file submitted to the compiler.
13454 (@value{GDBP}) info macro __STDC__
13455 Defined at /home/jimb/gdb/macros/play/sample.c:0
13462 @chapter Tracepoints
13463 @c This chapter is based on the documentation written by Michael
13464 @c Snyder, David Taylor, Jim Blandy, and Elena Zannoni.
13466 @cindex tracepoints
13467 In some applications, it is not feasible for the debugger to interrupt
13468 the program's execution long enough for the developer to learn
13469 anything helpful about its behavior. If the program's correctness
13470 depends on its real-time behavior, delays introduced by a debugger
13471 might cause the program to change its behavior drastically, or perhaps
13472 fail, even when the code itself is correct. It is useful to be able
13473 to observe the program's behavior without interrupting it.
13475 Using @value{GDBN}'s @code{trace} and @code{collect} commands, you can
13476 specify locations in the program, called @dfn{tracepoints}, and
13477 arbitrary expressions to evaluate when those tracepoints are reached.
13478 Later, using the @code{tfind} command, you can examine the values
13479 those expressions had when the program hit the tracepoints. The
13480 expressions may also denote objects in memory---structures or arrays,
13481 for example---whose values @value{GDBN} should record; while visiting
13482 a particular tracepoint, you may inspect those objects as if they were
13483 in memory at that moment. However, because @value{GDBN} records these
13484 values without interacting with you, it can do so quickly and
13485 unobtrusively, hopefully not disturbing the program's behavior.
13487 The tracepoint facility is currently available only for remote
13488 targets. @xref{Targets}. In addition, your remote target must know
13489 how to collect trace data. This functionality is implemented in the
13490 remote stub; however, none of the stubs distributed with @value{GDBN}
13491 support tracepoints as of this writing. The format of the remote
13492 packets used to implement tracepoints are described in @ref{Tracepoint
13495 It is also possible to get trace data from a file, in a manner reminiscent
13496 of corefiles; you specify the filename, and use @code{tfind} to search
13497 through the file. @xref{Trace Files}, for more details.
13499 This chapter describes the tracepoint commands and features.
13502 * Set Tracepoints::
13503 * Analyze Collected Data::
13504 * Tracepoint Variables::
13508 @node Set Tracepoints
13509 @section Commands to Set Tracepoints
13511 Before running such a @dfn{trace experiment}, an arbitrary number of
13512 tracepoints can be set. A tracepoint is actually a special type of
13513 breakpoint (@pxref{Set Breaks}), so you can manipulate it using
13514 standard breakpoint commands. For instance, as with breakpoints,
13515 tracepoint numbers are successive integers starting from one, and many
13516 of the commands associated with tracepoints take the tracepoint number
13517 as their argument, to identify which tracepoint to work on.
13519 For each tracepoint, you can specify, in advance, some arbitrary set
13520 of data that you want the target to collect in the trace buffer when
13521 it hits that tracepoint. The collected data can include registers,
13522 local variables, or global data. Later, you can use @value{GDBN}
13523 commands to examine the values these data had at the time the
13524 tracepoint was hit.
13526 Tracepoints do not support every breakpoint feature. Ignore counts on
13527 tracepoints have no effect, and tracepoints cannot run @value{GDBN}
13528 commands when they are hit. Tracepoints may not be thread-specific
13531 @cindex fast tracepoints
13532 Some targets may support @dfn{fast tracepoints}, which are inserted in
13533 a different way (such as with a jump instead of a trap), that is
13534 faster but possibly restricted in where they may be installed.
13536 @cindex static tracepoints
13537 @cindex markers, static tracepoints
13538 @cindex probing markers, static tracepoints
13539 Regular and fast tracepoints are dynamic tracing facilities, meaning
13540 that they can be used to insert tracepoints at (almost) any location
13541 in the target. Some targets may also support controlling @dfn{static
13542 tracepoints} from @value{GDBN}. With static tracing, a set of
13543 instrumentation points, also known as @dfn{markers}, are embedded in
13544 the target program, and can be activated or deactivated by name or
13545 address. These are usually placed at locations which facilitate
13546 investigating what the target is actually doing. @value{GDBN}'s
13547 support for static tracing includes being able to list instrumentation
13548 points, and attach them with @value{GDBN} defined high level
13549 tracepoints that expose the whole range of convenience of
13550 @value{GDBN}'s tracepoints support. Namely, support for collecting
13551 registers values and values of global or local (to the instrumentation
13552 point) variables; tracepoint conditions and trace state variables.
13553 The act of installing a @value{GDBN} static tracepoint on an
13554 instrumentation point, or marker, is referred to as @dfn{probing} a
13555 static tracepoint marker.
13557 @code{gdbserver} supports tracepoints on some target systems.
13558 @xref{Server,,Tracepoints support in @code{gdbserver}}.
13560 This section describes commands to set tracepoints and associated
13561 conditions and actions.
13564 * Create and Delete Tracepoints::
13565 * Enable and Disable Tracepoints::
13566 * Tracepoint Passcounts::
13567 * Tracepoint Conditions::
13568 * Trace State Variables::
13569 * Tracepoint Actions::
13570 * Listing Tracepoints::
13571 * Listing Static Tracepoint Markers::
13572 * Starting and Stopping Trace Experiments::
13573 * Tracepoint Restrictions::
13576 @node Create and Delete Tracepoints
13577 @subsection Create and Delete Tracepoints
13580 @cindex set tracepoint
13582 @item trace @var{location}
13583 The @code{trace} command is very similar to the @code{break} command.
13584 Its argument @var{location} can be any valid location.
13585 @xref{Specify Location}. The @code{trace} command defines a tracepoint,
13586 which is a point in the target program where the debugger will briefly stop,
13587 collect some data, and then allow the program to continue. Setting a tracepoint
13588 or changing its actions takes effect immediately if the remote stub
13589 supports the @samp{InstallInTrace} feature (@pxref{install tracepoint
13591 If remote stub doesn't support the @samp{InstallInTrace} feature, all
13592 these changes don't take effect until the next @code{tstart}
13593 command, and once a trace experiment is running, further changes will
13594 not have any effect until the next trace experiment starts. In addition,
13595 @value{GDBN} supports @dfn{pending tracepoints}---tracepoints whose
13596 address is not yet resolved. (This is similar to pending breakpoints.)
13597 Pending tracepoints are not downloaded to the target and not installed
13598 until they are resolved. The resolution of pending tracepoints requires
13599 @value{GDBN} support---when debugging with the remote target, and
13600 @value{GDBN} disconnects from the remote stub (@pxref{disconnected
13601 tracing}), pending tracepoints can not be resolved (and downloaded to
13602 the remote stub) while @value{GDBN} is disconnected.
13604 Here are some examples of using the @code{trace} command:
13607 (@value{GDBP}) @b{trace foo.c:121} // a source file and line number
13609 (@value{GDBP}) @b{trace +2} // 2 lines forward
13611 (@value{GDBP}) @b{trace my_function} // first source line of function
13613 (@value{GDBP}) @b{trace *my_function} // EXACT start address of function
13615 (@value{GDBP}) @b{trace *0x2117c4} // an address
13619 You can abbreviate @code{trace} as @code{tr}.
13621 @item trace @var{location} if @var{cond}
13622 Set a tracepoint with condition @var{cond}; evaluate the expression
13623 @var{cond} each time the tracepoint is reached, and collect data only
13624 if the value is nonzero---that is, if @var{cond} evaluates as true.
13625 @xref{Tracepoint Conditions, ,Tracepoint Conditions}, for more
13626 information on tracepoint conditions.
13628 @item ftrace @var{location} [ if @var{cond} ]
13629 @cindex set fast tracepoint
13630 @cindex fast tracepoints, setting
13632 The @code{ftrace} command sets a fast tracepoint. For targets that
13633 support them, fast tracepoints will use a more efficient but possibly
13634 less general technique to trigger data collection, such as a jump
13635 instruction instead of a trap, or some sort of hardware support. It
13636 may not be possible to create a fast tracepoint at the desired
13637 location, in which case the command will exit with an explanatory
13640 @value{GDBN} handles arguments to @code{ftrace} exactly as for
13643 On 32-bit x86-architecture systems, fast tracepoints normally need to
13644 be placed at an instruction that is 5 bytes or longer, but can be
13645 placed at 4-byte instructions if the low 64K of memory of the target
13646 program is available to install trampolines. Some Unix-type systems,
13647 such as @sc{gnu}/Linux, exclude low addresses from the program's
13648 address space; but for instance with the Linux kernel it is possible
13649 to let @value{GDBN} use this area by doing a @command{sysctl} command
13650 to set the @code{mmap_min_addr} kernel parameter, as in
13653 sudo sysctl -w vm.mmap_min_addr=32768
13657 which sets the low address to 32K, which leaves plenty of room for
13658 trampolines. The minimum address should be set to a page boundary.
13660 @item strace @var{location} [ if @var{cond} ]
13661 @cindex set static tracepoint
13662 @cindex static tracepoints, setting
13663 @cindex probe static tracepoint marker
13665 The @code{strace} command sets a static tracepoint. For targets that
13666 support it, setting a static tracepoint probes a static
13667 instrumentation point, or marker, found at @var{location}. It may not
13668 be possible to set a static tracepoint at the desired location, in
13669 which case the command will exit with an explanatory message.
13671 @value{GDBN} handles arguments to @code{strace} exactly as for
13672 @code{trace}, with the addition that the user can also specify
13673 @code{-m @var{marker}} as @var{location}. This probes the marker
13674 identified by the @var{marker} string identifier. This identifier
13675 depends on the static tracepoint backend library your program is
13676 using. You can find all the marker identifiers in the @samp{ID} field
13677 of the @code{info static-tracepoint-markers} command output.
13678 @xref{Listing Static Tracepoint Markers,,Listing Static Tracepoint
13679 Markers}. For example, in the following small program using the UST
13685 trace_mark(ust, bar33, "str %s", "FOOBAZ");
13690 the marker id is composed of joining the first two arguments to the
13691 @code{trace_mark} call with a slash, which translates to:
13694 (@value{GDBP}) info static-tracepoint-markers
13695 Cnt Enb ID Address What
13696 1 n ust/bar33 0x0000000000400ddc in main at stexample.c:22
13702 so you may probe the marker above with:
13705 (@value{GDBP}) strace -m ust/bar33
13708 Static tracepoints accept an extra collect action --- @code{collect
13709 $_sdata}. This collects arbitrary user data passed in the probe point
13710 call to the tracing library. In the UST example above, you'll see
13711 that the third argument to @code{trace_mark} is a printf-like format
13712 string. The user data is then the result of running that formating
13713 string against the following arguments. Note that @code{info
13714 static-tracepoint-markers} command output lists that format string in
13715 the @samp{Data:} field.
13717 You can inspect this data when analyzing the trace buffer, by printing
13718 the $_sdata variable like any other variable available to
13719 @value{GDBN}. @xref{Tracepoint Actions,,Tracepoint Action Lists}.
13722 @cindex last tracepoint number
13723 @cindex recent tracepoint number
13724 @cindex tracepoint number
13725 The convenience variable @code{$tpnum} records the tracepoint number
13726 of the most recently set tracepoint.
13728 @kindex delete tracepoint
13729 @cindex tracepoint deletion
13730 @item delete tracepoint @r{[}@var{num}@r{]}
13731 Permanently delete one or more tracepoints. With no argument, the
13732 default is to delete all tracepoints. Note that the regular
13733 @code{delete} command can remove tracepoints also.
13738 (@value{GDBP}) @b{delete trace 1 2 3} // remove three tracepoints
13740 (@value{GDBP}) @b{delete trace} // remove all tracepoints
13744 You can abbreviate this command as @code{del tr}.
13747 @node Enable and Disable Tracepoints
13748 @subsection Enable and Disable Tracepoints
13750 These commands are deprecated; they are equivalent to plain @code{disable} and @code{enable}.
13753 @kindex disable tracepoint
13754 @item disable tracepoint @r{[}@var{num}@r{]}
13755 Disable tracepoint @var{num}, or all tracepoints if no argument
13756 @var{num} is given. A disabled tracepoint will have no effect during
13757 a trace experiment, but it is not forgotten. You can re-enable
13758 a disabled tracepoint using the @code{enable tracepoint} command.
13759 If the command is issued during a trace experiment and the debug target
13760 has support for disabling tracepoints during a trace experiment, then the
13761 change will be effective immediately. Otherwise, it will be applied to the
13762 next trace experiment.
13764 @kindex enable tracepoint
13765 @item enable tracepoint @r{[}@var{num}@r{]}
13766 Enable tracepoint @var{num}, or all tracepoints. If this command is
13767 issued during a trace experiment and the debug target supports enabling
13768 tracepoints during a trace experiment, then the enabled tracepoints will
13769 become effective immediately. Otherwise, they will become effective the
13770 next time a trace experiment is run.
13773 @node Tracepoint Passcounts
13774 @subsection Tracepoint Passcounts
13778 @cindex tracepoint pass count
13779 @item passcount @r{[}@var{n} @r{[}@var{num}@r{]]}
13780 Set the @dfn{passcount} of a tracepoint. The passcount is a way to
13781 automatically stop a trace experiment. If a tracepoint's passcount is
13782 @var{n}, then the trace experiment will be automatically stopped on
13783 the @var{n}'th time that tracepoint is hit. If the tracepoint number
13784 @var{num} is not specified, the @code{passcount} command sets the
13785 passcount of the most recently defined tracepoint. If no passcount is
13786 given, the trace experiment will run until stopped explicitly by the
13792 (@value{GDBP}) @b{passcount 5 2} // Stop on the 5th execution of
13793 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// tracepoint 2}
13795 (@value{GDBP}) @b{passcount 12} // Stop on the 12th execution of the
13796 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// most recently defined tracepoint.}
13797 (@value{GDBP}) @b{trace foo}
13798 (@value{GDBP}) @b{pass 3}
13799 (@value{GDBP}) @b{trace bar}
13800 (@value{GDBP}) @b{pass 2}
13801 (@value{GDBP}) @b{trace baz}
13802 (@value{GDBP}) @b{pass 1} // Stop tracing when foo has been
13803 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// executed 3 times OR when bar has}
13804 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// been executed 2 times}
13805 @exdent @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @ @code{// OR when baz has been executed 1 time.}
13809 @node Tracepoint Conditions
13810 @subsection Tracepoint Conditions
13811 @cindex conditional tracepoints
13812 @cindex tracepoint conditions
13814 The simplest sort of tracepoint collects data every time your program
13815 reaches a specified place. You can also specify a @dfn{condition} for
13816 a tracepoint. A condition is just a Boolean expression in your
13817 programming language (@pxref{Expressions, ,Expressions}). A
13818 tracepoint with a condition evaluates the expression each time your
13819 program reaches it, and data collection happens only if the condition
13822 Tracepoint conditions can be specified when a tracepoint is set, by
13823 using @samp{if} in the arguments to the @code{trace} command.
13824 @xref{Create and Delete Tracepoints, ,Setting Tracepoints}. They can
13825 also be set or changed at any time with the @code{condition} command,
13826 just as with breakpoints.
13828 Unlike breakpoint conditions, @value{GDBN} does not actually evaluate
13829 the conditional expression itself. Instead, @value{GDBN} encodes the
13830 expression into an agent expression (@pxref{Agent Expressions})
13831 suitable for execution on the target, independently of @value{GDBN}.
13832 Global variables become raw memory locations, locals become stack
13833 accesses, and so forth.
13835 For instance, suppose you have a function that is usually called
13836 frequently, but should not be called after an error has occurred. You
13837 could use the following tracepoint command to collect data about calls
13838 of that function that happen while the error code is propagating
13839 through the program; an unconditional tracepoint could end up
13840 collecting thousands of useless trace frames that you would have to
13844 (@value{GDBP}) @kbd{trace normal_operation if errcode > 0}
13847 @node Trace State Variables
13848 @subsection Trace State Variables
13849 @cindex trace state variables
13851 A @dfn{trace state variable} is a special type of variable that is
13852 created and managed by target-side code. The syntax is the same as
13853 that for GDB's convenience variables (a string prefixed with ``$''),
13854 but they are stored on the target. They must be created explicitly,
13855 using a @code{tvariable} command. They are always 64-bit signed
13858 Trace state variables are remembered by @value{GDBN}, and downloaded
13859 to the target along with tracepoint information when the trace
13860 experiment starts. There are no intrinsic limits on the number of
13861 trace state variables, beyond memory limitations of the target.
13863 @cindex convenience variables, and trace state variables
13864 Although trace state variables are managed by the target, you can use
13865 them in print commands and expressions as if they were convenience
13866 variables; @value{GDBN} will get the current value from the target
13867 while the trace experiment is running. Trace state variables share
13868 the same namespace as other ``$'' variables, which means that you
13869 cannot have trace state variables with names like @code{$23} or
13870 @code{$pc}, nor can you have a trace state variable and a convenience
13871 variable with the same name.
13875 @item tvariable $@var{name} [ = @var{expression} ]
13877 The @code{tvariable} command creates a new trace state variable named
13878 @code{$@var{name}}, and optionally gives it an initial value of
13879 @var{expression}. The @var{expression} is evaluated when this command is
13880 entered; the result will be converted to an integer if possible,
13881 otherwise @value{GDBN} will report an error. A subsequent
13882 @code{tvariable} command specifying the same name does not create a
13883 variable, but instead assigns the supplied initial value to the
13884 existing variable of that name, overwriting any previous initial
13885 value. The default initial value is 0.
13887 @item info tvariables
13888 @kindex info tvariables
13889 List all the trace state variables along with their initial values.
13890 Their current values may also be displayed, if the trace experiment is
13893 @item delete tvariable @r{[} $@var{name} @dots{} @r{]}
13894 @kindex delete tvariable
13895 Delete the given trace state variables, or all of them if no arguments
13900 @node Tracepoint Actions
13901 @subsection Tracepoint Action Lists
13905 @cindex tracepoint actions
13906 @item actions @r{[}@var{num}@r{]}
13907 This command will prompt for a list of actions to be taken when the
13908 tracepoint is hit. If the tracepoint number @var{num} is not
13909 specified, this command sets the actions for the one that was most
13910 recently defined (so that you can define a tracepoint and then say
13911 @code{actions} without bothering about its number). You specify the
13912 actions themselves on the following lines, one action at a time, and
13913 terminate the actions list with a line containing just @code{end}. So
13914 far, the only defined actions are @code{collect}, @code{teval}, and
13915 @code{while-stepping}.
13917 @code{actions} is actually equivalent to @code{commands} (@pxref{Break
13918 Commands, ,Breakpoint Command Lists}), except that only the defined
13919 actions are allowed; any other @value{GDBN} command is rejected.
13921 @cindex remove actions from a tracepoint
13922 To remove all actions from a tracepoint, type @samp{actions @var{num}}
13923 and follow it immediately with @samp{end}.
13926 (@value{GDBP}) @b{collect @var{data}} // collect some data
13928 (@value{GDBP}) @b{while-stepping 5} // single-step 5 times, collect data
13930 (@value{GDBP}) @b{end} // signals the end of actions.
13933 In the following example, the action list begins with @code{collect}
13934 commands indicating the things to be collected when the tracepoint is
13935 hit. Then, in order to single-step and collect additional data
13936 following the tracepoint, a @code{while-stepping} command is used,
13937 followed by the list of things to be collected after each step in a
13938 sequence of single steps. The @code{while-stepping} command is
13939 terminated by its own separate @code{end} command. Lastly, the action
13940 list is terminated by an @code{end} command.
13943 (@value{GDBP}) @b{trace foo}
13944 (@value{GDBP}) @b{actions}
13945 Enter actions for tracepoint 1, one per line:
13948 > while-stepping 12
13949 > collect $pc, arr[i]
13954 @kindex collect @r{(tracepoints)}
13955 @item collect@r{[}/@var{mods}@r{]} @var{expr1}, @var{expr2}, @dots{}
13956 Collect values of the given expressions when the tracepoint is hit.
13957 This command accepts a comma-separated list of any valid expressions.
13958 In addition to global, static, or local variables, the following
13959 special arguments are supported:
13963 Collect all registers.
13966 Collect all function arguments.
13969 Collect all local variables.
13972 Collect the return address. This is helpful if you want to see more
13975 @emph{Note:} The return address location can not always be reliably
13976 determined up front, and the wrong address / registers may end up
13977 collected instead. On some architectures the reliability is higher
13978 for tracepoints at function entry, while on others it's the opposite.
13979 When this happens, backtracing will stop because the return address is
13980 found unavailable (unless another collect rule happened to match it).
13983 Collects the number of arguments from the static probe at which the
13984 tracepoint is located.
13985 @xref{Static Probe Points}.
13987 @item $_probe_arg@var{n}
13988 @var{n} is an integer between 0 and 11. Collects the @var{n}th argument
13989 from the static probe at which the tracepoint is located.
13990 @xref{Static Probe Points}.
13993 @vindex $_sdata@r{, collect}
13994 Collect static tracepoint marker specific data. Only available for
13995 static tracepoints. @xref{Tracepoint Actions,,Tracepoint Action
13996 Lists}. On the UST static tracepoints library backend, an
13997 instrumentation point resembles a @code{printf} function call. The
13998 tracing library is able to collect user specified data formatted to a
13999 character string using the format provided by the programmer that
14000 instrumented the program. Other backends have similar mechanisms.
14001 Here's an example of a UST marker call:
14004 const char master_name[] = "$your_name";
14005 trace_mark(channel1, marker1, "hello %s", master_name)
14008 In this case, collecting @code{$_sdata} collects the string
14009 @samp{hello $yourname}. When analyzing the trace buffer, you can
14010 inspect @samp{$_sdata} like any other variable available to
14014 You can give several consecutive @code{collect} commands, each one
14015 with a single argument, or one @code{collect} command with several
14016 arguments separated by commas; the effect is the same.
14018 The optional @var{mods} changes the usual handling of the arguments.
14019 @code{s} requests that pointers to chars be handled as strings, in
14020 particular collecting the contents of the memory being pointed at, up
14021 to the first zero. The upper bound is by default the value of the
14022 @code{print elements} variable; if @code{s} is followed by a decimal
14023 number, that is the upper bound instead. So for instance
14024 @samp{collect/s25 mystr} collects as many as 25 characters at
14027 The command @code{info scope} (@pxref{Symbols, info scope}) is
14028 particularly useful for figuring out what data to collect.
14030 @kindex teval @r{(tracepoints)}
14031 @item teval @var{expr1}, @var{expr2}, @dots{}
14032 Evaluate the given expressions when the tracepoint is hit. This
14033 command accepts a comma-separated list of expressions. The results
14034 are discarded, so this is mainly useful for assigning values to trace
14035 state variables (@pxref{Trace State Variables}) without adding those
14036 values to the trace buffer, as would be the case if the @code{collect}
14039 @kindex while-stepping @r{(tracepoints)}
14040 @item while-stepping @var{n}
14041 Perform @var{n} single-step instruction traces after the tracepoint,
14042 collecting new data after each step. The @code{while-stepping}
14043 command is followed by the list of what to collect while stepping
14044 (followed by its own @code{end} command):
14047 > while-stepping 12
14048 > collect $regs, myglobal
14054 Note that @code{$pc} is not automatically collected by
14055 @code{while-stepping}; you need to explicitly collect that register if
14056 you need it. You may abbreviate @code{while-stepping} as @code{ws} or
14059 @item set default-collect @var{expr1}, @var{expr2}, @dots{}
14060 @kindex set default-collect
14061 @cindex default collection action
14062 This variable is a list of expressions to collect at each tracepoint
14063 hit. It is effectively an additional @code{collect} action prepended
14064 to every tracepoint action list. The expressions are parsed
14065 individually for each tracepoint, so for instance a variable named
14066 @code{xyz} may be interpreted as a global for one tracepoint, and a
14067 local for another, as appropriate to the tracepoint's location.
14069 @item show default-collect
14070 @kindex show default-collect
14071 Show the list of expressions that are collected by default at each
14076 @node Listing Tracepoints
14077 @subsection Listing Tracepoints
14080 @kindex info tracepoints @r{[}@var{n}@dots{}@r{]}
14081 @kindex info tp @r{[}@var{n}@dots{}@r{]}
14082 @cindex information about tracepoints
14083 @item info tracepoints @r{[}@var{num}@dots{}@r{]}
14084 Display information about the tracepoint @var{num}. If you don't
14085 specify a tracepoint number, displays information about all the
14086 tracepoints defined so far. The format is similar to that used for
14087 @code{info breakpoints}; in fact, @code{info tracepoints} is the same
14088 command, simply restricting itself to tracepoints.
14090 A tracepoint's listing may include additional information specific to
14095 its passcount as given by the @code{passcount @var{n}} command
14098 the state about installed on target of each location
14102 (@value{GDBP}) @b{info trace}
14103 Num Type Disp Enb Address What
14104 1 tracepoint keep y 0x0804ab57 in foo() at main.cxx:7
14106 collect globfoo, $regs
14111 2 tracepoint keep y <MULTIPLE>
14113 2.1 y 0x0804859c in func4 at change-loc.h:35
14114 installed on target
14115 2.2 y 0xb7ffc480 in func4 at change-loc.h:35
14116 installed on target
14117 2.3 y <PENDING> set_tracepoint
14118 3 tracepoint keep y 0x080485b1 in foo at change-loc.c:29
14119 not installed on target
14124 This command can be abbreviated @code{info tp}.
14127 @node Listing Static Tracepoint Markers
14128 @subsection Listing Static Tracepoint Markers
14131 @kindex info static-tracepoint-markers
14132 @cindex information about static tracepoint markers
14133 @item info static-tracepoint-markers
14134 Display information about all static tracepoint markers defined in the
14137 For each marker, the following columns are printed:
14141 An incrementing counter, output to help readability. This is not a
14144 The marker ID, as reported by the target.
14145 @item Enabled or Disabled
14146 Probed markers are tagged with @samp{y}. @samp{n} identifies marks
14147 that are not enabled.
14149 Where the marker is in your program, as a memory address.
14151 Where the marker is in the source for your program, as a file and line
14152 number. If the debug information included in the program does not
14153 allow @value{GDBN} to locate the source of the marker, this column
14154 will be left blank.
14158 In addition, the following information may be printed for each marker:
14162 User data passed to the tracing library by the marker call. In the
14163 UST backend, this is the format string passed as argument to the
14165 @item Static tracepoints probing the marker
14166 The list of static tracepoints attached to the marker.
14170 (@value{GDBP}) info static-tracepoint-markers
14171 Cnt ID Enb Address What
14172 1 ust/bar2 y 0x0000000000400e1a in main at stexample.c:25
14173 Data: number1 %d number2 %d
14174 Probed by static tracepoints: #2
14175 2 ust/bar33 n 0x0000000000400c87 in main at stexample.c:24
14181 @node Starting and Stopping Trace Experiments
14182 @subsection Starting and Stopping Trace Experiments
14185 @kindex tstart [ @var{notes} ]
14186 @cindex start a new trace experiment
14187 @cindex collected data discarded
14189 This command starts the trace experiment, and begins collecting data.
14190 It has the side effect of discarding all the data collected in the
14191 trace buffer during the previous trace experiment. If any arguments
14192 are supplied, they are taken as a note and stored with the trace
14193 experiment's state. The notes may be arbitrary text, and are
14194 especially useful with disconnected tracing in a multi-user context;
14195 the notes can explain what the trace is doing, supply user contact
14196 information, and so forth.
14198 @kindex tstop [ @var{notes} ]
14199 @cindex stop a running trace experiment
14201 This command stops the trace experiment. If any arguments are
14202 supplied, they are recorded with the experiment as a note. This is
14203 useful if you are stopping a trace started by someone else, for
14204 instance if the trace is interfering with the system's behavior and
14205 needs to be stopped quickly.
14207 @strong{Note}: a trace experiment and data collection may stop
14208 automatically if any tracepoint's passcount is reached
14209 (@pxref{Tracepoint Passcounts}), or if the trace buffer becomes full.
14212 @cindex status of trace data collection
14213 @cindex trace experiment, status of
14215 This command displays the status of the current trace data
14219 Here is an example of the commands we described so far:
14222 (@value{GDBP}) @b{trace gdb_c_test}
14223 (@value{GDBP}) @b{actions}
14224 Enter actions for tracepoint #1, one per line.
14225 > collect $regs,$locals,$args
14226 > while-stepping 11
14230 (@value{GDBP}) @b{tstart}
14231 [time passes @dots{}]
14232 (@value{GDBP}) @b{tstop}
14235 @anchor{disconnected tracing}
14236 @cindex disconnected tracing
14237 You can choose to continue running the trace experiment even if
14238 @value{GDBN} disconnects from the target, voluntarily or
14239 involuntarily. For commands such as @code{detach}, the debugger will
14240 ask what you want to do with the trace. But for unexpected
14241 terminations (@value{GDBN} crash, network outage), it would be
14242 unfortunate to lose hard-won trace data, so the variable
14243 @code{disconnected-tracing} lets you decide whether the trace should
14244 continue running without @value{GDBN}.
14247 @item set disconnected-tracing on
14248 @itemx set disconnected-tracing off
14249 @kindex set disconnected-tracing
14250 Choose whether a tracing run should continue to run if @value{GDBN}
14251 has disconnected from the target. Note that @code{detach} or
14252 @code{quit} will ask you directly what to do about a running trace no
14253 matter what this variable's setting, so the variable is mainly useful
14254 for handling unexpected situations, such as loss of the network.
14256 @item show disconnected-tracing
14257 @kindex show disconnected-tracing
14258 Show the current choice for disconnected tracing.
14262 When you reconnect to the target, the trace experiment may or may not
14263 still be running; it might have filled the trace buffer in the
14264 meantime, or stopped for one of the other reasons. If it is running,
14265 it will continue after reconnection.
14267 Upon reconnection, the target will upload information about the
14268 tracepoints in effect. @value{GDBN} will then compare that
14269 information to the set of tracepoints currently defined, and attempt
14270 to match them up, allowing for the possibility that the numbers may
14271 have changed due to creation and deletion in the meantime. If one of
14272 the target's tracepoints does not match any in @value{GDBN}, the
14273 debugger will create a new tracepoint, so that you have a number with
14274 which to specify that tracepoint. This matching-up process is
14275 necessarily heuristic, and it may result in useless tracepoints being
14276 created; you may simply delete them if they are of no use.
14278 @cindex circular trace buffer
14279 If your target agent supports a @dfn{circular trace buffer}, then you
14280 can run a trace experiment indefinitely without filling the trace
14281 buffer; when space runs out, the agent deletes already-collected trace
14282 frames, oldest first, until there is enough room to continue
14283 collecting. This is especially useful if your tracepoints are being
14284 hit too often, and your trace gets terminated prematurely because the
14285 buffer is full. To ask for a circular trace buffer, simply set
14286 @samp{circular-trace-buffer} to on. You can set this at any time,
14287 including during tracing; if the agent can do it, it will change
14288 buffer handling on the fly, otherwise it will not take effect until
14292 @item set circular-trace-buffer on
14293 @itemx set circular-trace-buffer off
14294 @kindex set circular-trace-buffer
14295 Choose whether a tracing run should use a linear or circular buffer
14296 for trace data. A linear buffer will not lose any trace data, but may
14297 fill up prematurely, while a circular buffer will discard old trace
14298 data, but it will have always room for the latest tracepoint hits.
14300 @item show circular-trace-buffer
14301 @kindex show circular-trace-buffer
14302 Show the current choice for the trace buffer. Note that this may not
14303 match the agent's current buffer handling, nor is it guaranteed to
14304 match the setting that might have been in effect during a past run,
14305 for instance if you are looking at frames from a trace file.
14310 @item set trace-buffer-size @var{n}
14311 @itemx set trace-buffer-size unlimited
14312 @kindex set trace-buffer-size
14313 Request that the target use a trace buffer of @var{n} bytes. Not all
14314 targets will honor the request; they may have a compiled-in size for
14315 the trace buffer, or some other limitation. Set to a value of
14316 @code{unlimited} or @code{-1} to let the target use whatever size it
14317 likes. This is also the default.
14319 @item show trace-buffer-size
14320 @kindex show trace-buffer-size
14321 Show the current requested size for the trace buffer. Note that this
14322 will only match the actual size if the target supports size-setting,
14323 and was able to handle the requested size. For instance, if the
14324 target can only change buffer size between runs, this variable will
14325 not reflect the change until the next run starts. Use @code{tstatus}
14326 to get a report of the actual buffer size.
14330 @item set trace-user @var{text}
14331 @kindex set trace-user
14333 @item show trace-user
14334 @kindex show trace-user
14336 @item set trace-notes @var{text}
14337 @kindex set trace-notes
14338 Set the trace run's notes.
14340 @item show trace-notes
14341 @kindex show trace-notes
14342 Show the trace run's notes.
14344 @item set trace-stop-notes @var{text}
14345 @kindex set trace-stop-notes
14346 Set the trace run's stop notes. The handling of the note is as for
14347 @code{tstop} arguments; the set command is convenient way to fix a
14348 stop note that is mistaken or incomplete.
14350 @item show trace-stop-notes
14351 @kindex show trace-stop-notes
14352 Show the trace run's stop notes.
14356 @node Tracepoint Restrictions
14357 @subsection Tracepoint Restrictions
14359 @cindex tracepoint restrictions
14360 There are a number of restrictions on the use of tracepoints. As
14361 described above, tracepoint data gathering occurs on the target
14362 without interaction from @value{GDBN}. Thus the full capabilities of
14363 the debugger are not available during data gathering, and then at data
14364 examination time, you will be limited by only having what was
14365 collected. The following items describe some common problems, but it
14366 is not exhaustive, and you may run into additional difficulties not
14372 Tracepoint expressions are intended to gather objects (lvalues). Thus
14373 the full flexibility of GDB's expression evaluator is not available.
14374 You cannot call functions, cast objects to aggregate types, access
14375 convenience variables or modify values (except by assignment to trace
14376 state variables). Some language features may implicitly call
14377 functions (for instance Objective-C fields with accessors), and therefore
14378 cannot be collected either.
14381 Collection of local variables, either individually or in bulk with
14382 @code{$locals} or @code{$args}, during @code{while-stepping} may
14383 behave erratically. The stepping action may enter a new scope (for
14384 instance by stepping into a function), or the location of the variable
14385 may change (for instance it is loaded into a register). The
14386 tracepoint data recorded uses the location information for the
14387 variables that is correct for the tracepoint location. When the
14388 tracepoint is created, it is not possible, in general, to determine
14389 where the steps of a @code{while-stepping} sequence will advance the
14390 program---particularly if a conditional branch is stepped.
14393 Collection of an incompletely-initialized or partially-destroyed object
14394 may result in something that @value{GDBN} cannot display, or displays
14395 in a misleading way.
14398 When @value{GDBN} displays a pointer to character it automatically
14399 dereferences the pointer to also display characters of the string
14400 being pointed to. However, collecting the pointer during tracing does
14401 not automatically collect the string. You need to explicitly
14402 dereference the pointer and provide size information if you want to
14403 collect not only the pointer, but the memory pointed to. For example,
14404 @code{*ptr@@50} can be used to collect the 50 element array pointed to
14408 It is not possible to collect a complete stack backtrace at a
14409 tracepoint. Instead, you may collect the registers and a few hundred
14410 bytes from the stack pointer with something like @code{*(unsigned char *)$esp@@300}
14411 (adjust to use the name of the actual stack pointer register on your
14412 target architecture, and the amount of stack you wish to capture).
14413 Then the @code{backtrace} command will show a partial backtrace when
14414 using a trace frame. The number of stack frames that can be examined
14415 depends on the sizes of the frames in the collected stack. Note that
14416 if you ask for a block so large that it goes past the bottom of the
14417 stack, the target agent may report an error trying to read from an
14421 If you do not collect registers at a tracepoint, @value{GDBN} can
14422 infer that the value of @code{$pc} must be the same as the address of
14423 the tracepoint and use that when you are looking at a trace frame
14424 for that tracepoint. However, this cannot work if the tracepoint has
14425 multiple locations (for instance if it was set in a function that was
14426 inlined), or if it has a @code{while-stepping} loop. In those cases
14427 @value{GDBN} will warn you that it can't infer @code{$pc}, and default
14432 @node Analyze Collected Data
14433 @section Using the Collected Data
14435 After the tracepoint experiment ends, you use @value{GDBN} commands
14436 for examining the trace data. The basic idea is that each tracepoint
14437 collects a trace @dfn{snapshot} every time it is hit and another
14438 snapshot every time it single-steps. All these snapshots are
14439 consecutively numbered from zero and go into a buffer, and you can
14440 examine them later. The way you examine them is to @dfn{focus} on a
14441 specific trace snapshot. When the remote stub is focused on a trace
14442 snapshot, it will respond to all @value{GDBN} requests for memory and
14443 registers by reading from the buffer which belongs to that snapshot,
14444 rather than from @emph{real} memory or registers of the program being
14445 debugged. This means that @strong{all} @value{GDBN} commands
14446 (@code{print}, @code{info registers}, @code{backtrace}, etc.) will
14447 behave as if we were currently debugging the program state as it was
14448 when the tracepoint occurred. Any requests for data that are not in
14449 the buffer will fail.
14452 * tfind:: How to select a trace snapshot
14453 * tdump:: How to display all data for a snapshot
14454 * save tracepoints:: How to save tracepoints for a future run
14458 @subsection @code{tfind @var{n}}
14461 @cindex select trace snapshot
14462 @cindex find trace snapshot
14463 The basic command for selecting a trace snapshot from the buffer is
14464 @code{tfind @var{n}}, which finds trace snapshot number @var{n},
14465 counting from zero. If no argument @var{n} is given, the next
14466 snapshot is selected.
14468 Here are the various forms of using the @code{tfind} command.
14472 Find the first snapshot in the buffer. This is a synonym for
14473 @code{tfind 0} (since 0 is the number of the first snapshot).
14476 Stop debugging trace snapshots, resume @emph{live} debugging.
14479 Same as @samp{tfind none}.
14482 No argument means find the next trace snapshot or find the first
14483 one if no trace snapshot is selected.
14486 Find the previous trace snapshot before the current one. This permits
14487 retracing earlier steps.
14489 @item tfind tracepoint @var{num}
14490 Find the next snapshot associated with tracepoint @var{num}. Search
14491 proceeds forward from the last examined trace snapshot. If no
14492 argument @var{num} is given, it means find the next snapshot collected
14493 for the same tracepoint as the current snapshot.
14495 @item tfind pc @var{addr}
14496 Find the next snapshot associated with the value @var{addr} of the
14497 program counter. Search proceeds forward from the last examined trace
14498 snapshot. If no argument @var{addr} is given, it means find the next
14499 snapshot with the same value of PC as the current snapshot.
14501 @item tfind outside @var{addr1}, @var{addr2}
14502 Find the next snapshot whose PC is outside the given range of
14503 addresses (exclusive).
14505 @item tfind range @var{addr1}, @var{addr2}
14506 Find the next snapshot whose PC is between @var{addr1} and
14507 @var{addr2} (inclusive).
14509 @item tfind line @r{[}@var{file}:@r{]}@var{n}
14510 Find the next snapshot associated with the source line @var{n}. If
14511 the optional argument @var{file} is given, refer to line @var{n} in
14512 that source file. Search proceeds forward from the last examined
14513 trace snapshot. If no argument @var{n} is given, it means find the
14514 next line other than the one currently being examined; thus saying
14515 @code{tfind line} repeatedly can appear to have the same effect as
14516 stepping from line to line in a @emph{live} debugging session.
14519 The default arguments for the @code{tfind} commands are specifically
14520 designed to make it easy to scan through the trace buffer. For
14521 instance, @code{tfind} with no argument selects the next trace
14522 snapshot, and @code{tfind -} with no argument selects the previous
14523 trace snapshot. So, by giving one @code{tfind} command, and then
14524 simply hitting @key{RET} repeatedly you can examine all the trace
14525 snapshots in order. Or, by saying @code{tfind -} and then hitting
14526 @key{RET} repeatedly you can examine the snapshots in reverse order.
14527 The @code{tfind line} command with no argument selects the snapshot
14528 for the next source line executed. The @code{tfind pc} command with
14529 no argument selects the next snapshot with the same program counter
14530 (PC) as the current frame. The @code{tfind tracepoint} command with
14531 no argument selects the next trace snapshot collected by the same
14532 tracepoint as the current one.
14534 In addition to letting you scan through the trace buffer manually,
14535 these commands make it easy to construct @value{GDBN} scripts that
14536 scan through the trace buffer and print out whatever collected data
14537 you are interested in. Thus, if we want to examine the PC, FP, and SP
14538 registers from each trace frame in the buffer, we can say this:
14541 (@value{GDBP}) @b{tfind start}
14542 (@value{GDBP}) @b{while ($trace_frame != -1)}
14543 > printf "Frame %d, PC = %08X, SP = %08X, FP = %08X\n", \
14544 $trace_frame, $pc, $sp, $fp
14548 Frame 0, PC = 0020DC64, SP = 0030BF3C, FP = 0030BF44
14549 Frame 1, PC = 0020DC6C, SP = 0030BF38, FP = 0030BF44
14550 Frame 2, PC = 0020DC70, SP = 0030BF34, FP = 0030BF44
14551 Frame 3, PC = 0020DC74, SP = 0030BF30, FP = 0030BF44
14552 Frame 4, PC = 0020DC78, SP = 0030BF2C, FP = 0030BF44
14553 Frame 5, PC = 0020DC7C, SP = 0030BF28, FP = 0030BF44
14554 Frame 6, PC = 0020DC80, SP = 0030BF24, FP = 0030BF44
14555 Frame 7, PC = 0020DC84, SP = 0030BF20, FP = 0030BF44
14556 Frame 8, PC = 0020DC88, SP = 0030BF1C, FP = 0030BF44
14557 Frame 9, PC = 0020DC8E, SP = 0030BF18, FP = 0030BF44
14558 Frame 10, PC = 00203F6C, SP = 0030BE3C, FP = 0030BF14
14561 Or, if we want to examine the variable @code{X} at each source line in
14565 (@value{GDBP}) @b{tfind start}
14566 (@value{GDBP}) @b{while ($trace_frame != -1)}
14567 > printf "Frame %d, X == %d\n", $trace_frame, X
14577 @subsection @code{tdump}
14579 @cindex dump all data collected at tracepoint
14580 @cindex tracepoint data, display
14582 This command takes no arguments. It prints all the data collected at
14583 the current trace snapshot.
14586 (@value{GDBP}) @b{trace 444}
14587 (@value{GDBP}) @b{actions}
14588 Enter actions for tracepoint #2, one per line:
14589 > collect $regs, $locals, $args, gdb_long_test
14592 (@value{GDBP}) @b{tstart}
14594 (@value{GDBP}) @b{tfind line 444}
14595 #0 gdb_test (p1=0x11, p2=0x22, p3=0x33, p4=0x44, p5=0x55, p6=0x66)
14597 444 printp( "%s: arguments = 0x%X 0x%X 0x%X 0x%X 0x%X 0x%X\n", )
14599 (@value{GDBP}) @b{tdump}
14600 Data collected at tracepoint 2, trace frame 1:
14601 d0 0xc4aa0085 -995491707
14605 d4 0x71aea3d 119204413
14608 d7 0x380035 3670069
14609 a0 0x19e24a 1696330
14610 a1 0x3000668 50333288
14612 a3 0x322000 3284992
14613 a4 0x3000698 50333336
14614 a5 0x1ad3cc 1758156
14615 fp 0x30bf3c 0x30bf3c
14616 sp 0x30bf34 0x30bf34
14618 pc 0x20b2c8 0x20b2c8
14622 p = 0x20e5b4 "gdb-test"
14629 gdb_long_test = 17 '\021'
14634 @code{tdump} works by scanning the tracepoint's current collection
14635 actions and printing the value of each expression listed. So
14636 @code{tdump} can fail, if after a run, you change the tracepoint's
14637 actions to mention variables that were not collected during the run.
14639 Also, for tracepoints with @code{while-stepping} loops, @code{tdump}
14640 uses the collected value of @code{$pc} to distinguish between trace
14641 frames that were collected at the tracepoint hit, and frames that were
14642 collected while stepping. This allows it to correctly choose whether
14643 to display the basic list of collections, or the collections from the
14644 body of the while-stepping loop. However, if @code{$pc} was not collected,
14645 then @code{tdump} will always attempt to dump using the basic collection
14646 list, and may fail if a while-stepping frame does not include all the
14647 same data that is collected at the tracepoint hit.
14648 @c This is getting pretty arcane, example would be good.
14650 @node save tracepoints
14651 @subsection @code{save tracepoints @var{filename}}
14652 @kindex save tracepoints
14653 @kindex save-tracepoints
14654 @cindex save tracepoints for future sessions
14656 This command saves all current tracepoint definitions together with
14657 their actions and passcounts, into a file @file{@var{filename}}
14658 suitable for use in a later debugging session. To read the saved
14659 tracepoint definitions, use the @code{source} command (@pxref{Command
14660 Files}). The @w{@code{save-tracepoints}} command is a deprecated
14661 alias for @w{@code{save tracepoints}}
14663 @node Tracepoint Variables
14664 @section Convenience Variables for Tracepoints
14665 @cindex tracepoint variables
14666 @cindex convenience variables for tracepoints
14669 @vindex $trace_frame
14670 @item (int) $trace_frame
14671 The current trace snapshot (a.k.a.@: @dfn{frame}) number, or -1 if no
14672 snapshot is selected.
14674 @vindex $tracepoint
14675 @item (int) $tracepoint
14676 The tracepoint for the current trace snapshot.
14678 @vindex $trace_line
14679 @item (int) $trace_line
14680 The line number for the current trace snapshot.
14682 @vindex $trace_file
14683 @item (char []) $trace_file
14684 The source file for the current trace snapshot.
14686 @vindex $trace_func
14687 @item (char []) $trace_func
14688 The name of the function containing @code{$tracepoint}.
14691 Note: @code{$trace_file} is not suitable for use in @code{printf},
14692 use @code{output} instead.
14694 Here's a simple example of using these convenience variables for
14695 stepping through all the trace snapshots and printing some of their
14696 data. Note that these are not the same as trace state variables,
14697 which are managed by the target.
14700 (@value{GDBP}) @b{tfind start}
14702 (@value{GDBP}) @b{while $trace_frame != -1}
14703 > output $trace_file
14704 > printf ", line %d (tracepoint #%d)\n", $trace_line, $tracepoint
14710 @section Using Trace Files
14711 @cindex trace files
14713 In some situations, the target running a trace experiment may no
14714 longer be available; perhaps it crashed, or the hardware was needed
14715 for a different activity. To handle these cases, you can arrange to
14716 dump the trace data into a file, and later use that file as a source
14717 of trace data, via the @code{target tfile} command.
14722 @item tsave [ -r ] @var{filename}
14723 @itemx tsave [-ctf] @var{dirname}
14724 Save the trace data to @var{filename}. By default, this command
14725 assumes that @var{filename} refers to the host filesystem, so if
14726 necessary @value{GDBN} will copy raw trace data up from the target and
14727 then save it. If the target supports it, you can also supply the
14728 optional argument @code{-r} (``remote'') to direct the target to save
14729 the data directly into @var{filename} in its own filesystem, which may be
14730 more efficient if the trace buffer is very large. (Note, however, that
14731 @code{target tfile} can only read from files accessible to the host.)
14732 By default, this command will save trace frame in tfile format.
14733 You can supply the optional argument @code{-ctf} to save data in CTF
14734 format. The @dfn{Common Trace Format} (CTF) is proposed as a trace format
14735 that can be shared by multiple debugging and tracing tools. Please go to
14736 @indicateurl{http://www.efficios.com/ctf} to get more information.
14738 @kindex target tfile
14742 @item target tfile @var{filename}
14743 @itemx target ctf @var{dirname}
14744 Use the file named @var{filename} or directory named @var{dirname} as
14745 a source of trace data. Commands that examine data work as they do with
14746 a live target, but it is not possible to run any new trace experiments.
14747 @code{tstatus} will report the state of the trace run at the moment
14748 the data was saved, as well as the current trace frame you are examining.
14749 Both @var{filename} and @var{dirname} must be on a filesystem accessible to
14753 (@value{GDBP}) target ctf ctf.ctf
14754 (@value{GDBP}) tfind
14755 Found trace frame 0, tracepoint 2
14756 39 ++a; /* set tracepoint 1 here */
14757 (@value{GDBP}) tdump
14758 Data collected at tracepoint 2, trace frame 0:
14762 c = @{"123", "456", "789", "123", "456", "789"@}
14763 d = @{@{@{a = 1, b = 2@}, @{a = 3, b = 4@}@}, @{@{a = 5, b = 6@}, @{a = 7, b = 8@}@}@}
14771 @chapter Debugging Programs That Use Overlays
14774 If your program is too large to fit completely in your target system's
14775 memory, you can sometimes use @dfn{overlays} to work around this
14776 problem. @value{GDBN} provides some support for debugging programs that
14780 * How Overlays Work:: A general explanation of overlays.
14781 * Overlay Commands:: Managing overlays in @value{GDBN}.
14782 * Automatic Overlay Debugging:: @value{GDBN} can find out which overlays are
14783 mapped by asking the inferior.
14784 * Overlay Sample Program:: A sample program using overlays.
14787 @node How Overlays Work
14788 @section How Overlays Work
14789 @cindex mapped overlays
14790 @cindex unmapped overlays
14791 @cindex load address, overlay's
14792 @cindex mapped address
14793 @cindex overlay area
14795 Suppose you have a computer whose instruction address space is only 64
14796 kilobytes long, but which has much more memory which can be accessed by
14797 other means: special instructions, segment registers, or memory
14798 management hardware, for example. Suppose further that you want to
14799 adapt a program which is larger than 64 kilobytes to run on this system.
14801 One solution is to identify modules of your program which are relatively
14802 independent, and need not call each other directly; call these modules
14803 @dfn{overlays}. Separate the overlays from the main program, and place
14804 their machine code in the larger memory. Place your main program in
14805 instruction memory, but leave at least enough space there to hold the
14806 largest overlay as well.
14808 Now, to call a function located in an overlay, you must first copy that
14809 overlay's machine code from the large memory into the space set aside
14810 for it in the instruction memory, and then jump to its entry point
14813 @c NB: In the below the mapped area's size is greater or equal to the
14814 @c size of all overlays. This is intentional to remind the developer
14815 @c that overlays don't necessarily need to be the same size.
14819 Data Instruction Larger
14820 Address Space Address Space Address Space
14821 +-----------+ +-----------+ +-----------+
14823 +-----------+ +-----------+ +-----------+<-- overlay 1
14824 | program | | main | .----| overlay 1 | load address
14825 | variables | | program | | +-----------+
14826 | and heap | | | | | |
14827 +-----------+ | | | +-----------+<-- overlay 2
14828 | | +-----------+ | | | load address
14829 +-----------+ | | | .-| overlay 2 |
14831 mapped --->+-----------+ | | +-----------+
14832 address | | | | | |
14833 | overlay | <-' | | |
14834 | area | <---' +-----------+<-- overlay 3
14835 | | <---. | | load address
14836 +-----------+ `--| overlay 3 |
14843 @anchor{A code overlay}A code overlay
14847 The diagram (@pxref{A code overlay}) shows a system with separate data
14848 and instruction address spaces. To map an overlay, the program copies
14849 its code from the larger address space to the instruction address space.
14850 Since the overlays shown here all use the same mapped address, only one
14851 may be mapped at a time. For a system with a single address space for
14852 data and instructions, the diagram would be similar, except that the
14853 program variables and heap would share an address space with the main
14854 program and the overlay area.
14856 An overlay loaded into instruction memory and ready for use is called a
14857 @dfn{mapped} overlay; its @dfn{mapped address} is its address in the
14858 instruction memory. An overlay not present (or only partially present)
14859 in instruction memory is called @dfn{unmapped}; its @dfn{load address}
14860 is its address in the larger memory. The mapped address is also called
14861 the @dfn{virtual memory address}, or @dfn{VMA}; the load address is also
14862 called the @dfn{load memory address}, or @dfn{LMA}.
14864 Unfortunately, overlays are not a completely transparent way to adapt a
14865 program to limited instruction memory. They introduce a new set of
14866 global constraints you must keep in mind as you design your program:
14871 Before calling or returning to a function in an overlay, your program
14872 must make sure that overlay is actually mapped. Otherwise, the call or
14873 return will transfer control to the right address, but in the wrong
14874 overlay, and your program will probably crash.
14877 If the process of mapping an overlay is expensive on your system, you
14878 will need to choose your overlays carefully to minimize their effect on
14879 your program's performance.
14882 The executable file you load onto your system must contain each
14883 overlay's instructions, appearing at the overlay's load address, not its
14884 mapped address. However, each overlay's instructions must be relocated
14885 and its symbols defined as if the overlay were at its mapped address.
14886 You can use GNU linker scripts to specify different load and relocation
14887 addresses for pieces of your program; see @ref{Overlay Description,,,
14888 ld.info, Using ld: the GNU linker}.
14891 The procedure for loading executable files onto your system must be able
14892 to load their contents into the larger address space as well as the
14893 instruction and data spaces.
14897 The overlay system described above is rather simple, and could be
14898 improved in many ways:
14903 If your system has suitable bank switch registers or memory management
14904 hardware, you could use those facilities to make an overlay's load area
14905 contents simply appear at their mapped address in instruction space.
14906 This would probably be faster than copying the overlay to its mapped
14907 area in the usual way.
14910 If your overlays are small enough, you could set aside more than one
14911 overlay area, and have more than one overlay mapped at a time.
14914 You can use overlays to manage data, as well as instructions. In
14915 general, data overlays are even less transparent to your design than
14916 code overlays: whereas code overlays only require care when you call or
14917 return to functions, data overlays require care every time you access
14918 the data. Also, if you change the contents of a data overlay, you
14919 must copy its contents back out to its load address before you can copy a
14920 different data overlay into the same mapped area.
14925 @node Overlay Commands
14926 @section Overlay Commands
14928 To use @value{GDBN}'s overlay support, each overlay in your program must
14929 correspond to a separate section of the executable file. The section's
14930 virtual memory address and load memory address must be the overlay's
14931 mapped and load addresses. Identifying overlays with sections allows
14932 @value{GDBN} to determine the appropriate address of a function or
14933 variable, depending on whether the overlay is mapped or not.
14935 @value{GDBN}'s overlay commands all start with the word @code{overlay};
14936 you can abbreviate this as @code{ov} or @code{ovly}. The commands are:
14941 Disable @value{GDBN}'s overlay support. When overlay support is
14942 disabled, @value{GDBN} assumes that all functions and variables are
14943 always present at their mapped addresses. By default, @value{GDBN}'s
14944 overlay support is disabled.
14946 @item overlay manual
14947 @cindex manual overlay debugging
14948 Enable @dfn{manual} overlay debugging. In this mode, @value{GDBN}
14949 relies on you to tell it which overlays are mapped, and which are not,
14950 using the @code{overlay map-overlay} and @code{overlay unmap-overlay}
14951 commands described below.
14953 @item overlay map-overlay @var{overlay}
14954 @itemx overlay map @var{overlay}
14955 @cindex map an overlay
14956 Tell @value{GDBN} that @var{overlay} is now mapped; @var{overlay} must
14957 be the name of the object file section containing the overlay. When an
14958 overlay is mapped, @value{GDBN} assumes it can find the overlay's
14959 functions and variables at their mapped addresses. @value{GDBN} assumes
14960 that any other overlays whose mapped ranges overlap that of
14961 @var{overlay} are now unmapped.
14963 @item overlay unmap-overlay @var{overlay}
14964 @itemx overlay unmap @var{overlay}
14965 @cindex unmap an overlay
14966 Tell @value{GDBN} that @var{overlay} is no longer mapped; @var{overlay}
14967 must be the name of the object file section containing the overlay.
14968 When an overlay is unmapped, @value{GDBN} assumes it can find the
14969 overlay's functions and variables at their load addresses.
14972 Enable @dfn{automatic} overlay debugging. In this mode, @value{GDBN}
14973 consults a data structure the overlay manager maintains in the inferior
14974 to see which overlays are mapped. For details, see @ref{Automatic
14975 Overlay Debugging}.
14977 @item overlay load-target
14978 @itemx overlay load
14979 @cindex reloading the overlay table
14980 Re-read the overlay table from the inferior. Normally, @value{GDBN}
14981 re-reads the table @value{GDBN} automatically each time the inferior
14982 stops, so this command should only be necessary if you have changed the
14983 overlay mapping yourself using @value{GDBN}. This command is only
14984 useful when using automatic overlay debugging.
14986 @item overlay list-overlays
14987 @itemx overlay list
14988 @cindex listing mapped overlays
14989 Display a list of the overlays currently mapped, along with their mapped
14990 addresses, load addresses, and sizes.
14994 Normally, when @value{GDBN} prints a code address, it includes the name
14995 of the function the address falls in:
14998 (@value{GDBP}) print main
14999 $3 = @{int ()@} 0x11a0 <main>
15002 When overlay debugging is enabled, @value{GDBN} recognizes code in
15003 unmapped overlays, and prints the names of unmapped functions with
15004 asterisks around them. For example, if @code{foo} is a function in an
15005 unmapped overlay, @value{GDBN} prints it this way:
15008 (@value{GDBP}) overlay list
15009 No sections are mapped.
15010 (@value{GDBP}) print foo
15011 $5 = @{int (int)@} 0x100000 <*foo*>
15014 When @code{foo}'s overlay is mapped, @value{GDBN} prints the function's
15018 (@value{GDBP}) overlay list
15019 Section .ov.foo.text, loaded at 0x100000 - 0x100034,
15020 mapped at 0x1016 - 0x104a
15021 (@value{GDBP}) print foo
15022 $6 = @{int (int)@} 0x1016 <foo>
15025 When overlay debugging is enabled, @value{GDBN} can find the correct
15026 address for functions and variables in an overlay, whether or not the
15027 overlay is mapped. This allows most @value{GDBN} commands, like
15028 @code{break} and @code{disassemble}, to work normally, even on unmapped
15029 code. However, @value{GDBN}'s breakpoint support has some limitations:
15033 @cindex breakpoints in overlays
15034 @cindex overlays, setting breakpoints in
15035 You can set breakpoints in functions in unmapped overlays, as long as
15036 @value{GDBN} can write to the overlay at its load address.
15038 @value{GDBN} can not set hardware or simulator-based breakpoints in
15039 unmapped overlays. However, if you set a breakpoint at the end of your
15040 overlay manager (and tell @value{GDBN} which overlays are now mapped, if
15041 you are using manual overlay management), @value{GDBN} will re-set its
15042 breakpoints properly.
15046 @node Automatic Overlay Debugging
15047 @section Automatic Overlay Debugging
15048 @cindex automatic overlay debugging
15050 @value{GDBN} can automatically track which overlays are mapped and which
15051 are not, given some simple co-operation from the overlay manager in the
15052 inferior. If you enable automatic overlay debugging with the
15053 @code{overlay auto} command (@pxref{Overlay Commands}), @value{GDBN}
15054 looks in the inferior's memory for certain variables describing the
15055 current state of the overlays.
15057 Here are the variables your overlay manager must define to support
15058 @value{GDBN}'s automatic overlay debugging:
15062 @item @code{_ovly_table}:
15063 This variable must be an array of the following structures:
15068 /* The overlay's mapped address. */
15071 /* The size of the overlay, in bytes. */
15072 unsigned long size;
15074 /* The overlay's load address. */
15077 /* Non-zero if the overlay is currently mapped;
15079 unsigned long mapped;
15083 @item @code{_novlys}:
15084 This variable must be a four-byte signed integer, holding the total
15085 number of elements in @code{_ovly_table}.
15089 To decide whether a particular overlay is mapped or not, @value{GDBN}
15090 looks for an entry in @w{@code{_ovly_table}} whose @code{vma} and
15091 @code{lma} members equal the VMA and LMA of the overlay's section in the
15092 executable file. When @value{GDBN} finds a matching entry, it consults
15093 the entry's @code{mapped} member to determine whether the overlay is
15096 In addition, your overlay manager may define a function called
15097 @code{_ovly_debug_event}. If this function is defined, @value{GDBN}
15098 will silently set a breakpoint there. If the overlay manager then
15099 calls this function whenever it has changed the overlay table, this
15100 will enable @value{GDBN} to accurately keep track of which overlays
15101 are in program memory, and update any breakpoints that may be set
15102 in overlays. This will allow breakpoints to work even if the
15103 overlays are kept in ROM or other non-writable memory while they
15104 are not being executed.
15106 @node Overlay Sample Program
15107 @section Overlay Sample Program
15108 @cindex overlay example program
15110 When linking a program which uses overlays, you must place the overlays
15111 at their load addresses, while relocating them to run at their mapped
15112 addresses. To do this, you must write a linker script (@pxref{Overlay
15113 Description,,, ld.info, Using ld: the GNU linker}). Unfortunately,
15114 since linker scripts are specific to a particular host system, target
15115 architecture, and target memory layout, this manual cannot provide
15116 portable sample code demonstrating @value{GDBN}'s overlay support.
15118 However, the @value{GDBN} source distribution does contain an overlaid
15119 program, with linker scripts for a few systems, as part of its test
15120 suite. The program consists of the following files from
15121 @file{gdb/testsuite/gdb.base}:
15125 The main program file.
15127 A simple overlay manager, used by @file{overlays.c}.
15132 Overlay modules, loaded and used by @file{overlays.c}.
15135 Linker scripts for linking the test program on the @code{d10v-elf}
15136 and @code{m32r-elf} targets.
15139 You can build the test program using the @code{d10v-elf} GCC
15140 cross-compiler like this:
15143 $ d10v-elf-gcc -g -c overlays.c
15144 $ d10v-elf-gcc -g -c ovlymgr.c
15145 $ d10v-elf-gcc -g -c foo.c
15146 $ d10v-elf-gcc -g -c bar.c
15147 $ d10v-elf-gcc -g -c baz.c
15148 $ d10v-elf-gcc -g -c grbx.c
15149 $ d10v-elf-gcc -g overlays.o ovlymgr.o foo.o bar.o \
15150 baz.o grbx.o -Wl,-Td10v.ld -o overlays
15153 The build process is identical for any other architecture, except that
15154 you must substitute the appropriate compiler and linker script for the
15155 target system for @code{d10v-elf-gcc} and @code{d10v.ld}.
15159 @chapter Using @value{GDBN} with Different Languages
15162 Although programming languages generally have common aspects, they are
15163 rarely expressed in the same manner. For instance, in ANSI C,
15164 dereferencing a pointer @code{p} is accomplished by @code{*p}, but in
15165 Modula-2, it is accomplished by @code{p^}. Values can also be
15166 represented (and displayed) differently. Hex numbers in C appear as
15167 @samp{0x1ae}, while in Modula-2 they appear as @samp{1AEH}.
15169 @cindex working language
15170 Language-specific information is built into @value{GDBN} for some languages,
15171 allowing you to express operations like the above in your program's
15172 native language, and allowing @value{GDBN} to output values in a manner
15173 consistent with the syntax of your program's native language. The
15174 language you use to build expressions is called the @dfn{working
15178 * Setting:: Switching between source languages
15179 * Show:: Displaying the language
15180 * Checks:: Type and range checks
15181 * Supported Languages:: Supported languages
15182 * Unsupported Languages:: Unsupported languages
15186 @section Switching Between Source Languages
15188 There are two ways to control the working language---either have @value{GDBN}
15189 set it automatically, or select it manually yourself. You can use the
15190 @code{set language} command for either purpose. On startup, @value{GDBN}
15191 defaults to setting the language automatically. The working language is
15192 used to determine how expressions you type are interpreted, how values
15195 In addition to the working language, every source file that
15196 @value{GDBN} knows about has its own working language. For some object
15197 file formats, the compiler might indicate which language a particular
15198 source file is in. However, most of the time @value{GDBN} infers the
15199 language from the name of the file. The language of a source file
15200 controls whether C@t{++} names are demangled---this way @code{backtrace} can
15201 show each frame appropriately for its own language. There is no way to
15202 set the language of a source file from within @value{GDBN}, but you can
15203 set the language associated with a filename extension. @xref{Show, ,
15204 Displaying the Language}.
15206 This is most commonly a problem when you use a program, such
15207 as @code{cfront} or @code{f2c}, that generates C but is written in
15208 another language. In that case, make the
15209 program use @code{#line} directives in its C output; that way
15210 @value{GDBN} will know the correct language of the source code of the original
15211 program, and will display that source code, not the generated C code.
15214 * Filenames:: Filename extensions and languages.
15215 * Manually:: Setting the working language manually
15216 * Automatically:: Having @value{GDBN} infer the source language
15220 @subsection List of Filename Extensions and Languages
15222 If a source file name ends in one of the following extensions, then
15223 @value{GDBN} infers that its language is the one indicated.
15241 C@t{++} source file
15247 Objective-C source file
15251 Fortran source file
15254 Modula-2 source file
15258 Assembler source file. This actually behaves almost like C, but
15259 @value{GDBN} does not skip over function prologues when stepping.
15262 In addition, you may set the language associated with a filename
15263 extension. @xref{Show, , Displaying the Language}.
15266 @subsection Setting the Working Language
15268 If you allow @value{GDBN} to set the language automatically,
15269 expressions are interpreted the same way in your debugging session and
15272 @kindex set language
15273 If you wish, you may set the language manually. To do this, issue the
15274 command @samp{set language @var{lang}}, where @var{lang} is the name of
15275 a language, such as
15276 @code{c} or @code{modula-2}.
15277 For a list of the supported languages, type @samp{set language}.
15279 Setting the language manually prevents @value{GDBN} from updating the working
15280 language automatically. This can lead to confusion if you try
15281 to debug a program when the working language is not the same as the
15282 source language, when an expression is acceptable to both
15283 languages---but means different things. For instance, if the current
15284 source file were written in C, and @value{GDBN} was parsing Modula-2, a
15292 might not have the effect you intended. In C, this means to add
15293 @code{b} and @code{c} and place the result in @code{a}. The result
15294 printed would be the value of @code{a}. In Modula-2, this means to compare
15295 @code{a} to the result of @code{b+c}, yielding a @code{BOOLEAN} value.
15297 @node Automatically
15298 @subsection Having @value{GDBN} Infer the Source Language
15300 To have @value{GDBN} set the working language automatically, use
15301 @samp{set language local} or @samp{set language auto}. @value{GDBN}
15302 then infers the working language. That is, when your program stops in a
15303 frame (usually by encountering a breakpoint), @value{GDBN} sets the
15304 working language to the language recorded for the function in that
15305 frame. If the language for a frame is unknown (that is, if the function
15306 or block corresponding to the frame was defined in a source file that
15307 does not have a recognized extension), the current working language is
15308 not changed, and @value{GDBN} issues a warning.
15310 This may not seem necessary for most programs, which are written
15311 entirely in one source language. However, program modules and libraries
15312 written in one source language can be used by a main program written in
15313 a different source language. Using @samp{set language auto} in this
15314 case frees you from having to set the working language manually.
15317 @section Displaying the Language
15319 The following commands help you find out which language is the
15320 working language, and also what language source files were written in.
15323 @item show language
15324 @anchor{show language}
15325 @kindex show language
15326 Display the current working language. This is the
15327 language you can use with commands such as @code{print} to
15328 build and compute expressions that may involve variables in your program.
15331 @kindex info frame@r{, show the source language}
15332 Display the source language for this frame. This language becomes the
15333 working language if you use an identifier from this frame.
15334 @xref{Frame Info, ,Information about a Frame}, to identify the other
15335 information listed here.
15338 @kindex info source@r{, show the source language}
15339 Display the source language of this source file.
15340 @xref{Symbols, ,Examining the Symbol Table}, to identify the other
15341 information listed here.
15344 In unusual circumstances, you may have source files with extensions
15345 not in the standard list. You can then set the extension associated
15346 with a language explicitly:
15349 @item set extension-language @var{ext} @var{language}
15350 @kindex set extension-language
15351 Tell @value{GDBN} that source files with extension @var{ext} are to be
15352 assumed as written in the source language @var{language}.
15354 @item info extensions
15355 @kindex info extensions
15356 List all the filename extensions and the associated languages.
15360 @section Type and Range Checking
15362 Some languages are designed to guard you against making seemingly common
15363 errors through a series of compile- and run-time checks. These include
15364 checking the type of arguments to functions and operators and making
15365 sure mathematical overflows are caught at run time. Checks such as
15366 these help to ensure a program's correctness once it has been compiled
15367 by eliminating type mismatches and providing active checks for range
15368 errors when your program is running.
15370 By default @value{GDBN} checks for these errors according to the
15371 rules of the current source language. Although @value{GDBN} does not check
15372 the statements in your program, it can check expressions entered directly
15373 into @value{GDBN} for evaluation via the @code{print} command, for example.
15376 * Type Checking:: An overview of type checking
15377 * Range Checking:: An overview of range checking
15380 @cindex type checking
15381 @cindex checks, type
15382 @node Type Checking
15383 @subsection An Overview of Type Checking
15385 Some languages, such as C and C@t{++}, are strongly typed, meaning that the
15386 arguments to operators and functions have to be of the correct type,
15387 otherwise an error occurs. These checks prevent type mismatch
15388 errors from ever causing any run-time problems. For example,
15391 int klass::my_method(char *b) @{ return b ? 1 : 2; @}
15393 (@value{GDBP}) print obj.my_method (0)
15396 (@value{GDBP}) print obj.my_method (0x1234)
15397 Cannot resolve method klass::my_method to any overloaded instance
15400 The second example fails because in C@t{++} the integer constant
15401 @samp{0x1234} is not type-compatible with the pointer parameter type.
15403 For the expressions you use in @value{GDBN} commands, you can tell
15404 @value{GDBN} to not enforce strict type checking or
15405 to treat any mismatches as errors and abandon the expression;
15406 When type checking is disabled, @value{GDBN} successfully evaluates
15407 expressions like the second example above.
15409 Even if type checking is off, there may be other reasons
15410 related to type that prevent @value{GDBN} from evaluating an expression.
15411 For instance, @value{GDBN} does not know how to add an @code{int} and
15412 a @code{struct foo}. These particular type errors have nothing to do
15413 with the language in use and usually arise from expressions which make
15414 little sense to evaluate anyway.
15416 @value{GDBN} provides some additional commands for controlling type checking:
15418 @kindex set check type
15419 @kindex show check type
15421 @item set check type on
15422 @itemx set check type off
15423 Set strict type checking on or off. If any type mismatches occur in
15424 evaluating an expression while type checking is on, @value{GDBN} prints a
15425 message and aborts evaluation of the expression.
15427 @item show check type
15428 Show the current setting of type checking and whether @value{GDBN}
15429 is enforcing strict type checking rules.
15432 @cindex range checking
15433 @cindex checks, range
15434 @node Range Checking
15435 @subsection An Overview of Range Checking
15437 In some languages (such as Modula-2), it is an error to exceed the
15438 bounds of a type; this is enforced with run-time checks. Such range
15439 checking is meant to ensure program correctness by making sure
15440 computations do not overflow, or indices on an array element access do
15441 not exceed the bounds of the array.
15443 For expressions you use in @value{GDBN} commands, you can tell
15444 @value{GDBN} to treat range errors in one of three ways: ignore them,
15445 always treat them as errors and abandon the expression, or issue
15446 warnings but evaluate the expression anyway.
15448 A range error can result from numerical overflow, from exceeding an
15449 array index bound, or when you type a constant that is not a member
15450 of any type. Some languages, however, do not treat overflows as an
15451 error. In many implementations of C, mathematical overflow causes the
15452 result to ``wrap around'' to lower values---for example, if @var{m} is
15453 the largest integer value, and @var{s} is the smallest, then
15456 @var{m} + 1 @result{} @var{s}
15459 This, too, is specific to individual languages, and in some cases
15460 specific to individual compilers or machines. @xref{Supported Languages, ,
15461 Supported Languages}, for further details on specific languages.
15463 @value{GDBN} provides some additional commands for controlling the range checker:
15465 @kindex set check range
15466 @kindex show check range
15468 @item set check range auto
15469 Set range checking on or off based on the current working language.
15470 @xref{Supported Languages, ,Supported Languages}, for the default settings for
15473 @item set check range on
15474 @itemx set check range off
15475 Set range checking on or off, overriding the default setting for the
15476 current working language. A warning is issued if the setting does not
15477 match the language default. If a range error occurs and range checking is on,
15478 then a message is printed and evaluation of the expression is aborted.
15480 @item set check range warn
15481 Output messages when the @value{GDBN} range checker detects a range error,
15482 but attempt to evaluate the expression anyway. Evaluating the
15483 expression may still be impossible for other reasons, such as accessing
15484 memory that the process does not own (a typical example from many Unix
15488 Show the current setting of the range checker, and whether or not it is
15489 being set automatically by @value{GDBN}.
15492 @node Supported Languages
15493 @section Supported Languages
15495 @value{GDBN} supports C, C@t{++}, D, Go, Objective-C, Fortran,
15496 OpenCL C, Pascal, Rust, assembly, Modula-2, and Ada.
15497 @c This is false ...
15498 Some @value{GDBN} features may be used in expressions regardless of the
15499 language you use: the @value{GDBN} @code{@@} and @code{::} operators,
15500 and the @samp{@{type@}addr} construct (@pxref{Expressions,
15501 ,Expressions}) can be used with the constructs of any supported
15504 The following sections detail to what degree each source language is
15505 supported by @value{GDBN}. These sections are not meant to be language
15506 tutorials or references, but serve only as a reference guide to what the
15507 @value{GDBN} expression parser accepts, and what input and output
15508 formats should look like for different languages. There are many good
15509 books written on each of these languages; please look to these for a
15510 language reference or tutorial.
15513 * C:: C and C@t{++}
15516 * Objective-C:: Objective-C
15517 * OpenCL C:: OpenCL C
15518 * Fortran:: Fortran
15521 * Modula-2:: Modula-2
15526 @subsection C and C@t{++}
15528 @cindex C and C@t{++}
15529 @cindex expressions in C or C@t{++}
15531 Since C and C@t{++} are so closely related, many features of @value{GDBN} apply
15532 to both languages. Whenever this is the case, we discuss those languages
15536 @cindex @code{g++}, @sc{gnu} C@t{++} compiler
15537 @cindex @sc{gnu} C@t{++}
15538 The C@t{++} debugging facilities are jointly implemented by the C@t{++}
15539 compiler and @value{GDBN}. Therefore, to debug your C@t{++} code
15540 effectively, you must compile your C@t{++} programs with a supported
15541 C@t{++} compiler, such as @sc{gnu} @code{g++}, or the HP ANSI C@t{++}
15542 compiler (@code{aCC}).
15545 * C Operators:: C and C@t{++} operators
15546 * C Constants:: C and C@t{++} constants
15547 * C Plus Plus Expressions:: C@t{++} expressions
15548 * C Defaults:: Default settings for C and C@t{++}
15549 * C Checks:: C and C@t{++} type and range checks
15550 * Debugging C:: @value{GDBN} and C
15551 * Debugging C Plus Plus:: @value{GDBN} features for C@t{++}
15552 * Decimal Floating Point:: Numbers in Decimal Floating Point format
15556 @subsubsection C and C@t{++} Operators
15558 @cindex C and C@t{++} operators
15560 Operators must be defined on values of specific types. For instance,
15561 @code{+} is defined on numbers, but not on structures. Operators are
15562 often defined on groups of types.
15564 For the purposes of C and C@t{++}, the following definitions hold:
15569 @emph{Integral types} include @code{int} with any of its storage-class
15570 specifiers; @code{char}; @code{enum}; and, for C@t{++}, @code{bool}.
15573 @emph{Floating-point types} include @code{float}, @code{double}, and
15574 @code{long double} (if supported by the target platform).
15577 @emph{Pointer types} include all types defined as @code{(@var{type} *)}.
15580 @emph{Scalar types} include all of the above.
15585 The following operators are supported. They are listed here
15586 in order of increasing precedence:
15590 The comma or sequencing operator. Expressions in a comma-separated list
15591 are evaluated from left to right, with the result of the entire
15592 expression being the last expression evaluated.
15595 Assignment. The value of an assignment expression is the value
15596 assigned. Defined on scalar types.
15599 Used in an expression of the form @w{@code{@var{a} @var{op}= @var{b}}},
15600 and translated to @w{@code{@var{a} = @var{a op b}}}.
15601 @w{@code{@var{op}=}} and @code{=} have the same precedence. The operator
15602 @var{op} is any one of the operators @code{|}, @code{^}, @code{&},
15603 @code{<<}, @code{>>}, @code{+}, @code{-}, @code{*}, @code{/}, @code{%}.
15606 The ternary operator. @code{@var{a} ? @var{b} : @var{c}} can be thought
15607 of as: if @var{a} then @var{b} else @var{c}. The argument @var{a}
15608 should be of an integral type.
15611 Logical @sc{or}. Defined on integral types.
15614 Logical @sc{and}. Defined on integral types.
15617 Bitwise @sc{or}. Defined on integral types.
15620 Bitwise exclusive-@sc{or}. Defined on integral types.
15623 Bitwise @sc{and}. Defined on integral types.
15626 Equality and inequality. Defined on scalar types. The value of these
15627 expressions is 0 for false and non-zero for true.
15629 @item <@r{, }>@r{, }<=@r{, }>=
15630 Less than, greater than, less than or equal, greater than or equal.
15631 Defined on scalar types. The value of these expressions is 0 for false
15632 and non-zero for true.
15635 left shift, and right shift. Defined on integral types.
15638 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
15641 Addition and subtraction. Defined on integral types, floating-point types and
15644 @item *@r{, }/@r{, }%
15645 Multiplication, division, and modulus. Multiplication and division are
15646 defined on integral and floating-point types. Modulus is defined on
15650 Increment and decrement. When appearing before a variable, the
15651 operation is performed before the variable is used in an expression;
15652 when appearing after it, the variable's value is used before the
15653 operation takes place.
15656 Pointer dereferencing. Defined on pointer types. Same precedence as
15660 Address operator. Defined on variables. Same precedence as @code{++}.
15662 For debugging C@t{++}, @value{GDBN} implements a use of @samp{&} beyond what is
15663 allowed in the C@t{++} language itself: you can use @samp{&(&@var{ref})}
15664 to examine the address
15665 where a C@t{++} reference variable (declared with @samp{&@var{ref}}) is
15669 Negative. Defined on integral and floating-point types. Same
15670 precedence as @code{++}.
15673 Logical negation. Defined on integral types. Same precedence as
15677 Bitwise complement operator. Defined on integral types. Same precedence as
15682 Structure member, and pointer-to-structure member. For convenience,
15683 @value{GDBN} regards the two as equivalent, choosing whether to dereference a
15684 pointer based on the stored type information.
15685 Defined on @code{struct} and @code{union} data.
15688 Dereferences of pointers to members.
15691 Array indexing. @code{@var{a}[@var{i}]} is defined as
15692 @code{*(@var{a}+@var{i})}. Same precedence as @code{->}.
15695 Function parameter list. Same precedence as @code{->}.
15698 C@t{++} scope resolution operator. Defined on @code{struct}, @code{union},
15699 and @code{class} types.
15702 Doubled colons also represent the @value{GDBN} scope operator
15703 (@pxref{Expressions, ,Expressions}). Same precedence as @code{::},
15707 If an operator is redefined in the user code, @value{GDBN} usually
15708 attempts to invoke the redefined version instead of using the operator's
15709 predefined meaning.
15712 @subsubsection C and C@t{++} Constants
15714 @cindex C and C@t{++} constants
15716 @value{GDBN} allows you to express the constants of C and C@t{++} in the
15721 Integer constants are a sequence of digits. Octal constants are
15722 specified by a leading @samp{0} (i.e.@: zero), and hexadecimal constants
15723 by a leading @samp{0x} or @samp{0X}. Constants may also end with a letter
15724 @samp{l}, specifying that the constant should be treated as a
15728 Floating point constants are a sequence of digits, followed by a decimal
15729 point, followed by a sequence of digits, and optionally followed by an
15730 exponent. An exponent is of the form:
15731 @samp{@w{e@r{[[}+@r{]|}-@r{]}@var{nnn}}}, where @var{nnn} is another
15732 sequence of digits. The @samp{+} is optional for positive exponents.
15733 A floating-point constant may also end with a letter @samp{f} or
15734 @samp{F}, specifying that the constant should be treated as being of
15735 the @code{float} (as opposed to the default @code{double}) type; or with
15736 a letter @samp{l} or @samp{L}, which specifies a @code{long double}
15740 Enumerated constants consist of enumerated identifiers, or their
15741 integral equivalents.
15744 Character constants are a single character surrounded by single quotes
15745 (@code{'}), or a number---the ordinal value of the corresponding character
15746 (usually its @sc{ascii} value). Within quotes, the single character may
15747 be represented by a letter or by @dfn{escape sequences}, which are of
15748 the form @samp{\@var{nnn}}, where @var{nnn} is the octal representation
15749 of the character's ordinal value; or of the form @samp{\@var{x}}, where
15750 @samp{@var{x}} is a predefined special character---for example,
15751 @samp{\n} for newline.
15753 Wide character constants can be written by prefixing a character
15754 constant with @samp{L}, as in C. For example, @samp{L'x'} is the wide
15755 form of @samp{x}. The target wide character set is used when
15756 computing the value of this constant (@pxref{Character Sets}).
15759 String constants are a sequence of character constants surrounded by
15760 double quotes (@code{"}). Any valid character constant (as described
15761 above) may appear. Double quotes within the string must be preceded by
15762 a backslash, so for instance @samp{"a\"b'c"} is a string of five
15765 Wide string constants can be written by prefixing a string constant
15766 with @samp{L}, as in C. The target wide character set is used when
15767 computing the value of this constant (@pxref{Character Sets}).
15770 Pointer constants are an integral value. You can also write pointers
15771 to constants using the C operator @samp{&}.
15774 Array constants are comma-separated lists surrounded by braces @samp{@{}
15775 and @samp{@}}; for example, @samp{@{1,2,3@}} is a three-element array of
15776 integers, @samp{@{@{1,2@}, @{3,4@}, @{5,6@}@}} is a three-by-two array,
15777 and @samp{@{&"hi", &"there", &"fred"@}} is a three-element array of pointers.
15780 @node C Plus Plus Expressions
15781 @subsubsection C@t{++} Expressions
15783 @cindex expressions in C@t{++}
15784 @value{GDBN} expression handling can interpret most C@t{++} expressions.
15786 @cindex debugging C@t{++} programs
15787 @cindex C@t{++} compilers
15788 @cindex debug formats and C@t{++}
15789 @cindex @value{NGCC} and C@t{++}
15791 @emph{Warning:} @value{GDBN} can only debug C@t{++} code if you use
15792 the proper compiler and the proper debug format. Currently,
15793 @value{GDBN} works best when debugging C@t{++} code that is compiled
15794 with the most recent version of @value{NGCC} possible. The DWARF
15795 debugging format is preferred; @value{NGCC} defaults to this on most
15796 popular platforms. Other compilers and/or debug formats are likely to
15797 work badly or not at all when using @value{GDBN} to debug C@t{++}
15798 code. @xref{Compilation}.
15803 @cindex member functions
15805 Member function calls are allowed; you can use expressions like
15808 count = aml->GetOriginal(x, y)
15811 @vindex this@r{, inside C@t{++} member functions}
15812 @cindex namespace in C@t{++}
15814 While a member function is active (in the selected stack frame), your
15815 expressions have the same namespace available as the member function;
15816 that is, @value{GDBN} allows implicit references to the class instance
15817 pointer @code{this} following the same rules as C@t{++}. @code{using}
15818 declarations in the current scope are also respected by @value{GDBN}.
15820 @cindex call overloaded functions
15821 @cindex overloaded functions, calling
15822 @cindex type conversions in C@t{++}
15824 You can call overloaded functions; @value{GDBN} resolves the function
15825 call to the right definition, with some restrictions. @value{GDBN} does not
15826 perform overload resolution involving user-defined type conversions,
15827 calls to constructors, or instantiations of templates that do not exist
15828 in the program. It also cannot handle ellipsis argument lists or
15831 It does perform integral conversions and promotions, floating-point
15832 promotions, arithmetic conversions, pointer conversions, conversions of
15833 class objects to base classes, and standard conversions such as those of
15834 functions or arrays to pointers; it requires an exact match on the
15835 number of function arguments.
15837 Overload resolution is always performed, unless you have specified
15838 @code{set overload-resolution off}. @xref{Debugging C Plus Plus,
15839 ,@value{GDBN} Features for C@t{++}}.
15841 You must specify @code{set overload-resolution off} in order to use an
15842 explicit function signature to call an overloaded function, as in
15844 p 'foo(char,int)'('x', 13)
15847 The @value{GDBN} command-completion facility can simplify this;
15848 see @ref{Completion, ,Command Completion}.
15850 @cindex reference declarations
15852 @value{GDBN} understands variables declared as C@t{++} lvalue or rvalue
15853 references; you can use them in expressions just as you do in C@t{++}
15854 source---they are automatically dereferenced.
15856 In the parameter list shown when @value{GDBN} displays a frame, the values of
15857 reference variables are not displayed (unlike other variables); this
15858 avoids clutter, since references are often used for large structures.
15859 The @emph{address} of a reference variable is always shown, unless
15860 you have specified @samp{set print address off}.
15863 @value{GDBN} supports the C@t{++} name resolution operator @code{::}---your
15864 expressions can use it just as expressions in your program do. Since
15865 one scope may be defined in another, you can use @code{::} repeatedly if
15866 necessary, for example in an expression like
15867 @samp{@var{scope1}::@var{scope2}::@var{name}}. @value{GDBN} also allows
15868 resolving name scope by reference to source files, in both C and C@t{++}
15869 debugging (@pxref{Variables, ,Program Variables}).
15872 @value{GDBN} performs argument-dependent lookup, following the C@t{++}
15877 @subsubsection C and C@t{++} Defaults
15879 @cindex C and C@t{++} defaults
15881 If you allow @value{GDBN} to set range checking automatically, it
15882 defaults to @code{off} whenever the working language changes to
15883 C or C@t{++}. This happens regardless of whether you or @value{GDBN}
15884 selects the working language.
15886 If you allow @value{GDBN} to set the language automatically, it
15887 recognizes source files whose names end with @file{.c}, @file{.C}, or
15888 @file{.cc}, etc, and when @value{GDBN} enters code compiled from one of
15889 these files, it sets the working language to C or C@t{++}.
15890 @xref{Automatically, ,Having @value{GDBN} Infer the Source Language},
15891 for further details.
15894 @subsubsection C and C@t{++} Type and Range Checks
15896 @cindex C and C@t{++} checks
15898 By default, when @value{GDBN} parses C or C@t{++} expressions, strict type
15899 checking is used. However, if you turn type checking off, @value{GDBN}
15900 will allow certain non-standard conversions, such as promoting integer
15901 constants to pointers.
15903 Range checking, if turned on, is done on mathematical operations. Array
15904 indices are not checked, since they are often used to index a pointer
15905 that is not itself an array.
15908 @subsubsection @value{GDBN} and C
15910 The @code{set print union} and @code{show print union} commands apply to
15911 the @code{union} type. When set to @samp{on}, any @code{union} that is
15912 inside a @code{struct} or @code{class} is also printed. Otherwise, it
15913 appears as @samp{@{...@}}.
15915 The @code{@@} operator aids in the debugging of dynamic arrays, formed
15916 with pointers and a memory allocation function. @xref{Expressions,
15919 @node Debugging C Plus Plus
15920 @subsubsection @value{GDBN} Features for C@t{++}
15922 @cindex commands for C@t{++}
15924 Some @value{GDBN} commands are particularly useful with C@t{++}, and some are
15925 designed specifically for use with C@t{++}. Here is a summary:
15928 @cindex break in overloaded functions
15929 @item @r{breakpoint menus}
15930 When you want a breakpoint in a function whose name is overloaded,
15931 @value{GDBN} has the capability to display a menu of possible breakpoint
15932 locations to help you specify which function definition you want.
15933 @xref{Ambiguous Expressions,,Ambiguous Expressions}.
15935 @cindex overloading in C@t{++}
15936 @item rbreak @var{regex}
15937 Setting breakpoints using regular expressions is helpful for setting
15938 breakpoints on overloaded functions that are not members of any special
15940 @xref{Set Breaks, ,Setting Breakpoints}.
15942 @cindex C@t{++} exception handling
15944 @itemx catch rethrow
15946 Debug C@t{++} exception handling using these commands. @xref{Set
15947 Catchpoints, , Setting Catchpoints}.
15949 @cindex inheritance
15950 @item ptype @var{typename}
15951 Print inheritance relationships as well as other information for type
15953 @xref{Symbols, ,Examining the Symbol Table}.
15955 @item info vtbl @var{expression}.
15956 The @code{info vtbl} command can be used to display the virtual
15957 method tables of the object computed by @var{expression}. This shows
15958 one entry per virtual table; there may be multiple virtual tables when
15959 multiple inheritance is in use.
15961 @cindex C@t{++} demangling
15962 @item demangle @var{name}
15963 Demangle @var{name}.
15964 @xref{Symbols}, for a more complete description of the @code{demangle} command.
15966 @cindex C@t{++} symbol display
15967 @item set print demangle
15968 @itemx show print demangle
15969 @itemx set print asm-demangle
15970 @itemx show print asm-demangle
15971 Control whether C@t{++} symbols display in their source form, both when
15972 displaying code as C@t{++} source and when displaying disassemblies.
15973 @xref{Print Settings, ,Print Settings}.
15975 @item set print object
15976 @itemx show print object
15977 Choose whether to print derived (actual) or declared types of objects.
15978 @xref{Print Settings, ,Print Settings}.
15980 @item set print vtbl
15981 @itemx show print vtbl
15982 Control the format for printing virtual function tables.
15983 @xref{Print Settings, ,Print Settings}.
15984 (The @code{vtbl} commands do not work on programs compiled with the HP
15985 ANSI C@t{++} compiler (@code{aCC}).)
15987 @kindex set overload-resolution
15988 @cindex overloaded functions, overload resolution
15989 @item set overload-resolution on
15990 Enable overload resolution for C@t{++} expression evaluation. The default
15991 is on. For overloaded functions, @value{GDBN} evaluates the arguments
15992 and searches for a function whose signature matches the argument types,
15993 using the standard C@t{++} conversion rules (see @ref{C Plus Plus
15994 Expressions, ,C@t{++} Expressions}, for details).
15995 If it cannot find a match, it emits a message.
15997 @item set overload-resolution off
15998 Disable overload resolution for C@t{++} expression evaluation. For
15999 overloaded functions that are not class member functions, @value{GDBN}
16000 chooses the first function of the specified name that it finds in the
16001 symbol table, whether or not its arguments are of the correct type. For
16002 overloaded functions that are class member functions, @value{GDBN}
16003 searches for a function whose signature @emph{exactly} matches the
16006 @kindex show overload-resolution
16007 @item show overload-resolution
16008 Show the current setting of overload resolution.
16010 @item @r{Overloaded symbol names}
16011 You can specify a particular definition of an overloaded symbol, using
16012 the same notation that is used to declare such symbols in C@t{++}: type
16013 @code{@var{symbol}(@var{types})} rather than just @var{symbol}. You can
16014 also use the @value{GDBN} command-line word completion facilities to list the
16015 available choices, or to finish the type list for you.
16016 @xref{Completion,, Command Completion}, for details on how to do this.
16018 @item @r{Breakpoints in functions with ABI tags}
16020 The GNU C@t{++} compiler introduced the notion of ABI ``tags'', which
16021 correspond to changes in the ABI of a type, function, or variable that
16022 would not otherwise be reflected in a mangled name. See
16023 @url{https://developers.redhat.com/blog/2015/02/05/gcc5-and-the-c11-abi/}
16026 The ABI tags are visible in C@t{++} demangled names. For example, a
16027 function that returns a std::string:
16030 std::string function(int);
16034 when compiled for the C++11 ABI is marked with the @code{cxx11} ABI
16035 tag, and @value{GDBN} displays the symbol like this:
16038 function[abi:cxx11](int)
16041 You can set a breakpoint on such functions simply as if they had no
16045 (gdb) b function(int)
16046 Breakpoint 2 at 0x40060d: file main.cc, line 10.
16047 (gdb) info breakpoints
16048 Num Type Disp Enb Address What
16049 1 breakpoint keep y 0x0040060d in function[abi:cxx11](int)
16053 On the rare occasion you need to disambiguate between different ABI
16054 tags, you can do so by simply including the ABI tag in the function
16058 (@value{GDBP}) b ambiguous[abi:other_tag](int)
16062 @node Decimal Floating Point
16063 @subsubsection Decimal Floating Point format
16064 @cindex decimal floating point format
16066 @value{GDBN} can examine, set and perform computations with numbers in
16067 decimal floating point format, which in the C language correspond to the
16068 @code{_Decimal32}, @code{_Decimal64} and @code{_Decimal128} types as
16069 specified by the extension to support decimal floating-point arithmetic.
16071 There are two encodings in use, depending on the architecture: BID (Binary
16072 Integer Decimal) for x86 and x86-64, and DPD (Densely Packed Decimal) for
16073 PowerPC and S/390. @value{GDBN} will use the appropriate encoding for the
16076 Because of a limitation in @file{libdecnumber}, the library used by @value{GDBN}
16077 to manipulate decimal floating point numbers, it is not possible to convert
16078 (using a cast, for example) integers wider than 32-bit to decimal float.
16080 In addition, in order to imitate @value{GDBN}'s behaviour with binary floating
16081 point computations, error checking in decimal float operations ignores
16082 underflow, overflow and divide by zero exceptions.
16084 In the PowerPC architecture, @value{GDBN} provides a set of pseudo-registers
16085 to inspect @code{_Decimal128} values stored in floating point registers.
16086 See @ref{PowerPC,,PowerPC} for more details.
16092 @value{GDBN} can be used to debug programs written in D and compiled with
16093 GDC, LDC or DMD compilers. Currently @value{GDBN} supports only one D
16094 specific feature --- dynamic arrays.
16099 @cindex Go (programming language)
16100 @value{GDBN} can be used to debug programs written in Go and compiled with
16101 @file{gccgo} or @file{6g} compilers.
16103 Here is a summary of the Go-specific features and restrictions:
16106 @cindex current Go package
16107 @item The current Go package
16108 The name of the current package does not need to be specified when
16109 specifying global variables and functions.
16111 For example, given the program:
16115 var myglob = "Shall we?"
16121 When stopped inside @code{main} either of these work:
16125 (gdb) p main.myglob
16128 @cindex builtin Go types
16129 @item Builtin Go types
16130 The @code{string} type is recognized by @value{GDBN} and is printed
16133 @cindex builtin Go functions
16134 @item Builtin Go functions
16135 The @value{GDBN} expression parser recognizes the @code{unsafe.Sizeof}
16136 function and handles it internally.
16138 @cindex restrictions on Go expressions
16139 @item Restrictions on Go expressions
16140 All Go operators are supported except @code{&^}.
16141 The Go @code{_} ``blank identifier'' is not supported.
16142 Automatic dereferencing of pointers is not supported.
16146 @subsection Objective-C
16148 @cindex Objective-C
16149 This section provides information about some commands and command
16150 options that are useful for debugging Objective-C code. See also
16151 @ref{Symbols, info classes}, and @ref{Symbols, info selectors}, for a
16152 few more commands specific to Objective-C support.
16155 * Method Names in Commands::
16156 * The Print Command with Objective-C::
16159 @node Method Names in Commands
16160 @subsubsection Method Names in Commands
16162 The following commands have been extended to accept Objective-C method
16163 names as line specifications:
16165 @kindex clear@r{, and Objective-C}
16166 @kindex break@r{, and Objective-C}
16167 @kindex info line@r{, and Objective-C}
16168 @kindex jump@r{, and Objective-C}
16169 @kindex list@r{, and Objective-C}
16173 @item @code{info line}
16178 A fully qualified Objective-C method name is specified as
16181 -[@var{Class} @var{methodName}]
16184 where the minus sign is used to indicate an instance method and a
16185 plus sign (not shown) is used to indicate a class method. The class
16186 name @var{Class} and method name @var{methodName} are enclosed in
16187 brackets, similar to the way messages are specified in Objective-C
16188 source code. For example, to set a breakpoint at the @code{create}
16189 instance method of class @code{Fruit} in the program currently being
16193 break -[Fruit create]
16196 To list ten program lines around the @code{initialize} class method,
16200 list +[NSText initialize]
16203 In the current version of @value{GDBN}, the plus or minus sign is
16204 required. In future versions of @value{GDBN}, the plus or minus
16205 sign will be optional, but you can use it to narrow the search. It
16206 is also possible to specify just a method name:
16212 You must specify the complete method name, including any colons. If
16213 your program's source files contain more than one @code{create} method,
16214 you'll be presented with a numbered list of classes that implement that
16215 method. Indicate your choice by number, or type @samp{0} to exit if
16218 As another example, to clear a breakpoint established at the
16219 @code{makeKeyAndOrderFront:} method of the @code{NSWindow} class, enter:
16222 clear -[NSWindow makeKeyAndOrderFront:]
16225 @node The Print Command with Objective-C
16226 @subsubsection The Print Command With Objective-C
16227 @cindex Objective-C, print objects
16228 @kindex print-object
16229 @kindex po @r{(@code{print-object})}
16231 The print command has also been extended to accept methods. For example:
16234 print -[@var{object} hash]
16237 @cindex print an Objective-C object description
16238 @cindex @code{_NSPrintForDebugger}, and printing Objective-C objects
16240 will tell @value{GDBN} to send the @code{hash} message to @var{object}
16241 and print the result. Also, an additional command has been added,
16242 @code{print-object} or @code{po} for short, which is meant to print
16243 the description of an object. However, this command may only work
16244 with certain Objective-C libraries that have a particular hook
16245 function, @code{_NSPrintForDebugger}, defined.
16248 @subsection OpenCL C
16251 This section provides information about @value{GDBN}s OpenCL C support.
16254 * OpenCL C Datatypes::
16255 * OpenCL C Expressions::
16256 * OpenCL C Operators::
16259 @node OpenCL C Datatypes
16260 @subsubsection OpenCL C Datatypes
16262 @cindex OpenCL C Datatypes
16263 @value{GDBN} supports the builtin scalar and vector datatypes specified
16264 by OpenCL 1.1. In addition the half- and double-precision floating point
16265 data types of the @code{cl_khr_fp16} and @code{cl_khr_fp64} OpenCL
16266 extensions are also known to @value{GDBN}.
16268 @node OpenCL C Expressions
16269 @subsubsection OpenCL C Expressions
16271 @cindex OpenCL C Expressions
16272 @value{GDBN} supports accesses to vector components including the access as
16273 lvalue where possible. Since OpenCL C is based on C99 most C expressions
16274 supported by @value{GDBN} can be used as well.
16276 @node OpenCL C Operators
16277 @subsubsection OpenCL C Operators
16279 @cindex OpenCL C Operators
16280 @value{GDBN} supports the operators specified by OpenCL 1.1 for scalar and
16284 @subsection Fortran
16285 @cindex Fortran-specific support in @value{GDBN}
16287 @value{GDBN} can be used to debug programs written in Fortran, but it
16288 currently supports only the features of Fortran 77 language.
16290 @cindex trailing underscore, in Fortran symbols
16291 Some Fortran compilers (@sc{gnu} Fortran 77 and Fortran 95 compilers
16292 among them) append an underscore to the names of variables and
16293 functions. When you debug programs compiled by those compilers, you
16294 will need to refer to variables and functions with a trailing
16298 * Fortran Operators:: Fortran operators and expressions
16299 * Fortran Defaults:: Default settings for Fortran
16300 * Special Fortran Commands:: Special @value{GDBN} commands for Fortran
16303 @node Fortran Operators
16304 @subsubsection Fortran Operators and Expressions
16306 @cindex Fortran operators and expressions
16308 Operators must be defined on values of specific types. For instance,
16309 @code{+} is defined on numbers, but not on characters or other non-
16310 arithmetic types. Operators are often defined on groups of types.
16314 The exponentiation operator. It raises the first operand to the power
16318 The range operator. Normally used in the form of array(low:high) to
16319 represent a section of array.
16322 The access component operator. Normally used to access elements in derived
16323 types. Also suitable for unions. As unions aren't part of regular Fortran,
16324 this can only happen when accessing a register that uses a gdbarch-defined
16328 @node Fortran Defaults
16329 @subsubsection Fortran Defaults
16331 @cindex Fortran Defaults
16333 Fortran symbols are usually case-insensitive, so @value{GDBN} by
16334 default uses case-insensitive matches for Fortran symbols. You can
16335 change that with the @samp{set case-insensitive} command, see
16336 @ref{Symbols}, for the details.
16338 @node Special Fortran Commands
16339 @subsubsection Special Fortran Commands
16341 @cindex Special Fortran commands
16343 @value{GDBN} has some commands to support Fortran-specific features,
16344 such as displaying common blocks.
16347 @cindex @code{COMMON} blocks, Fortran
16348 @kindex info common
16349 @item info common @r{[}@var{common-name}@r{]}
16350 This command prints the values contained in the Fortran @code{COMMON}
16351 block whose name is @var{common-name}. With no argument, the names of
16352 all @code{COMMON} blocks visible at the current program location are
16359 @cindex Pascal support in @value{GDBN}, limitations
16360 Debugging Pascal programs which use sets, subranges, file variables, or
16361 nested functions does not currently work. @value{GDBN} does not support
16362 entering expressions, printing values, or similar features using Pascal
16365 The Pascal-specific command @code{set print pascal_static-members}
16366 controls whether static members of Pascal objects are displayed.
16367 @xref{Print Settings, pascal_static-members}.
16372 @value{GDBN} supports the @url{https://www.rust-lang.org/, Rust
16373 Programming Language}. Type- and value-printing, and expression
16374 parsing, are reasonably complete. However, there are a few
16375 peculiarities and holes to be aware of.
16379 Linespecs (@pxref{Specify Location}) are never relative to the current
16380 crate. Instead, they act as if there were a global namespace of
16381 crates, somewhat similar to the way @code{extern crate} behaves.
16383 That is, if @value{GDBN} is stopped at a breakpoint in a function in
16384 crate @samp{A}, module @samp{B}, then @code{break B::f} will attempt
16385 to set a breakpoint in a function named @samp{f} in a crate named
16388 As a consequence of this approach, linespecs also cannot refer to
16389 items using @samp{self::} or @samp{super::}.
16392 Because @value{GDBN} implements Rust name-lookup semantics in
16393 expressions, it will sometimes prepend the current crate to a name.
16394 For example, if @value{GDBN} is stopped at a breakpoint in the crate
16395 @samp{K}, then @code{print ::x::y} will try to find the symbol
16398 However, since it is useful to be able to refer to other crates when
16399 debugging, @value{GDBN} provides the @code{extern} extension to
16400 circumvent this. To use the extension, just put @code{extern} before
16401 a path expression to refer to the otherwise unavailable ``global''
16404 In the above example, if you wanted to refer to the symbol @samp{y} in
16405 the crate @samp{x}, you would use @code{print extern x::y}.
16408 The Rust expression evaluator does not support ``statement-like''
16409 expressions such as @code{if} or @code{match}, or lambda expressions.
16412 Tuple expressions are not implemented.
16415 The Rust expression evaluator does not currently implement the
16416 @code{Drop} trait. Objects that may be created by the evaluator will
16417 never be destroyed.
16420 @value{GDBN} does not implement type inference for generics. In order
16421 to call generic functions or otherwise refer to generic items, you
16422 will have to specify the type parameters manually.
16425 @value{GDBN} currently uses the C@t{++} demangler for Rust. In most
16426 cases this does not cause any problems. However, in an expression
16427 context, completing a generic function name will give syntactically
16428 invalid results. This happens because Rust requires the @samp{::}
16429 operator between the function name and its generic arguments. For
16430 example, @value{GDBN} might provide a completion like
16431 @code{crate::f<u32>}, where the parser would require
16432 @code{crate::f::<u32>}.
16435 As of this writing, the Rust compiler (version 1.8) has a few holes in
16436 the debugging information it generates. These holes prevent certain
16437 features from being implemented by @value{GDBN}:
16441 Method calls cannot be made via traits.
16444 Operator overloading is not implemented.
16447 When debugging in a monomorphized function, you cannot use the generic
16451 The type @code{Self} is not available.
16454 @code{use} statements are not available, so some names may not be
16455 available in the crate.
16460 @subsection Modula-2
16462 @cindex Modula-2, @value{GDBN} support
16464 The extensions made to @value{GDBN} to support Modula-2 only support
16465 output from the @sc{gnu} Modula-2 compiler (which is currently being
16466 developed). Other Modula-2 compilers are not currently supported, and
16467 attempting to debug executables produced by them is most likely
16468 to give an error as @value{GDBN} reads in the executable's symbol
16471 @cindex expressions in Modula-2
16473 * M2 Operators:: Built-in operators
16474 * Built-In Func/Proc:: Built-in functions and procedures
16475 * M2 Constants:: Modula-2 constants
16476 * M2 Types:: Modula-2 types
16477 * M2 Defaults:: Default settings for Modula-2
16478 * Deviations:: Deviations from standard Modula-2
16479 * M2 Checks:: Modula-2 type and range checks
16480 * M2 Scope:: The scope operators @code{::} and @code{.}
16481 * GDB/M2:: @value{GDBN} and Modula-2
16485 @subsubsection Operators
16486 @cindex Modula-2 operators
16488 Operators must be defined on values of specific types. For instance,
16489 @code{+} is defined on numbers, but not on structures. Operators are
16490 often defined on groups of types. For the purposes of Modula-2, the
16491 following definitions hold:
16496 @emph{Integral types} consist of @code{INTEGER}, @code{CARDINAL}, and
16500 @emph{Character types} consist of @code{CHAR} and its subranges.
16503 @emph{Floating-point types} consist of @code{REAL}.
16506 @emph{Pointer types} consist of anything declared as @code{POINTER TO
16510 @emph{Scalar types} consist of all of the above.
16513 @emph{Set types} consist of @code{SET} and @code{BITSET} types.
16516 @emph{Boolean types} consist of @code{BOOLEAN}.
16520 The following operators are supported, and appear in order of
16521 increasing precedence:
16525 Function argument or array index separator.
16528 Assignment. The value of @var{var} @code{:=} @var{value} is
16532 Less than, greater than on integral, floating-point, or enumerated
16536 Less than or equal to, greater than or equal to
16537 on integral, floating-point and enumerated types, or set inclusion on
16538 set types. Same precedence as @code{<}.
16540 @item =@r{, }<>@r{, }#
16541 Equality and two ways of expressing inequality, valid on scalar types.
16542 Same precedence as @code{<}. In @value{GDBN} scripts, only @code{<>} is
16543 available for inequality, since @code{#} conflicts with the script
16547 Set membership. Defined on set types and the types of their members.
16548 Same precedence as @code{<}.
16551 Boolean disjunction. Defined on boolean types.
16554 Boolean conjunction. Defined on boolean types.
16557 The @value{GDBN} ``artificial array'' operator (@pxref{Expressions, ,Expressions}).
16560 Addition and subtraction on integral and floating-point types, or union
16561 and difference on set types.
16564 Multiplication on integral and floating-point types, or set intersection
16568 Division on floating-point types, or symmetric set difference on set
16569 types. Same precedence as @code{*}.
16572 Integer division and remainder. Defined on integral types. Same
16573 precedence as @code{*}.
16576 Negative. Defined on @code{INTEGER} and @code{REAL} data.
16579 Pointer dereferencing. Defined on pointer types.
16582 Boolean negation. Defined on boolean types. Same precedence as
16586 @code{RECORD} field selector. Defined on @code{RECORD} data. Same
16587 precedence as @code{^}.
16590 Array indexing. Defined on @code{ARRAY} data. Same precedence as @code{^}.
16593 Procedure argument list. Defined on @code{PROCEDURE} objects. Same precedence
16597 @value{GDBN} and Modula-2 scope operators.
16601 @emph{Warning:} Set expressions and their operations are not yet supported, so @value{GDBN}
16602 treats the use of the operator @code{IN}, or the use of operators
16603 @code{+}, @code{-}, @code{*}, @code{/}, @code{=}, , @code{<>}, @code{#},
16604 @code{<=}, and @code{>=} on sets as an error.
16608 @node Built-In Func/Proc
16609 @subsubsection Built-in Functions and Procedures
16610 @cindex Modula-2 built-ins
16612 Modula-2 also makes available several built-in procedures and functions.
16613 In describing these, the following metavariables are used:
16618 represents an @code{ARRAY} variable.
16621 represents a @code{CHAR} constant or variable.
16624 represents a variable or constant of integral type.
16627 represents an identifier that belongs to a set. Generally used in the
16628 same function with the metavariable @var{s}. The type of @var{s} should
16629 be @code{SET OF @var{mtype}} (where @var{mtype} is the type of @var{m}).
16632 represents a variable or constant of integral or floating-point type.
16635 represents a variable or constant of floating-point type.
16641 represents a variable.
16644 represents a variable or constant of one of many types. See the
16645 explanation of the function for details.
16648 All Modula-2 built-in procedures also return a result, described below.
16652 Returns the absolute value of @var{n}.
16655 If @var{c} is a lower case letter, it returns its upper case
16656 equivalent, otherwise it returns its argument.
16659 Returns the character whose ordinal value is @var{i}.
16662 Decrements the value in the variable @var{v} by one. Returns the new value.
16664 @item DEC(@var{v},@var{i})
16665 Decrements the value in the variable @var{v} by @var{i}. Returns the
16668 @item EXCL(@var{m},@var{s})
16669 Removes the element @var{m} from the set @var{s}. Returns the new
16672 @item FLOAT(@var{i})
16673 Returns the floating point equivalent of the integer @var{i}.
16675 @item HIGH(@var{a})
16676 Returns the index of the last member of @var{a}.
16679 Increments the value in the variable @var{v} by one. Returns the new value.
16681 @item INC(@var{v},@var{i})
16682 Increments the value in the variable @var{v} by @var{i}. Returns the
16685 @item INCL(@var{m},@var{s})
16686 Adds the element @var{m} to the set @var{s} if it is not already
16687 there. Returns the new set.
16690 Returns the maximum value of the type @var{t}.
16693 Returns the minimum value of the type @var{t}.
16696 Returns boolean TRUE if @var{i} is an odd number.
16699 Returns the ordinal value of its argument. For example, the ordinal
16700 value of a character is its @sc{ascii} value (on machines supporting
16701 the @sc{ascii} character set). The argument @var{x} must be of an
16702 ordered type, which include integral, character and enumerated types.
16704 @item SIZE(@var{x})
16705 Returns the size of its argument. The argument @var{x} can be a
16706 variable or a type.
16708 @item TRUNC(@var{r})
16709 Returns the integral part of @var{r}.
16711 @item TSIZE(@var{x})
16712 Returns the size of its argument. The argument @var{x} can be a
16713 variable or a type.
16715 @item VAL(@var{t},@var{i})
16716 Returns the member of the type @var{t} whose ordinal value is @var{i}.
16720 @emph{Warning:} Sets and their operations are not yet supported, so
16721 @value{GDBN} treats the use of procedures @code{INCL} and @code{EXCL} as
16725 @cindex Modula-2 constants
16727 @subsubsection Constants
16729 @value{GDBN} allows you to express the constants of Modula-2 in the following
16735 Integer constants are simply a sequence of digits. When used in an
16736 expression, a constant is interpreted to be type-compatible with the
16737 rest of the expression. Hexadecimal integers are specified by a
16738 trailing @samp{H}, and octal integers by a trailing @samp{B}.
16741 Floating point constants appear as a sequence of digits, followed by a
16742 decimal point and another sequence of digits. An optional exponent can
16743 then be specified, in the form @samp{E@r{[}+@r{|}-@r{]}@var{nnn}}, where
16744 @samp{@r{[}+@r{|}-@r{]}@var{nnn}} is the desired exponent. All of the
16745 digits of the floating point constant must be valid decimal (base 10)
16749 Character constants consist of a single character enclosed by a pair of
16750 like quotes, either single (@code{'}) or double (@code{"}). They may
16751 also be expressed by their ordinal value (their @sc{ascii} value, usually)
16752 followed by a @samp{C}.
16755 String constants consist of a sequence of characters enclosed by a
16756 pair of like quotes, either single (@code{'}) or double (@code{"}).
16757 Escape sequences in the style of C are also allowed. @xref{C
16758 Constants, ,C and C@t{++} Constants}, for a brief explanation of escape
16762 Enumerated constants consist of an enumerated identifier.
16765 Boolean constants consist of the identifiers @code{TRUE} and
16769 Pointer constants consist of integral values only.
16772 Set constants are not yet supported.
16776 @subsubsection Modula-2 Types
16777 @cindex Modula-2 types
16779 Currently @value{GDBN} can print the following data types in Modula-2
16780 syntax: array types, record types, set types, pointer types, procedure
16781 types, enumerated types, subrange types and base types. You can also
16782 print the contents of variables declared using these type.
16783 This section gives a number of simple source code examples together with
16784 sample @value{GDBN} sessions.
16786 The first example contains the following section of code:
16795 and you can request @value{GDBN} to interrogate the type and value of
16796 @code{r} and @code{s}.
16799 (@value{GDBP}) print s
16801 (@value{GDBP}) ptype s
16803 (@value{GDBP}) print r
16805 (@value{GDBP}) ptype r
16810 Likewise if your source code declares @code{s} as:
16814 s: SET ['A'..'Z'] ;
16818 then you may query the type of @code{s} by:
16821 (@value{GDBP}) ptype s
16822 type = SET ['A'..'Z']
16826 Note that at present you cannot interactively manipulate set
16827 expressions using the debugger.
16829 The following example shows how you might declare an array in Modula-2
16830 and how you can interact with @value{GDBN} to print its type and contents:
16834 s: ARRAY [-10..10] OF CHAR ;
16838 (@value{GDBP}) ptype s
16839 ARRAY [-10..10] OF CHAR
16842 Note that the array handling is not yet complete and although the type
16843 is printed correctly, expression handling still assumes that all
16844 arrays have a lower bound of zero and not @code{-10} as in the example
16847 Here are some more type related Modula-2 examples:
16851 colour = (blue, red, yellow, green) ;
16852 t = [blue..yellow] ;
16860 The @value{GDBN} interaction shows how you can query the data type
16861 and value of a variable.
16864 (@value{GDBP}) print s
16866 (@value{GDBP}) ptype t
16867 type = [blue..yellow]
16871 In this example a Modula-2 array is declared and its contents
16872 displayed. Observe that the contents are written in the same way as
16873 their @code{C} counterparts.
16877 s: ARRAY [1..5] OF CARDINAL ;
16883 (@value{GDBP}) print s
16884 $1 = @{1, 0, 0, 0, 0@}
16885 (@value{GDBP}) ptype s
16886 type = ARRAY [1..5] OF CARDINAL
16889 The Modula-2 language interface to @value{GDBN} also understands
16890 pointer types as shown in this example:
16894 s: POINTER TO ARRAY [1..5] OF CARDINAL ;
16901 and you can request that @value{GDBN} describes the type of @code{s}.
16904 (@value{GDBP}) ptype s
16905 type = POINTER TO ARRAY [1..5] OF CARDINAL
16908 @value{GDBN} handles compound types as we can see in this example.
16909 Here we combine array types, record types, pointer types and subrange
16920 myarray = ARRAY myrange OF CARDINAL ;
16921 myrange = [-2..2] ;
16923 s: POINTER TO ARRAY myrange OF foo ;
16927 and you can ask @value{GDBN} to describe the type of @code{s} as shown
16931 (@value{GDBP}) ptype s
16932 type = POINTER TO ARRAY [-2..2] OF foo = RECORD
16935 f3 : ARRAY [-2..2] OF CARDINAL;
16940 @subsubsection Modula-2 Defaults
16941 @cindex Modula-2 defaults
16943 If type and range checking are set automatically by @value{GDBN}, they
16944 both default to @code{on} whenever the working language changes to
16945 Modula-2. This happens regardless of whether you or @value{GDBN}
16946 selected the working language.
16948 If you allow @value{GDBN} to set the language automatically, then entering
16949 code compiled from a file whose name ends with @file{.mod} sets the
16950 working language to Modula-2. @xref{Automatically, ,Having @value{GDBN}
16951 Infer the Source Language}, for further details.
16954 @subsubsection Deviations from Standard Modula-2
16955 @cindex Modula-2, deviations from
16957 A few changes have been made to make Modula-2 programs easier to debug.
16958 This is done primarily via loosening its type strictness:
16962 Unlike in standard Modula-2, pointer constants can be formed by
16963 integers. This allows you to modify pointer variables during
16964 debugging. (In standard Modula-2, the actual address contained in a
16965 pointer variable is hidden from you; it can only be modified
16966 through direct assignment to another pointer variable or expression that
16967 returned a pointer.)
16970 C escape sequences can be used in strings and characters to represent
16971 non-printable characters. @value{GDBN} prints out strings with these
16972 escape sequences embedded. Single non-printable characters are
16973 printed using the @samp{CHR(@var{nnn})} format.
16976 The assignment operator (@code{:=}) returns the value of its right-hand
16980 All built-in procedures both modify @emph{and} return their argument.
16984 @subsubsection Modula-2 Type and Range Checks
16985 @cindex Modula-2 checks
16988 @emph{Warning:} in this release, @value{GDBN} does not yet perform type or
16991 @c FIXME remove warning when type/range checks added
16993 @value{GDBN} considers two Modula-2 variables type equivalent if:
16997 They are of types that have been declared equivalent via a @code{TYPE
16998 @var{t1} = @var{t2}} statement
17001 They have been declared on the same line. (Note: This is true of the
17002 @sc{gnu} Modula-2 compiler, but it may not be true of other compilers.)
17005 As long as type checking is enabled, any attempt to combine variables
17006 whose types are not equivalent is an error.
17008 Range checking is done on all mathematical operations, assignment, array
17009 index bounds, and all built-in functions and procedures.
17012 @subsubsection The Scope Operators @code{::} and @code{.}
17014 @cindex @code{.}, Modula-2 scope operator
17015 @cindex colon, doubled as scope operator
17017 @vindex colon-colon@r{, in Modula-2}
17018 @c Info cannot handle :: but TeX can.
17021 @vindex ::@r{, in Modula-2}
17024 There are a few subtle differences between the Modula-2 scope operator
17025 (@code{.}) and the @value{GDBN} scope operator (@code{::}). The two have
17030 @var{module} . @var{id}
17031 @var{scope} :: @var{id}
17035 where @var{scope} is the name of a module or a procedure,
17036 @var{module} the name of a module, and @var{id} is any declared
17037 identifier within your program, except another module.
17039 Using the @code{::} operator makes @value{GDBN} search the scope
17040 specified by @var{scope} for the identifier @var{id}. If it is not
17041 found in the specified scope, then @value{GDBN} searches all scopes
17042 enclosing the one specified by @var{scope}.
17044 Using the @code{.} operator makes @value{GDBN} search the current scope for
17045 the identifier specified by @var{id} that was imported from the
17046 definition module specified by @var{module}. With this operator, it is
17047 an error if the identifier @var{id} was not imported from definition
17048 module @var{module}, or if @var{id} is not an identifier in
17052 @subsubsection @value{GDBN} and Modula-2
17054 Some @value{GDBN} commands have little use when debugging Modula-2 programs.
17055 Five subcommands of @code{set print} and @code{show print} apply
17056 specifically to C and C@t{++}: @samp{vtbl}, @samp{demangle},
17057 @samp{asm-demangle}, @samp{object}, and @samp{union}. The first four
17058 apply to C@t{++}, and the last to the C @code{union} type, which has no direct
17059 analogue in Modula-2.
17061 The @code{@@} operator (@pxref{Expressions, ,Expressions}), while available
17062 with any language, is not useful with Modula-2. Its
17063 intent is to aid the debugging of @dfn{dynamic arrays}, which cannot be
17064 created in Modula-2 as they can in C or C@t{++}. However, because an
17065 address can be specified by an integral constant, the construct
17066 @samp{@{@var{type}@}@var{adrexp}} is still useful.
17068 @cindex @code{#} in Modula-2
17069 In @value{GDBN} scripts, the Modula-2 inequality operator @code{#} is
17070 interpreted as the beginning of a comment. Use @code{<>} instead.
17076 The extensions made to @value{GDBN} for Ada only support
17077 output from the @sc{gnu} Ada (GNAT) compiler.
17078 Other Ada compilers are not currently supported, and
17079 attempting to debug executables produced by them is most likely
17083 @cindex expressions in Ada
17085 * Ada Mode Intro:: General remarks on the Ada syntax
17086 and semantics supported by Ada mode
17088 * Omissions from Ada:: Restrictions on the Ada expression syntax.
17089 * Additions to Ada:: Extensions of the Ada expression syntax.
17090 * Overloading support for Ada:: Support for expressions involving overloaded
17092 * Stopping Before Main Program:: Debugging the program during elaboration.
17093 * Ada Exceptions:: Ada Exceptions
17094 * Ada Tasks:: Listing and setting breakpoints in tasks.
17095 * Ada Tasks and Core Files:: Tasking Support when Debugging Core Files
17096 * Ravenscar Profile:: Tasking Support when using the Ravenscar
17098 * Ada Settings:: New settable GDB parameters for Ada.
17099 * Ada Glitches:: Known peculiarities of Ada mode.
17102 @node Ada Mode Intro
17103 @subsubsection Introduction
17104 @cindex Ada mode, general
17106 The Ada mode of @value{GDBN} supports a fairly large subset of Ada expression
17107 syntax, with some extensions.
17108 The philosophy behind the design of this subset is
17112 That @value{GDBN} should provide basic literals and access to operations for
17113 arithmetic, dereferencing, field selection, indexing, and subprogram calls,
17114 leaving more sophisticated computations to subprograms written into the
17115 program (which therefore may be called from @value{GDBN}).
17118 That type safety and strict adherence to Ada language restrictions
17119 are not particularly important to the @value{GDBN} user.
17122 That brevity is important to the @value{GDBN} user.
17125 Thus, for brevity, the debugger acts as if all names declared in
17126 user-written packages are directly visible, even if they are not visible
17127 according to Ada rules, thus making it unnecessary to fully qualify most
17128 names with their packages, regardless of context. Where this causes
17129 ambiguity, @value{GDBN} asks the user's intent.
17131 The debugger will start in Ada mode if it detects an Ada main program.
17132 As for other languages, it will enter Ada mode when stopped in a program that
17133 was translated from an Ada source file.
17135 While in Ada mode, you may use `@t{--}' for comments. This is useful
17136 mostly for documenting command files. The standard @value{GDBN} comment
17137 (@samp{#}) still works at the beginning of a line in Ada mode, but not in the
17138 middle (to allow based literals).
17140 @node Omissions from Ada
17141 @subsubsection Omissions from Ada
17142 @cindex Ada, omissions from
17144 Here are the notable omissions from the subset:
17148 Only a subset of the attributes are supported:
17152 @t{'First}, @t{'Last}, and @t{'Length}
17153 on array objects (not on types and subtypes).
17156 @t{'Min} and @t{'Max}.
17159 @t{'Pos} and @t{'Val}.
17165 @t{'Range} on array objects (not subtypes), but only as the right
17166 operand of the membership (@code{in}) operator.
17169 @t{'Access}, @t{'Unchecked_Access}, and
17170 @t{'Unrestricted_Access} (a GNAT extension).
17178 @code{Characters.Latin_1} are not available and
17179 concatenation is not implemented. Thus, escape characters in strings are
17180 not currently available.
17183 Equality tests (@samp{=} and @samp{/=}) on arrays test for bitwise
17184 equality of representations. They will generally work correctly
17185 for strings and arrays whose elements have integer or enumeration types.
17186 They may not work correctly for arrays whose element
17187 types have user-defined equality, for arrays of real values
17188 (in particular, IEEE-conformant floating point, because of negative
17189 zeroes and NaNs), and for arrays whose elements contain unused bits with
17190 indeterminate values.
17193 The other component-by-component array operations (@code{and}, @code{or},
17194 @code{xor}, @code{not}, and relational tests other than equality)
17195 are not implemented.
17198 @cindex array aggregates (Ada)
17199 @cindex record aggregates (Ada)
17200 @cindex aggregates (Ada)
17201 There is limited support for array and record aggregates. They are
17202 permitted only on the right sides of assignments, as in these examples:
17205 (@value{GDBP}) set An_Array := (1, 2, 3, 4, 5, 6)
17206 (@value{GDBP}) set An_Array := (1, others => 0)
17207 (@value{GDBP}) set An_Array := (0|4 => 1, 1..3 => 2, 5 => 6)
17208 (@value{GDBP}) set A_2D_Array := ((1, 2, 3), (4, 5, 6), (7, 8, 9))
17209 (@value{GDBP}) set A_Record := (1, "Peter", True);
17210 (@value{GDBP}) set A_Record := (Name => "Peter", Id => 1, Alive => True)
17214 discriminant's value by assigning an aggregate has an
17215 undefined effect if that discriminant is used within the record.
17216 However, you can first modify discriminants by directly assigning to
17217 them (which normally would not be allowed in Ada), and then performing an
17218 aggregate assignment. For example, given a variable @code{A_Rec}
17219 declared to have a type such as:
17222 type Rec (Len : Small_Integer := 0) is record
17224 Vals : IntArray (1 .. Len);
17228 you can assign a value with a different size of @code{Vals} with two
17232 (@value{GDBP}) set A_Rec.Len := 4
17233 (@value{GDBP}) set A_Rec := (Id => 42, Vals => (1, 2, 3, 4))
17236 As this example also illustrates, @value{GDBN} is very loose about the usual
17237 rules concerning aggregates. You may leave out some of the
17238 components of an array or record aggregate (such as the @code{Len}
17239 component in the assignment to @code{A_Rec} above); they will retain their
17240 original values upon assignment. You may freely use dynamic values as
17241 indices in component associations. You may even use overlapping or
17242 redundant component associations, although which component values are
17243 assigned in such cases is not defined.
17246 Calls to dispatching subprograms are not implemented.
17249 The overloading algorithm is much more limited (i.e., less selective)
17250 than that of real Ada. It makes only limited use of the context in
17251 which a subexpression appears to resolve its meaning, and it is much
17252 looser in its rules for allowing type matches. As a result, some
17253 function calls will be ambiguous, and the user will be asked to choose
17254 the proper resolution.
17257 The @code{new} operator is not implemented.
17260 Entry calls are not implemented.
17263 Aside from printing, arithmetic operations on the native VAX floating-point
17264 formats are not supported.
17267 It is not possible to slice a packed array.
17270 The names @code{True} and @code{False}, when not part of a qualified name,
17271 are interpreted as if implicitly prefixed by @code{Standard}, regardless of
17273 Should your program
17274 redefine these names in a package or procedure (at best a dubious practice),
17275 you will have to use fully qualified names to access their new definitions.
17278 @node Additions to Ada
17279 @subsubsection Additions to Ada
17280 @cindex Ada, deviations from
17282 As it does for other languages, @value{GDBN} makes certain generic
17283 extensions to Ada (@pxref{Expressions}):
17287 If the expression @var{E} is a variable residing in memory (typically
17288 a local variable or array element) and @var{N} is a positive integer,
17289 then @code{@var{E}@@@var{N}} displays the values of @var{E} and the
17290 @var{N}-1 adjacent variables following it in memory as an array. In
17291 Ada, this operator is generally not necessary, since its prime use is
17292 in displaying parts of an array, and slicing will usually do this in
17293 Ada. However, there are occasional uses when debugging programs in
17294 which certain debugging information has been optimized away.
17297 @code{@var{B}::@var{var}} means ``the variable named @var{var} that
17298 appears in function or file @var{B}.'' When @var{B} is a file name,
17299 you must typically surround it in single quotes.
17302 The expression @code{@{@var{type}@} @var{addr}} means ``the variable of type
17303 @var{type} that appears at address @var{addr}.''
17306 A name starting with @samp{$} is a convenience variable
17307 (@pxref{Convenience Vars}) or a machine register (@pxref{Registers}).
17310 In addition, @value{GDBN} provides a few other shortcuts and outright
17311 additions specific to Ada:
17315 The assignment statement is allowed as an expression, returning
17316 its right-hand operand as its value. Thus, you may enter
17319 (@value{GDBP}) set x := y + 3
17320 (@value{GDBP}) print A(tmp := y + 1)
17324 The semicolon is allowed as an ``operator,'' returning as its value
17325 the value of its right-hand operand.
17326 This allows, for example,
17327 complex conditional breaks:
17330 (@value{GDBP}) break f
17331 (@value{GDBP}) condition 1 (report(i); k += 1; A(k) > 100)
17335 Rather than use catenation and symbolic character names to introduce special
17336 characters into strings, one may instead use a special bracket notation,
17337 which is also used to print strings. A sequence of characters of the form
17338 @samp{["@var{XX}"]} within a string or character literal denotes the
17339 (single) character whose numeric encoding is @var{XX} in hexadecimal. The
17340 sequence of characters @samp{["""]} also denotes a single quotation mark
17341 in strings. For example,
17343 "One line.["0a"]Next line.["0a"]"
17346 contains an ASCII newline character (@code{Ada.Characters.Latin_1.LF})
17350 The subtype used as a prefix for the attributes @t{'Pos}, @t{'Min}, and
17351 @t{'Max} is optional (and is ignored in any case). For example, it is valid
17355 (@value{GDBP}) print 'max(x, y)
17359 When printing arrays, @value{GDBN} uses positional notation when the
17360 array has a lower bound of 1, and uses a modified named notation otherwise.
17361 For example, a one-dimensional array of three integers with a lower bound
17362 of 3 might print as
17369 That is, in contrast to valid Ada, only the first component has a @code{=>}
17373 You may abbreviate attributes in expressions with any unique,
17374 multi-character subsequence of
17375 their names (an exact match gets preference).
17376 For example, you may use @t{a'len}, @t{a'gth}, or @t{a'lh}
17377 in place of @t{a'length}.
17380 @cindex quoting Ada internal identifiers
17381 Since Ada is case-insensitive, the debugger normally maps identifiers you type
17382 to lower case. The GNAT compiler uses upper-case characters for
17383 some of its internal identifiers, which are normally of no interest to users.
17384 For the rare occasions when you actually have to look at them,
17385 enclose them in angle brackets to avoid the lower-case mapping.
17388 (@value{GDBP}) print <JMPBUF_SAVE>[0]
17392 Printing an object of class-wide type or dereferencing an
17393 access-to-class-wide value will display all the components of the object's
17394 specific type (as indicated by its run-time tag). Likewise, component
17395 selection on such a value will operate on the specific type of the
17400 @node Overloading support for Ada
17401 @subsubsection Overloading support for Ada
17402 @cindex overloading, Ada
17404 The debugger supports limited overloading. Given a subprogram call in which
17405 the function symbol has multiple definitions, it will use the number of
17406 actual parameters and some information about their types to attempt to narrow
17407 the set of definitions. It also makes very limited use of context, preferring
17408 procedures to functions in the context of the @code{call} command, and
17409 functions to procedures elsewhere.
17411 If, after narrowing, the set of matching definitions still contains more than
17412 one definition, @value{GDBN} will display a menu to query which one it should
17416 (@value{GDBP}) print f(1)
17417 Multiple matches for f
17419 [1] foo.f (integer) return boolean at foo.adb:23
17420 [2] foo.f (foo.new_integer) return boolean at foo.adb:28
17424 In this case, just select one menu entry either to cancel expression evaluation
17425 (type @kbd{0} and press @key{RET}) or to continue evaluation with a specific
17426 instance (type the corresponding number and press @key{RET}).
17428 Here are a couple of commands to customize @value{GDBN}'s behavior in this
17433 @kindex set ada print-signatures
17434 @item set ada print-signatures
17435 Control whether parameter types and return types are displayed in overloads
17436 selection menus. It is @code{on} by default.
17437 @xref{Overloading support for Ada}.
17439 @kindex show ada print-signatures
17440 @item show ada print-signatures
17441 Show the current setting for displaying parameter types and return types in
17442 overloads selection menu.
17443 @xref{Overloading support for Ada}.
17447 @node Stopping Before Main Program
17448 @subsubsection Stopping at the Very Beginning
17450 @cindex breakpointing Ada elaboration code
17451 It is sometimes necessary to debug the program during elaboration, and
17452 before reaching the main procedure.
17453 As defined in the Ada Reference
17454 Manual, the elaboration code is invoked from a procedure called
17455 @code{adainit}. To run your program up to the beginning of
17456 elaboration, simply use the following two commands:
17457 @code{tbreak adainit} and @code{run}.
17459 @node Ada Exceptions
17460 @subsubsection Ada Exceptions
17462 A command is provided to list all Ada exceptions:
17465 @kindex info exceptions
17466 @item info exceptions
17467 @itemx info exceptions @var{regexp}
17468 The @code{info exceptions} command allows you to list all Ada exceptions
17469 defined within the program being debugged, as well as their addresses.
17470 With a regular expression, @var{regexp}, as argument, only those exceptions
17471 whose names match @var{regexp} are listed.
17474 Below is a small example, showing how the command can be used, first
17475 without argument, and next with a regular expression passed as an
17479 (@value{GDBP}) info exceptions
17480 All defined Ada exceptions:
17481 constraint_error: 0x613da0
17482 program_error: 0x613d20
17483 storage_error: 0x613ce0
17484 tasking_error: 0x613ca0
17485 const.aint_global_e: 0x613b00
17486 (@value{GDBP}) info exceptions const.aint
17487 All Ada exceptions matching regular expression "const.aint":
17488 constraint_error: 0x613da0
17489 const.aint_global_e: 0x613b00
17492 It is also possible to ask @value{GDBN} to stop your program's execution
17493 when an exception is raised. For more details, see @ref{Set Catchpoints}.
17496 @subsubsection Extensions for Ada Tasks
17497 @cindex Ada, tasking
17499 Support for Ada tasks is analogous to that for threads (@pxref{Threads}).
17500 @value{GDBN} provides the following task-related commands:
17505 This command shows a list of current Ada tasks, as in the following example:
17512 (@value{GDBP}) info tasks
17513 ID TID P-ID Pri State Name
17514 1 8088000 0 15 Child Activation Wait main_task
17515 2 80a4000 1 15 Accept Statement b
17516 3 809a800 1 15 Child Activation Wait a
17517 * 4 80ae800 3 15 Runnable c
17522 In this listing, the asterisk before the last task indicates it to be the
17523 task currently being inspected.
17527 Represents @value{GDBN}'s internal task number.
17533 The parent's task ID (@value{GDBN}'s internal task number).
17536 The base priority of the task.
17539 Current state of the task.
17543 The task has been created but has not been activated. It cannot be
17547 The task is not blocked for any reason known to Ada. (It may be waiting
17548 for a mutex, though.) It is conceptually "executing" in normal mode.
17551 The task is terminated, in the sense of ARM 9.3 (5). Any dependents
17552 that were waiting on terminate alternatives have been awakened and have
17553 terminated themselves.
17555 @item Child Activation Wait
17556 The task is waiting for created tasks to complete activation.
17558 @item Accept Statement
17559 The task is waiting on an accept or selective wait statement.
17561 @item Waiting on entry call
17562 The task is waiting on an entry call.
17564 @item Async Select Wait
17565 The task is waiting to start the abortable part of an asynchronous
17569 The task is waiting on a select statement with only a delay
17572 @item Child Termination Wait
17573 The task is sleeping having completed a master within itself, and is
17574 waiting for the tasks dependent on that master to become terminated or
17575 waiting on a terminate Phase.
17577 @item Wait Child in Term Alt
17578 The task is sleeping waiting for tasks on terminate alternatives to
17579 finish terminating.
17581 @item Accepting RV with @var{taskno}
17582 The task is accepting a rendez-vous with the task @var{taskno}.
17586 Name of the task in the program.
17590 @kindex info task @var{taskno}
17591 @item info task @var{taskno}
17592 This command shows detailled informations on the specified task, as in
17593 the following example:
17598 (@value{GDBP}) info tasks
17599 ID TID P-ID Pri State Name
17600 1 8077880 0 15 Child Activation Wait main_task
17601 * 2 807c468 1 15 Runnable task_1
17602 (@value{GDBP}) info task 2
17603 Ada Task: 0x807c468
17607 Parent: 1 (main_task)
17613 @kindex task@r{ (Ada)}
17614 @cindex current Ada task ID
17615 This command prints the ID of the current task.
17621 (@value{GDBP}) info tasks
17622 ID TID P-ID Pri State Name
17623 1 8077870 0 15 Child Activation Wait main_task
17624 * 2 807c458 1 15 Runnable t
17625 (@value{GDBP}) task
17626 [Current task is 2]
17629 @item task @var{taskno}
17630 @cindex Ada task switching
17631 This command is like the @code{thread @var{thread-id}}
17632 command (@pxref{Threads}). It switches the context of debugging
17633 from the current task to the given task.
17639 (@value{GDBP}) info tasks
17640 ID TID P-ID Pri State Name
17641 1 8077870 0 15 Child Activation Wait main_task
17642 * 2 807c458 1 15 Runnable t
17643 (@value{GDBP}) task 1
17644 [Switching to task 1]
17645 #0 0x8067726 in pthread_cond_wait ()
17647 #0 0x8067726 in pthread_cond_wait ()
17648 #1 0x8056714 in system.os_interface.pthread_cond_wait ()
17649 #2 0x805cb63 in system.task_primitives.operations.sleep ()
17650 #3 0x806153e in system.tasking.stages.activate_tasks ()
17651 #4 0x804aacc in un () at un.adb:5
17654 @item break @var{location} task @var{taskno}
17655 @itemx break @var{location} task @var{taskno} if @dots{}
17656 @cindex breakpoints and tasks, in Ada
17657 @cindex task breakpoints, in Ada
17658 @kindex break @dots{} task @var{taskno}@r{ (Ada)}
17659 These commands are like the @code{break @dots{} thread @dots{}}
17660 command (@pxref{Thread Stops}). The
17661 @var{location} argument specifies source lines, as described
17662 in @ref{Specify Location}.
17664 Use the qualifier @samp{task @var{taskno}} with a breakpoint command
17665 to specify that you only want @value{GDBN} to stop the program when a
17666 particular Ada task reaches this breakpoint. The @var{taskno} is one of the
17667 numeric task identifiers assigned by @value{GDBN}, shown in the first
17668 column of the @samp{info tasks} display.
17670 If you do not specify @samp{task @var{taskno}} when you set a
17671 breakpoint, the breakpoint applies to @emph{all} tasks of your
17674 You can use the @code{task} qualifier on conditional breakpoints as
17675 well; in this case, place @samp{task @var{taskno}} before the
17676 breakpoint condition (before the @code{if}).
17684 (@value{GDBP}) info tasks
17685 ID TID P-ID Pri State Name
17686 1 140022020 0 15 Child Activation Wait main_task
17687 2 140045060 1 15 Accept/Select Wait t2
17688 3 140044840 1 15 Runnable t1
17689 * 4 140056040 1 15 Runnable t3
17690 (@value{GDBP}) b 15 task 2
17691 Breakpoint 5 at 0x120044cb0: file test_task_debug.adb, line 15.
17692 (@value{GDBP}) cont
17697 Breakpoint 5, test_task_debug () at test_task_debug.adb:15
17699 (@value{GDBP}) info tasks
17700 ID TID P-ID Pri State Name
17701 1 140022020 0 15 Child Activation Wait main_task
17702 * 2 140045060 1 15 Runnable t2
17703 3 140044840 1 15 Runnable t1
17704 4 140056040 1 15 Delay Sleep t3
17708 @node Ada Tasks and Core Files
17709 @subsubsection Tasking Support when Debugging Core Files
17710 @cindex Ada tasking and core file debugging
17712 When inspecting a core file, as opposed to debugging a live program,
17713 tasking support may be limited or even unavailable, depending on
17714 the platform being used.
17715 For instance, on x86-linux, the list of tasks is available, but task
17716 switching is not supported.
17718 On certain platforms, the debugger needs to perform some
17719 memory writes in order to provide Ada tasking support. When inspecting
17720 a core file, this means that the core file must be opened with read-write
17721 privileges, using the command @samp{"set write on"} (@pxref{Patching}).
17722 Under these circumstances, you should make a backup copy of the core
17723 file before inspecting it with @value{GDBN}.
17725 @node Ravenscar Profile
17726 @subsubsection Tasking Support when using the Ravenscar Profile
17727 @cindex Ravenscar Profile
17729 The @dfn{Ravenscar Profile} is a subset of the Ada tasking features,
17730 specifically designed for systems with safety-critical real-time
17734 @kindex set ravenscar task-switching on
17735 @cindex task switching with program using Ravenscar Profile
17736 @item set ravenscar task-switching on
17737 Allows task switching when debugging a program that uses the Ravenscar
17738 Profile. This is the default.
17740 @kindex set ravenscar task-switching off
17741 @item set ravenscar task-switching off
17742 Turn off task switching when debugging a program that uses the Ravenscar
17743 Profile. This is mostly intended to disable the code that adds support
17744 for the Ravenscar Profile, in case a bug in either @value{GDBN} or in
17745 the Ravenscar runtime is preventing @value{GDBN} from working properly.
17746 To be effective, this command should be run before the program is started.
17748 @kindex show ravenscar task-switching
17749 @item show ravenscar task-switching
17750 Show whether it is possible to switch from task to task in a program
17751 using the Ravenscar Profile.
17756 @subsubsection Ada Settings
17757 @cindex Ada settings
17760 @kindex set varsize-limit
17761 @item set varsize-limit @var{size}
17762 Prevent @value{GDBN} from attempting to evaluate objects whose size
17763 is above the given limit (@var{size}) when those sizes are computed
17764 from run-time quantities. This is typically the case when the object
17765 has a variable size, such as an array whose bounds are not known at
17766 compile time for example. Setting @var{size} to @code{unlimited}
17767 removes the size limitation. By default, the limit is about 65KB.
17769 The purpose of having such a limit is to prevent @value{GDBN} from
17770 trying to grab enormous chunks of virtual memory when asked to evaluate
17771 a quantity whose bounds have been corrupted or have not yet been fully
17772 initialized. The limit applies to the results of some subexpressions
17773 as well as to complete expressions. For example, an expression denoting
17774 a simple integer component, such as @code{x.y.z}, may fail if the size of
17775 @code{x.y} is variable and exceeds @code{size}. On the other hand,
17776 @value{GDBN} is sometimes clever; the expression @code{A(i)}, where
17777 @code{A} is an array variable with non-constant size, will generally
17778 succeed regardless of the bounds on @code{A}, as long as the component
17779 size is less than @var{size}.
17781 @kindex show varsize-limit
17782 @item show varsize-limit
17783 Show the limit on types whose size is determined by run-time quantities.
17787 @subsubsection Known Peculiarities of Ada Mode
17788 @cindex Ada, problems
17790 Besides the omissions listed previously (@pxref{Omissions from Ada}),
17791 we know of several problems with and limitations of Ada mode in
17793 some of which will be fixed with planned future releases of the debugger
17794 and the GNU Ada compiler.
17798 Static constants that the compiler chooses not to materialize as objects in
17799 storage are invisible to the debugger.
17802 Named parameter associations in function argument lists are ignored (the
17803 argument lists are treated as positional).
17806 Many useful library packages are currently invisible to the debugger.
17809 Fixed-point arithmetic, conversions, input, and output is carried out using
17810 floating-point arithmetic, and may give results that only approximate those on
17814 The GNAT compiler never generates the prefix @code{Standard} for any of
17815 the standard symbols defined by the Ada language. @value{GDBN} knows about
17816 this: it will strip the prefix from names when you use it, and will never
17817 look for a name you have so qualified among local symbols, nor match against
17818 symbols in other packages or subprograms. If you have
17819 defined entities anywhere in your program other than parameters and
17820 local variables whose simple names match names in @code{Standard},
17821 GNAT's lack of qualification here can cause confusion. When this happens,
17822 you can usually resolve the confusion
17823 by qualifying the problematic names with package
17824 @code{Standard} explicitly.
17827 Older versions of the compiler sometimes generate erroneous debugging
17828 information, resulting in the debugger incorrectly printing the value
17829 of affected entities. In some cases, the debugger is able to work
17830 around an issue automatically. In other cases, the debugger is able
17831 to work around the issue, but the work-around has to be specifically
17834 @kindex set ada trust-PAD-over-XVS
17835 @kindex show ada trust-PAD-over-XVS
17838 @item set ada trust-PAD-over-XVS on
17839 Configure GDB to strictly follow the GNAT encoding when computing the
17840 value of Ada entities, particularly when @code{PAD} and @code{PAD___XVS}
17841 types are involved (see @code{ada/exp_dbug.ads} in the GCC sources for
17842 a complete description of the encoding used by the GNAT compiler).
17843 This is the default.
17845 @item set ada trust-PAD-over-XVS off
17846 This is related to the encoding using by the GNAT compiler. If @value{GDBN}
17847 sometimes prints the wrong value for certain entities, changing @code{ada
17848 trust-PAD-over-XVS} to @code{off} activates a work-around which may fix
17849 the issue. It is always safe to set @code{ada trust-PAD-over-XVS} to
17850 @code{off}, but this incurs a slight performance penalty, so it is
17851 recommended to leave this setting to @code{on} unless necessary.
17855 @cindex GNAT descriptive types
17856 @cindex GNAT encoding
17857 Internally, the debugger also relies on the compiler following a number
17858 of conventions known as the @samp{GNAT Encoding}, all documented in
17859 @file{gcc/ada/exp_dbug.ads} in the GCC sources. This encoding describes
17860 how the debugging information should be generated for certain types.
17861 In particular, this convention makes use of @dfn{descriptive types},
17862 which are artificial types generated purely to help the debugger.
17864 These encodings were defined at a time when the debugging information
17865 format used was not powerful enough to describe some of the more complex
17866 types available in Ada. Since DWARF allows us to express nearly all
17867 Ada features, the long-term goal is to slowly replace these descriptive
17868 types by their pure DWARF equivalent. To facilitate that transition,
17869 a new maintenance option is available to force the debugger to ignore
17870 those descriptive types. It allows the user to quickly evaluate how
17871 well @value{GDBN} works without them.
17875 @kindex maint ada set ignore-descriptive-types
17876 @item maintenance ada set ignore-descriptive-types [on|off]
17877 Control whether the debugger should ignore descriptive types.
17878 The default is not to ignore descriptives types (@code{off}).
17880 @kindex maint ada show ignore-descriptive-types
17881 @item maintenance ada show ignore-descriptive-types
17882 Show if descriptive types are ignored by @value{GDBN}.
17886 @node Unsupported Languages
17887 @section Unsupported Languages
17889 @cindex unsupported languages
17890 @cindex minimal language
17891 In addition to the other fully-supported programming languages,
17892 @value{GDBN} also provides a pseudo-language, called @code{minimal}.
17893 It does not represent a real programming language, but provides a set
17894 of capabilities close to what the C or assembly languages provide.
17895 This should allow most simple operations to be performed while debugging
17896 an application that uses a language currently not supported by @value{GDBN}.
17898 If the language is set to @code{auto}, @value{GDBN} will automatically
17899 select this language if the current frame corresponds to an unsupported
17903 @chapter Examining the Symbol Table
17905 The commands described in this chapter allow you to inquire about the
17906 symbols (names of variables, functions and types) defined in your
17907 program. This information is inherent in the text of your program and
17908 does not change as your program executes. @value{GDBN} finds it in your
17909 program's symbol table, in the file indicated when you started @value{GDBN}
17910 (@pxref{File Options, ,Choosing Files}), or by one of the
17911 file-management commands (@pxref{Files, ,Commands to Specify Files}).
17913 @cindex symbol names
17914 @cindex names of symbols
17915 @cindex quoting names
17916 @anchor{quoting names}
17917 Occasionally, you may need to refer to symbols that contain unusual
17918 characters, which @value{GDBN} ordinarily treats as word delimiters. The
17919 most frequent case is in referring to static variables in other
17920 source files (@pxref{Variables,,Program Variables}). File names
17921 are recorded in object files as debugging symbols, but @value{GDBN} would
17922 ordinarily parse a typical file name, like @file{foo.c}, as the three words
17923 @samp{foo} @samp{.} @samp{c}. To allow @value{GDBN} to recognize
17924 @samp{foo.c} as a single symbol, enclose it in single quotes; for example,
17931 looks up the value of @code{x} in the scope of the file @file{foo.c}.
17934 @cindex case-insensitive symbol names
17935 @cindex case sensitivity in symbol names
17936 @kindex set case-sensitive
17937 @item set case-sensitive on
17938 @itemx set case-sensitive off
17939 @itemx set case-sensitive auto
17940 Normally, when @value{GDBN} looks up symbols, it matches their names
17941 with case sensitivity determined by the current source language.
17942 Occasionally, you may wish to control that. The command @code{set
17943 case-sensitive} lets you do that by specifying @code{on} for
17944 case-sensitive matches or @code{off} for case-insensitive ones. If
17945 you specify @code{auto}, case sensitivity is reset to the default
17946 suitable for the source language. The default is case-sensitive
17947 matches for all languages except for Fortran, for which the default is
17948 case-insensitive matches.
17950 @kindex show case-sensitive
17951 @item show case-sensitive
17952 This command shows the current setting of case sensitivity for symbols
17955 @kindex set print type methods
17956 @item set print type methods
17957 @itemx set print type methods on
17958 @itemx set print type methods off
17959 Normally, when @value{GDBN} prints a class, it displays any methods
17960 declared in that class. You can control this behavior either by
17961 passing the appropriate flag to @code{ptype}, or using @command{set
17962 print type methods}. Specifying @code{on} will cause @value{GDBN} to
17963 display the methods; this is the default. Specifying @code{off} will
17964 cause @value{GDBN} to omit the methods.
17966 @kindex show print type methods
17967 @item show print type methods
17968 This command shows the current setting of method display when printing
17971 @kindex set print type nested-type-limit
17972 @item set print type nested-type-limit @var{limit}
17973 @itemx set print type nested-type-limit unlimited
17974 Set the limit of displayed nested types that the type printer will
17975 show. A @var{limit} of @code{unlimited} or @code{-1} will show all
17976 nested definitions. By default, the type printer will not show any nested
17977 types defined in classes.
17979 @kindex show print type nested-type-limit
17980 @item show print type nested-type-limit
17981 This command shows the current display limit of nested types when
17984 @kindex set print type typedefs
17985 @item set print type typedefs
17986 @itemx set print type typedefs on
17987 @itemx set print type typedefs off
17989 Normally, when @value{GDBN} prints a class, it displays any typedefs
17990 defined in that class. You can control this behavior either by
17991 passing the appropriate flag to @code{ptype}, or using @command{set
17992 print type typedefs}. Specifying @code{on} will cause @value{GDBN} to
17993 display the typedef definitions; this is the default. Specifying
17994 @code{off} will cause @value{GDBN} to omit the typedef definitions.
17995 Note that this controls whether the typedef definition itself is
17996 printed, not whether typedef names are substituted when printing other
17999 @kindex show print type typedefs
18000 @item show print type typedefs
18001 This command shows the current setting of typedef display when
18004 @kindex info address
18005 @cindex address of a symbol
18006 @item info address @var{symbol}
18007 Describe where the data for @var{symbol} is stored. For a register
18008 variable, this says which register it is kept in. For a non-register
18009 local variable, this prints the stack-frame offset at which the variable
18012 Note the contrast with @samp{print &@var{symbol}}, which does not work
18013 at all for a register variable, and for a stack local variable prints
18014 the exact address of the current instantiation of the variable.
18016 @kindex info symbol
18017 @cindex symbol from address
18018 @cindex closest symbol and offset for an address
18019 @item info symbol @var{addr}
18020 Print the name of a symbol which is stored at the address @var{addr}.
18021 If no symbol is stored exactly at @var{addr}, @value{GDBN} prints the
18022 nearest symbol and an offset from it:
18025 (@value{GDBP}) info symbol 0x54320
18026 _initialize_vx + 396 in section .text
18030 This is the opposite of the @code{info address} command. You can use
18031 it to find out the name of a variable or a function given its address.
18033 For dynamically linked executables, the name of executable or shared
18034 library containing the symbol is also printed:
18037 (@value{GDBP}) info symbol 0x400225
18038 _start + 5 in section .text of /tmp/a.out
18039 (@value{GDBP}) info symbol 0x2aaaac2811cf
18040 __read_nocancel + 6 in section .text of /usr/lib64/libc.so.6
18045 @item demangle @r{[}-l @var{language}@r{]} @r{[}@var{--}@r{]} @var{name}
18046 Demangle @var{name}.
18047 If @var{language} is provided it is the name of the language to demangle
18048 @var{name} in. Otherwise @var{name} is demangled in the current language.
18050 The @samp{--} option specifies the end of options,
18051 and is useful when @var{name} begins with a dash.
18053 The parameter @code{demangle-style} specifies how to interpret the kind
18054 of mangling used. @xref{Print Settings}.
18057 @item whatis[/@var{flags}] [@var{arg}]
18058 Print the data type of @var{arg}, which can be either an expression
18059 or a name of a data type. With no argument, print the data type of
18060 @code{$}, the last value in the value history.
18062 If @var{arg} is an expression (@pxref{Expressions, ,Expressions}), it
18063 is not actually evaluated, and any side-effecting operations (such as
18064 assignments or function calls) inside it do not take place.
18066 If @var{arg} is a variable or an expression, @code{whatis} prints its
18067 literal type as it is used in the source code. If the type was
18068 defined using a @code{typedef}, @code{whatis} will @emph{not} print
18069 the data type underlying the @code{typedef}. If the type of the
18070 variable or the expression is a compound data type, such as
18071 @code{struct} or @code{class}, @code{whatis} never prints their
18072 fields or methods. It just prints the @code{struct}/@code{class}
18073 name (a.k.a.@: its @dfn{tag}). If you want to see the members of
18074 such a compound data type, use @code{ptype}.
18076 If @var{arg} is a type name that was defined using @code{typedef},
18077 @code{whatis} @dfn{unrolls} only one level of that @code{typedef}.
18078 Unrolling means that @code{whatis} will show the underlying type used
18079 in the @code{typedef} declaration of @var{arg}. However, if that
18080 underlying type is also a @code{typedef}, @code{whatis} will not
18083 For C code, the type names may also have the form @samp{class
18084 @var{class-name}}, @samp{struct @var{struct-tag}}, @samp{union
18085 @var{union-tag}} or @samp{enum @var{enum-tag}}.
18087 @var{flags} can be used to modify how the type is displayed.
18088 Available flags are:
18092 Display in ``raw'' form. Normally, @value{GDBN} substitutes template
18093 parameters and typedefs defined in a class when printing the class'
18094 members. The @code{/r} flag disables this.
18097 Do not print methods defined in the class.
18100 Print methods defined in the class. This is the default, but the flag
18101 exists in case you change the default with @command{set print type methods}.
18104 Do not print typedefs defined in the class. Note that this controls
18105 whether the typedef definition itself is printed, not whether typedef
18106 names are substituted when printing other types.
18109 Print typedefs defined in the class. This is the default, but the flag
18110 exists in case you change the default with @command{set print type typedefs}.
18113 Print the offsets and sizes of fields in a struct, similar to what the
18114 @command{pahole} tool does. This option implies the @code{/tm} flags.
18116 For example, given the following declarations:
18152 Issuing a @kbd{ptype /o struct tuv} command would print:
18155 (@value{GDBP}) ptype /o struct tuv
18156 /* offset | size */ type = struct tuv @{
18157 /* 0 | 4 */ int a1;
18158 /* XXX 4-byte hole */
18159 /* 8 | 8 */ char *a2;
18160 /* 16 | 4 */ int a3;
18162 /* total size (bytes): 24 */
18166 Notice the format of the first column of comments. There, you can
18167 find two parts separated by the @samp{|} character: the @emph{offset},
18168 which indicates where the field is located inside the struct, in
18169 bytes, and the @emph{size} of the field. Another interesting line is
18170 the marker of a @emph{hole} in the struct, indicating that it may be
18171 possible to pack the struct and make it use less space by reorganizing
18174 It is also possible to print offsets inside an union:
18177 (@value{GDBP}) ptype /o union qwe
18178 /* offset | size */ type = union qwe @{
18179 /* 24 */ struct tuv @{
18180 /* 0 | 4 */ int a1;
18181 /* XXX 4-byte hole */
18182 /* 8 | 8 */ char *a2;
18183 /* 16 | 4 */ int a3;
18185 /* total size (bytes): 24 */
18187 /* 40 */ struct xyz @{
18188 /* 0 | 4 */ int f1;
18189 /* 4 | 1 */ char f2;
18190 /* XXX 3-byte hole */
18191 /* 8 | 8 */ void *f3;
18192 /* 16 | 24 */ struct tuv @{
18193 /* 16 | 4 */ int a1;
18194 /* XXX 4-byte hole */
18195 /* 24 | 8 */ char *a2;
18196 /* 32 | 4 */ int a3;
18198 /* total size (bytes): 24 */
18201 /* total size (bytes): 40 */
18204 /* total size (bytes): 40 */
18208 In this case, since @code{struct tuv} and @code{struct xyz} occupy the
18209 same space (because we are dealing with an union), the offset is not
18210 printed for them. However, you can still examine the offset of each
18211 of these structures' fields.
18213 Another useful scenario is printing the offsets of a struct containing
18217 (@value{GDBP}) ptype /o struct tyu
18218 /* offset | size */ type = struct tyu @{
18219 /* 0:31 | 4 */ int a1 : 1;
18220 /* 0:28 | 4 */ int a2 : 3;
18221 /* 0: 5 | 4 */ int a3 : 23;
18222 /* 3: 3 | 1 */ signed char a4 : 2;
18223 /* XXX 3-bit hole */
18224 /* XXX 4-byte hole */
18225 /* 8 | 8 */ int64_t a5;
18226 /* 16: 0 | 4 */ int a6 : 5;
18227 /* 16: 5 | 8 */ int64_t a7 : 3;
18228 "/* XXX 7-byte padding */
18230 /* total size (bytes): 24 */
18234 Note how the offset information is now extended to also include the
18235 first bit of the bitfield.
18239 @item ptype[/@var{flags}] [@var{arg}]
18240 @code{ptype} accepts the same arguments as @code{whatis}, but prints a
18241 detailed description of the type, instead of just the name of the type.
18242 @xref{Expressions, ,Expressions}.
18244 Contrary to @code{whatis}, @code{ptype} always unrolls any
18245 @code{typedef}s in its argument declaration, whether the argument is
18246 a variable, expression, or a data type. This means that @code{ptype}
18247 of a variable or an expression will not print literally its type as
18248 present in the source code---use @code{whatis} for that. @code{typedef}s at
18249 the pointer or reference targets are also unrolled. Only @code{typedef}s of
18250 fields, methods and inner @code{class typedef}s of @code{struct}s,
18251 @code{class}es and @code{union}s are not unrolled even with @code{ptype}.
18253 For example, for this variable declaration:
18256 typedef double real_t;
18257 struct complex @{ real_t real; double imag; @};
18258 typedef struct complex complex_t;
18260 real_t *real_pointer_var;
18264 the two commands give this output:
18268 (@value{GDBP}) whatis var
18270 (@value{GDBP}) ptype var
18271 type = struct complex @{
18275 (@value{GDBP}) whatis complex_t
18276 type = struct complex
18277 (@value{GDBP}) whatis struct complex
18278 type = struct complex
18279 (@value{GDBP}) ptype struct complex
18280 type = struct complex @{
18284 (@value{GDBP}) whatis real_pointer_var
18286 (@value{GDBP}) ptype real_pointer_var
18292 As with @code{whatis}, using @code{ptype} without an argument refers to
18293 the type of @code{$}, the last value in the value history.
18295 @cindex incomplete type
18296 Sometimes, programs use opaque data types or incomplete specifications
18297 of complex data structure. If the debug information included in the
18298 program does not allow @value{GDBN} to display a full declaration of
18299 the data type, it will say @samp{<incomplete type>}. For example,
18300 given these declarations:
18304 struct foo *fooptr;
18308 but no definition for @code{struct foo} itself, @value{GDBN} will say:
18311 (@value{GDBP}) ptype foo
18312 $1 = <incomplete type>
18316 ``Incomplete type'' is C terminology for data types that are not
18317 completely specified.
18319 @cindex unknown type
18320 Othertimes, information about a variable's type is completely absent
18321 from the debug information included in the program. This most often
18322 happens when the program or library where the variable is defined
18323 includes no debug information at all. @value{GDBN} knows the variable
18324 exists from inspecting the linker/loader symbol table (e.g., the ELF
18325 dynamic symbol table), but such symbols do not contain type
18326 information. Inspecting the type of a (global) variable for which
18327 @value{GDBN} has no type information shows:
18330 (@value{GDBP}) ptype var
18331 type = <data variable, no debug info>
18334 @xref{Variables, no debug info variables}, for how to print the values
18338 @item info types @var{regexp}
18340 Print a brief description of all types whose names match the regular
18341 expression @var{regexp} (or all types in your program, if you supply
18342 no argument). Each complete typename is matched as though it were a
18343 complete line; thus, @samp{i type value} gives information on all
18344 types in your program whose names include the string @code{value}, but
18345 @samp{i type ^value$} gives information only on types whose complete
18346 name is @code{value}.
18348 In programs using different languages, @value{GDBN} chooses the syntax
18349 to print the type description according to the
18350 @samp{set language} value: using @samp{set language auto}
18351 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18352 language of the type, other values mean to use
18353 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18355 This command differs from @code{ptype} in two ways: first, like
18356 @code{whatis}, it does not print a detailed description; second, it
18357 lists all source files and line numbers where a type is defined.
18359 @kindex info type-printers
18360 @item info type-printers
18361 Versions of @value{GDBN} that ship with Python scripting enabled may
18362 have ``type printers'' available. When using @command{ptype} or
18363 @command{whatis}, these printers are consulted when the name of a type
18364 is needed. @xref{Type Printing API}, for more information on writing
18367 @code{info type-printers} displays all the available type printers.
18369 @kindex enable type-printer
18370 @kindex disable type-printer
18371 @item enable type-printer @var{name}@dots{}
18372 @item disable type-printer @var{name}@dots{}
18373 These commands can be used to enable or disable type printers.
18376 @cindex local variables
18377 @item info scope @var{location}
18378 List all the variables local to a particular scope. This command
18379 accepts a @var{location} argument---a function name, a source line, or
18380 an address preceded by a @samp{*}, and prints all the variables local
18381 to the scope defined by that location. (@xref{Specify Location}, for
18382 details about supported forms of @var{location}.) For example:
18385 (@value{GDBP}) @b{info scope command_line_handler}
18386 Scope for command_line_handler:
18387 Symbol rl is an argument at stack/frame offset 8, length 4.
18388 Symbol linebuffer is in static storage at address 0x150a18, length 4.
18389 Symbol linelength is in static storage at address 0x150a1c, length 4.
18390 Symbol p is a local variable in register $esi, length 4.
18391 Symbol p1 is a local variable in register $ebx, length 4.
18392 Symbol nline is a local variable in register $edx, length 4.
18393 Symbol repeat is a local variable at frame offset -8, length 4.
18397 This command is especially useful for determining what data to collect
18398 during a @dfn{trace experiment}, see @ref{Tracepoint Actions,
18401 @kindex info source
18403 Show information about the current source file---that is, the source file for
18404 the function containing the current point of execution:
18407 the name of the source file, and the directory containing it,
18409 the directory it was compiled in,
18411 its length, in lines,
18413 which programming language it is written in,
18415 if the debug information provides it, the program that compiled the file
18416 (which may include, e.g., the compiler version and command line arguments),
18418 whether the executable includes debugging information for that file, and
18419 if so, what format the information is in (e.g., STABS, Dwarf 2, etc.), and
18421 whether the debugging information includes information about
18422 preprocessor macros.
18426 @kindex info sources
18428 Print the names of all source files in your program for which there is
18429 debugging information, organized into two lists: files whose symbols
18430 have already been read, and files whose symbols will be read when needed.
18432 @kindex info functions
18433 @item info functions [-q]
18434 Print the names and data types of all defined functions.
18435 Similarly to @samp{info types}, this command groups its output by source
18436 files and annotates each function definition with its source line
18439 In programs using different languages, @value{GDBN} chooses the syntax
18440 to print the function name and type according to the
18441 @samp{set language} value: using @samp{set language auto}
18442 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18443 language of the function, other values mean to use
18444 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18446 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18447 printing header information and messages explaining why no functions
18450 @item info functions [-q] [-t @var{type_regexp}] [@var{regexp}]
18451 Like @samp{info functions}, but only print the names and data types
18452 of the functions selected with the provided regexp(s).
18454 If @var{regexp} is provided, print only the functions whose names
18455 match the regular expression @var{regexp}.
18456 Thus, @samp{info fun step} finds all functions whose
18457 names include @code{step}; @samp{info fun ^step} finds those whose names
18458 start with @code{step}. If a function name contains characters that
18459 conflict with the regular expression language (e.g.@:
18460 @samp{operator*()}), they may be quoted with a backslash.
18462 If @var{type_regexp} is provided, print only the functions whose
18463 types, as printed by the @code{whatis} command, match
18464 the regular expression @var{type_regexp}.
18465 If @var{type_regexp} contains space(s), it should be enclosed in
18466 quote characters. If needed, use backslash to escape the meaning
18467 of special characters or quotes.
18468 Thus, @samp{info fun -t '^int ('} finds the functions that return
18469 an integer; @samp{info fun -t '(.*int.*'} finds the functions that
18470 have an argument type containing int; @samp{info fun -t '^int (' ^step}
18471 finds the functions whose names start with @code{step} and that return
18474 If both @var{regexp} and @var{type_regexp} are provided, a function
18475 is printed only if its name matches @var{regexp} and its type matches
18479 @kindex info variables
18480 @item info variables [-q]
18481 Print the names and data types of all variables that are defined
18482 outside of functions (i.e.@: excluding local variables).
18483 The printed variables are grouped by source files and annotated with
18484 their respective source line numbers.
18486 In programs using different languages, @value{GDBN} chooses the syntax
18487 to print the variable name and type according to the
18488 @samp{set language} value: using @samp{set language auto}
18489 (see @ref{Automatically, ,Set Language Automatically}) means to use the
18490 language of the variable, other values mean to use
18491 the manually specified language (see @ref{Manually, ,Set Language Manually}).
18493 The optional flag @samp{-q}, which stands for @samp{quiet}, disables
18494 printing header information and messages explaining why no variables
18497 @item info variables [-q] [-t @var{type_regexp}] [@var{regexp}]
18498 Like @kbd{info variables}, but only print the variables selected
18499 with the provided regexp(s).
18501 If @var{regexp} is provided, print only the variables whose names
18502 match the regular expression @var{regexp}.
18504 If @var{type_regexp} is provided, print only the variables whose
18505 types, as printed by the @code{whatis} command, match
18506 the regular expression @var{type_regexp}.
18507 If @var{type_regexp} contains space(s), it should be enclosed in
18508 quote characters. If needed, use backslash to escape the meaning
18509 of special characters or quotes.
18511 If both @var{regexp} and @var{type_regexp} are provided, an argument
18512 is printed only if its name matches @var{regexp} and its type matches
18515 @kindex info classes
18516 @cindex Objective-C, classes and selectors
18518 @itemx info classes @var{regexp}
18519 Display all Objective-C classes in your program, or
18520 (with the @var{regexp} argument) all those matching a particular regular
18523 @kindex info selectors
18524 @item info selectors
18525 @itemx info selectors @var{regexp}
18526 Display all Objective-C selectors in your program, or
18527 (with the @var{regexp} argument) all those matching a particular regular
18531 This was never implemented.
18532 @kindex info methods
18534 @itemx info methods @var{regexp}
18535 The @code{info methods} command permits the user to examine all defined
18536 methods within C@t{++} program, or (with the @var{regexp} argument) a
18537 specific set of methods found in the various C@t{++} classes. Many
18538 C@t{++} classes provide a large number of methods. Thus, the output
18539 from the @code{ptype} command can be overwhelming and hard to use. The
18540 @code{info-methods} command filters the methods, printing only those
18541 which match the regular-expression @var{regexp}.
18544 @cindex opaque data types
18545 @kindex set opaque-type-resolution
18546 @item set opaque-type-resolution on
18547 Tell @value{GDBN} to resolve opaque types. An opaque type is a type
18548 declared as a pointer to a @code{struct}, @code{class}, or
18549 @code{union}---for example, @code{struct MyType *}---that is used in one
18550 source file although the full declaration of @code{struct MyType} is in
18551 another source file. The default is on.
18553 A change in the setting of this subcommand will not take effect until
18554 the next time symbols for a file are loaded.
18556 @item set opaque-type-resolution off
18557 Tell @value{GDBN} not to resolve opaque types. In this case, the type
18558 is printed as follows:
18560 @{<no data fields>@}
18563 @kindex show opaque-type-resolution
18564 @item show opaque-type-resolution
18565 Show whether opaque types are resolved or not.
18567 @kindex set print symbol-loading
18568 @cindex print messages when symbols are loaded
18569 @item set print symbol-loading
18570 @itemx set print symbol-loading full
18571 @itemx set print symbol-loading brief
18572 @itemx set print symbol-loading off
18573 The @code{set print symbol-loading} command allows you to control the
18574 printing of messages when @value{GDBN} loads symbol information.
18575 By default a message is printed for the executable and one for each
18576 shared library, and normally this is what you want. However, when
18577 debugging apps with large numbers of shared libraries these messages
18579 When set to @code{brief} a message is printed for each executable,
18580 and when @value{GDBN} loads a collection of shared libraries at once
18581 it will only print one message regardless of the number of shared
18582 libraries. When set to @code{off} no messages are printed.
18584 @kindex show print symbol-loading
18585 @item show print symbol-loading
18586 Show whether messages will be printed when a @value{GDBN} command
18587 entered from the keyboard causes symbol information to be loaded.
18589 @kindex maint print symbols
18590 @cindex symbol dump
18591 @kindex maint print psymbols
18592 @cindex partial symbol dump
18593 @kindex maint print msymbols
18594 @cindex minimal symbol dump
18595 @item maint print symbols @r{[}-pc @var{address}@r{]} @r{[}@var{filename}@r{]}
18596 @itemx maint print symbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18597 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-pc @var{address}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18598 @itemx maint print psymbols @r{[}-objfile @var{objfile}@r{]} @r{[}-source @var{source}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18599 @itemx maint print msymbols @r{[}-objfile @var{objfile}@r{]} @r{[}--@r{]} @r{[}@var{filename}@r{]}
18600 Write a dump of debugging symbol data into the file @var{filename} or
18601 the terminal if @var{filename} is unspecified.
18602 If @code{-objfile @var{objfile}} is specified, only dump symbols for
18604 If @code{-pc @var{address}} is specified, only dump symbols for the file
18605 with code at that address. Note that @var{address} may be a symbol like
18607 If @code{-source @var{source}} is specified, only dump symbols for that
18610 These commands are used to debug the @value{GDBN} symbol-reading code.
18611 These commands do not modify internal @value{GDBN} state, therefore
18612 @samp{maint print symbols} will only print symbols for already expanded symbol
18614 You can use the command @code{info sources} to find out which files these are.
18615 If you use @samp{maint print psymbols} instead, the dump shows information
18616 about symbols that @value{GDBN} only knows partially---that is, symbols
18617 defined in files that @value{GDBN} has skimmed, but not yet read completely.
18618 Finally, @samp{maint print msymbols} just dumps ``minimal symbols'', e.g.,
18621 @xref{Files, ,Commands to Specify Files}, for a discussion of how
18622 @value{GDBN} reads symbols (in the description of @code{symbol-file}).
18624 @kindex maint info symtabs
18625 @kindex maint info psymtabs
18626 @cindex listing @value{GDBN}'s internal symbol tables
18627 @cindex symbol tables, listing @value{GDBN}'s internal
18628 @cindex full symbol tables, listing @value{GDBN}'s internal
18629 @cindex partial symbol tables, listing @value{GDBN}'s internal
18630 @item maint info symtabs @r{[} @var{regexp} @r{]}
18631 @itemx maint info psymtabs @r{[} @var{regexp} @r{]}
18633 List the @code{struct symtab} or @code{struct partial_symtab}
18634 structures whose names match @var{regexp}. If @var{regexp} is not
18635 given, list them all. The output includes expressions which you can
18636 copy into a @value{GDBN} debugging this one to examine a particular
18637 structure in more detail. For example:
18640 (@value{GDBP}) maint info psymtabs dwarf2read
18641 @{ objfile /home/gnu/build/gdb/gdb
18642 ((struct objfile *) 0x82e69d0)
18643 @{ psymtab /home/gnu/src/gdb/dwarf2read.c
18644 ((struct partial_symtab *) 0x8474b10)
18647 text addresses 0x814d3c8 -- 0x8158074
18648 globals (* (struct partial_symbol **) 0x8507a08 @@ 9)
18649 statics (* (struct partial_symbol **) 0x40e95b78 @@ 2882)
18650 dependencies (none)
18653 (@value{GDBP}) maint info symtabs
18657 We see that there is one partial symbol table whose filename contains
18658 the string @samp{dwarf2read}, belonging to the @samp{gdb} executable;
18659 and we see that @value{GDBN} has not read in any symtabs yet at all.
18660 If we set a breakpoint on a function, that will cause @value{GDBN} to
18661 read the symtab for the compilation unit containing that function:
18664 (@value{GDBP}) break dwarf2_psymtab_to_symtab
18665 Breakpoint 1 at 0x814e5da: file /home/gnu/src/gdb/dwarf2read.c,
18667 (@value{GDBP}) maint info symtabs
18668 @{ objfile /home/gnu/build/gdb/gdb
18669 ((struct objfile *) 0x82e69d0)
18670 @{ symtab /home/gnu/src/gdb/dwarf2read.c
18671 ((struct symtab *) 0x86c1f38)
18674 blockvector ((struct blockvector *) 0x86c1bd0) (primary)
18675 linetable ((struct linetable *) 0x8370fa0)
18676 debugformat DWARF 2
18682 @kindex maint info line-table
18683 @cindex listing @value{GDBN}'s internal line tables
18684 @cindex line tables, listing @value{GDBN}'s internal
18685 @item maint info line-table @r{[} @var{regexp} @r{]}
18687 List the @code{struct linetable} from all @code{struct symtab}
18688 instances whose name matches @var{regexp}. If @var{regexp} is not
18689 given, list the @code{struct linetable} from all @code{struct symtab}.
18691 @kindex maint set symbol-cache-size
18692 @cindex symbol cache size
18693 @item maint set symbol-cache-size @var{size}
18694 Set the size of the symbol cache to @var{size}.
18695 The default size is intended to be good enough for debugging
18696 most applications. This option exists to allow for experimenting
18697 with different sizes.
18699 @kindex maint show symbol-cache-size
18700 @item maint show symbol-cache-size
18701 Show the size of the symbol cache.
18703 @kindex maint print symbol-cache
18704 @cindex symbol cache, printing its contents
18705 @item maint print symbol-cache
18706 Print the contents of the symbol cache.
18707 This is useful when debugging symbol cache issues.
18709 @kindex maint print symbol-cache-statistics
18710 @cindex symbol cache, printing usage statistics
18711 @item maint print symbol-cache-statistics
18712 Print symbol cache usage statistics.
18713 This helps determine how well the cache is being utilized.
18715 @kindex maint flush-symbol-cache
18716 @cindex symbol cache, flushing
18717 @item maint flush-symbol-cache
18718 Flush the contents of the symbol cache, all entries are removed.
18719 This command is useful when debugging the symbol cache.
18720 It is also useful when collecting performance data.
18725 @chapter Altering Execution
18727 Once you think you have found an error in your program, you might want to
18728 find out for certain whether correcting the apparent error would lead to
18729 correct results in the rest of the run. You can find the answer by
18730 experiment, using the @value{GDBN} features for altering execution of the
18733 For example, you can store new values into variables or memory
18734 locations, give your program a signal, restart it at a different
18735 address, or even return prematurely from a function.
18738 * Assignment:: Assignment to variables
18739 * Jumping:: Continuing at a different address
18740 * Signaling:: Giving your program a signal
18741 * Returning:: Returning from a function
18742 * Calling:: Calling your program's functions
18743 * Patching:: Patching your program
18744 * Compiling and Injecting Code:: Compiling and injecting code in @value{GDBN}
18748 @section Assignment to Variables
18751 @cindex setting variables
18752 To alter the value of a variable, evaluate an assignment expression.
18753 @xref{Expressions, ,Expressions}. For example,
18760 stores the value 4 into the variable @code{x}, and then prints the
18761 value of the assignment expression (which is 4).
18762 @xref{Languages, ,Using @value{GDBN} with Different Languages}, for more
18763 information on operators in supported languages.
18765 @kindex set variable
18766 @cindex variables, setting
18767 If you are not interested in seeing the value of the assignment, use the
18768 @code{set} command instead of the @code{print} command. @code{set} is
18769 really the same as @code{print} except that the expression's value is
18770 not printed and is not put in the value history (@pxref{Value History,
18771 ,Value History}). The expression is evaluated only for its effects.
18773 If the beginning of the argument string of the @code{set} command
18774 appears identical to a @code{set} subcommand, use the @code{set
18775 variable} command instead of just @code{set}. This command is identical
18776 to @code{set} except for its lack of subcommands. For example, if your
18777 program has a variable @code{width}, you get an error if you try to set
18778 a new value with just @samp{set width=13}, because @value{GDBN} has the
18779 command @code{set width}:
18782 (@value{GDBP}) whatis width
18784 (@value{GDBP}) p width
18786 (@value{GDBP}) set width=47
18787 Invalid syntax in expression.
18791 The invalid expression, of course, is @samp{=47}. In
18792 order to actually set the program's variable @code{width}, use
18795 (@value{GDBP}) set var width=47
18798 Because the @code{set} command has many subcommands that can conflict
18799 with the names of program variables, it is a good idea to use the
18800 @code{set variable} command instead of just @code{set}. For example, if
18801 your program has a variable @code{g}, you run into problems if you try
18802 to set a new value with just @samp{set g=4}, because @value{GDBN} has
18803 the command @code{set gnutarget}, abbreviated @code{set g}:
18807 (@value{GDBP}) whatis g
18811 (@value{GDBP}) set g=4
18815 The program being debugged has been started already.
18816 Start it from the beginning? (y or n) y
18817 Starting program: /home/smith/cc_progs/a.out
18818 "/home/smith/cc_progs/a.out": can't open to read symbols:
18819 Invalid bfd target.
18820 (@value{GDBP}) show g
18821 The current BFD target is "=4".
18826 The program variable @code{g} did not change, and you silently set the
18827 @code{gnutarget} to an invalid value. In order to set the variable
18831 (@value{GDBP}) set var g=4
18834 @value{GDBN} allows more implicit conversions in assignments than C; you can
18835 freely store an integer value into a pointer variable or vice versa,
18836 and you can convert any structure to any other structure that is the
18837 same length or shorter.
18838 @comment FIXME: how do structs align/pad in these conversions?
18839 @comment /doc@cygnus.com 18dec1990
18841 To store values into arbitrary places in memory, use the @samp{@{@dots{}@}}
18842 construct to generate a value of specified type at a specified address
18843 (@pxref{Expressions, ,Expressions}). For example, @code{@{int@}0x83040} refers
18844 to memory location @code{0x83040} as an integer (which implies a certain size
18845 and representation in memory), and
18848 set @{int@}0x83040 = 4
18852 stores the value 4 into that memory location.
18855 @section Continuing at a Different Address
18857 Ordinarily, when you continue your program, you do so at the place where
18858 it stopped, with the @code{continue} command. You can instead continue at
18859 an address of your own choosing, with the following commands:
18863 @kindex j @r{(@code{jump})}
18864 @item jump @var{location}
18865 @itemx j @var{location}
18866 Resume execution at @var{location}. Execution stops again immediately
18867 if there is a breakpoint there. @xref{Specify Location}, for a description
18868 of the different forms of @var{location}. It is common
18869 practice to use the @code{tbreak} command in conjunction with
18870 @code{jump}. @xref{Set Breaks, ,Setting Breakpoints}.
18872 The @code{jump} command does not change the current stack frame, or
18873 the stack pointer, or the contents of any memory location or any
18874 register other than the program counter. If @var{location} is in
18875 a different function from the one currently executing, the results may
18876 be bizarre if the two functions expect different patterns of arguments or
18877 of local variables. For this reason, the @code{jump} command requests
18878 confirmation if the specified line is not in the function currently
18879 executing. However, even bizarre results are predictable if you are
18880 well acquainted with the machine-language code of your program.
18883 On many systems, you can get much the same effect as the @code{jump}
18884 command by storing a new value into the register @code{$pc}. The
18885 difference is that this does not start your program running; it only
18886 changes the address of where it @emph{will} run when you continue. For
18894 makes the next @code{continue} command or stepping command execute at
18895 address @code{0x485}, rather than at the address where your program stopped.
18896 @xref{Continuing and Stepping, ,Continuing and Stepping}.
18898 The most common occasion to use the @code{jump} command is to back
18899 up---perhaps with more breakpoints set---over a portion of a program
18900 that has already executed, in order to examine its execution in more
18905 @section Giving your Program a Signal
18906 @cindex deliver a signal to a program
18910 @item signal @var{signal}
18911 Resume execution where your program is stopped, but immediately give it the
18912 signal @var{signal}. The @var{signal} can be the name or the number of a
18913 signal. For example, on many systems @code{signal 2} and @code{signal
18914 SIGINT} are both ways of sending an interrupt signal.
18916 Alternatively, if @var{signal} is zero, continue execution without
18917 giving a signal. This is useful when your program stopped on account of
18918 a signal and would ordinarily see the signal when resumed with the
18919 @code{continue} command; @samp{signal 0} causes it to resume without a
18922 @emph{Note:} When resuming a multi-threaded program, @var{signal} is
18923 delivered to the currently selected thread, not the thread that last
18924 reported a stop. This includes the situation where a thread was
18925 stopped due to a signal. So if you want to continue execution
18926 suppressing the signal that stopped a thread, you should select that
18927 same thread before issuing the @samp{signal 0} command. If you issue
18928 the @samp{signal 0} command with another thread as the selected one,
18929 @value{GDBN} detects that and asks for confirmation.
18931 Invoking the @code{signal} command is not the same as invoking the
18932 @code{kill} utility from the shell. Sending a signal with @code{kill}
18933 causes @value{GDBN} to decide what to do with the signal depending on
18934 the signal handling tables (@pxref{Signals}). The @code{signal} command
18935 passes the signal directly to your program.
18937 @code{signal} does not repeat when you press @key{RET} a second time
18938 after executing the command.
18940 @kindex queue-signal
18941 @item queue-signal @var{signal}
18942 Queue @var{signal} to be delivered immediately to the current thread
18943 when execution of the thread resumes. The @var{signal} can be the name or
18944 the number of a signal. For example, on many systems @code{signal 2} and
18945 @code{signal SIGINT} are both ways of sending an interrupt signal.
18946 The handling of the signal must be set to pass the signal to the program,
18947 otherwise @value{GDBN} will report an error.
18948 You can control the handling of signals from @value{GDBN} with the
18949 @code{handle} command (@pxref{Signals}).
18951 Alternatively, if @var{signal} is zero, any currently queued signal
18952 for the current thread is discarded and when execution resumes no signal
18953 will be delivered. This is useful when your program stopped on account
18954 of a signal and would ordinarily see the signal when resumed with the
18955 @code{continue} command.
18957 This command differs from the @code{signal} command in that the signal
18958 is just queued, execution is not resumed. And @code{queue-signal} cannot
18959 be used to pass a signal whose handling state has been set to @code{nopass}
18964 @xref{stepping into signal handlers}, for information on how stepping
18965 commands behave when the thread has a signal queued.
18968 @section Returning from a Function
18971 @cindex returning from a function
18974 @itemx return @var{expression}
18975 You can cancel execution of a function call with the @code{return}
18976 command. If you give an
18977 @var{expression} argument, its value is used as the function's return
18981 When you use @code{return}, @value{GDBN} discards the selected stack frame
18982 (and all frames within it). You can think of this as making the
18983 discarded frame return prematurely. If you wish to specify a value to
18984 be returned, give that value as the argument to @code{return}.
18986 This pops the selected stack frame (@pxref{Selection, ,Selecting a
18987 Frame}), and any other frames inside of it, leaving its caller as the
18988 innermost remaining frame. That frame becomes selected. The
18989 specified value is stored in the registers used for returning values
18992 The @code{return} command does not resume execution; it leaves the
18993 program stopped in the state that would exist if the function had just
18994 returned. In contrast, the @code{finish} command (@pxref{Continuing
18995 and Stepping, ,Continuing and Stepping}) resumes execution until the
18996 selected stack frame returns naturally.
18998 @value{GDBN} needs to know how the @var{expression} argument should be set for
18999 the inferior. The concrete registers assignment depends on the OS ABI and the
19000 type being returned by the selected stack frame. For example it is common for
19001 OS ABI to return floating point values in FPU registers while integer values in
19002 CPU registers. Still some ABIs return even floating point values in CPU
19003 registers. Larger integer widths (such as @code{long long int}) also have
19004 specific placement rules. @value{GDBN} already knows the OS ABI from its
19005 current target so it needs to find out also the type being returned to make the
19006 assignment into the right register(s).
19008 Normally, the selected stack frame has debug info. @value{GDBN} will always
19009 use the debug info instead of the implicit type of @var{expression} when the
19010 debug info is available. For example, if you type @kbd{return -1}, and the
19011 function in the current stack frame is declared to return a @code{long long
19012 int}, @value{GDBN} transparently converts the implicit @code{int} value of -1
19013 into a @code{long long int}:
19016 Breakpoint 1, func () at gdb.base/return-nodebug.c:29
19018 (@value{GDBP}) return -1
19019 Make func return now? (y or n) y
19020 #0 0x004004f6 in main () at gdb.base/return-nodebug.c:43
19021 43 printf ("result=%lld\n", func ());
19025 However, if the selected stack frame does not have a debug info, e.g., if the
19026 function was compiled without debug info, @value{GDBN} has to find out the type
19027 to return from user. Specifying a different type by mistake may set the value
19028 in different inferior registers than the caller code expects. For example,
19029 typing @kbd{return -1} with its implicit type @code{int} would set only a part
19030 of a @code{long long int} result for a debug info less function (on 32-bit
19031 architectures). Therefore the user is required to specify the return type by
19032 an appropriate cast explicitly:
19035 Breakpoint 2, 0x0040050b in func ()
19036 (@value{GDBP}) return -1
19037 Return value type not available for selected stack frame.
19038 Please use an explicit cast of the value to return.
19039 (@value{GDBP}) return (long long int) -1
19040 Make selected stack frame return now? (y or n) y
19041 #0 0x00400526 in main ()
19046 @section Calling Program Functions
19049 @cindex calling functions
19050 @cindex inferior functions, calling
19051 @item print @var{expr}
19052 Evaluate the expression @var{expr} and display the resulting value.
19053 The expression may include calls to functions in the program being
19057 @item call @var{expr}
19058 Evaluate the expression @var{expr} without displaying @code{void}
19061 You can use this variant of the @code{print} command if you want to
19062 execute a function from your program that does not return anything
19063 (a.k.a.@: @dfn{a void function}), but without cluttering the output
19064 with @code{void} returned values that @value{GDBN} will otherwise
19065 print. If the result is not void, it is printed and saved in the
19069 It is possible for the function you call via the @code{print} or
19070 @code{call} command to generate a signal (e.g., if there's a bug in
19071 the function, or if you passed it incorrect arguments). What happens
19072 in that case is controlled by the @code{set unwindonsignal} command.
19074 Similarly, with a C@t{++} program it is possible for the function you
19075 call via the @code{print} or @code{call} command to generate an
19076 exception that is not handled due to the constraints of the dummy
19077 frame. In this case, any exception that is raised in the frame, but has
19078 an out-of-frame exception handler will not be found. GDB builds a
19079 dummy-frame for the inferior function call, and the unwinder cannot
19080 seek for exception handlers outside of this dummy-frame. What happens
19081 in that case is controlled by the
19082 @code{set unwind-on-terminating-exception} command.
19085 @item set unwindonsignal
19086 @kindex set unwindonsignal
19087 @cindex unwind stack in called functions
19088 @cindex call dummy stack unwinding
19089 Set unwinding of the stack if a signal is received while in a function
19090 that @value{GDBN} called in the program being debugged. If set to on,
19091 @value{GDBN} unwinds the stack it created for the call and restores
19092 the context to what it was before the call. If set to off (the
19093 default), @value{GDBN} stops in the frame where the signal was
19096 @item show unwindonsignal
19097 @kindex show unwindonsignal
19098 Show the current setting of stack unwinding in the functions called by
19101 @item set unwind-on-terminating-exception
19102 @kindex set unwind-on-terminating-exception
19103 @cindex unwind stack in called functions with unhandled exceptions
19104 @cindex call dummy stack unwinding on unhandled exception.
19105 Set unwinding of the stack if a C@t{++} exception is raised, but left
19106 unhandled while in a function that @value{GDBN} called in the program being
19107 debugged. If set to on (the default), @value{GDBN} unwinds the stack
19108 it created for the call and restores the context to what it was before
19109 the call. If set to off, @value{GDBN} the exception is delivered to
19110 the default C@t{++} exception handler and the inferior terminated.
19112 @item show unwind-on-terminating-exception
19113 @kindex show unwind-on-terminating-exception
19114 Show the current setting of stack unwinding in the functions called by
19117 @item set may-call-functions
19118 @kindex set may-call-functions
19119 @cindex disabling calling functions in the program
19120 @cindex calling functions in the program, disabling
19121 Set permission to call functions in the program.
19122 This controls whether @value{GDBN} will attempt to call functions in
19123 the program, such as with expressions in the @code{print} command. It
19124 defaults to @code{on}.
19126 To call a function in the program, @value{GDBN} has to temporarily
19127 modify the state of the inferior. This has potentially undesired side
19128 effects. Also, having @value{GDBN} call nested functions is likely to
19129 be erroneous and may even crash the program being debugged. You can
19130 avoid such hazards by forbidding @value{GDBN} from calling functions
19131 in the program being debugged. If calling functions in the program
19132 is forbidden, GDB will throw an error when a command (such as printing
19133 an expression) starts a function call in the program.
19135 @item show may-call-functions
19136 @kindex show may-call-functions
19137 Show permission to call functions in the program.
19141 @subsection Calling functions with no debug info
19143 @cindex no debug info functions
19144 Sometimes, a function you wish to call is missing debug information.
19145 In such case, @value{GDBN} does not know the type of the function,
19146 including the types of the function's parameters. To avoid calling
19147 the inferior function incorrectly, which could result in the called
19148 function functioning erroneously and even crash, @value{GDBN} refuses
19149 to call the function unless you tell it the type of the function.
19151 For prototyped (i.e.@: ANSI/ISO style) functions, there are two ways
19152 to do that. The simplest is to cast the call to the function's
19153 declared return type. For example:
19156 (@value{GDBP}) p getenv ("PATH")
19157 'getenv' has unknown return type; cast the call to its declared return type
19158 (@value{GDBP}) p (char *) getenv ("PATH")
19159 $1 = 0x7fffffffe7ba "/usr/local/bin:/"...
19162 Casting the return type of a no-debug function is equivalent to
19163 casting the function to a pointer to a prototyped function that has a
19164 prototype that matches the types of the passed-in arguments, and
19165 calling that. I.e., the call above is equivalent to:
19168 (@value{GDBP}) p ((char * (*) (const char *)) getenv) ("PATH")
19172 and given this prototyped C or C++ function with float parameters:
19175 float multiply (float v1, float v2) @{ return v1 * v2; @}
19179 these calls are equivalent:
19182 (@value{GDBP}) p (float) multiply (2.0f, 3.0f)
19183 (@value{GDBP}) p ((float (*) (float, float)) multiply) (2.0f, 3.0f)
19186 If the function you wish to call is declared as unprototyped (i.e.@:
19187 old K&R style), you must use the cast-to-function-pointer syntax, so
19188 that @value{GDBN} knows that it needs to apply default argument
19189 promotions (promote float arguments to double). @xref{ABI, float
19190 promotion}. For example, given this unprototyped C function with
19191 float parameters, and no debug info:
19195 multiply_noproto (v1, v2)
19203 you call it like this:
19206 (@value{GDBP}) p ((float (*) ()) multiply_noproto) (2.0f, 3.0f)
19210 @section Patching Programs
19212 @cindex patching binaries
19213 @cindex writing into executables
19214 @cindex writing into corefiles
19216 By default, @value{GDBN} opens the file containing your program's
19217 executable code (or the corefile) read-only. This prevents accidental
19218 alterations to machine code; but it also prevents you from intentionally
19219 patching your program's binary.
19221 If you'd like to be able to patch the binary, you can specify that
19222 explicitly with the @code{set write} command. For example, you might
19223 want to turn on internal debugging flags, or even to make emergency
19229 @itemx set write off
19230 If you specify @samp{set write on}, @value{GDBN} opens executable and
19231 core files for both reading and writing; if you specify @kbd{set write
19232 off} (the default), @value{GDBN} opens them read-only.
19234 If you have already loaded a file, you must load it again (using the
19235 @code{exec-file} or @code{core-file} command) after changing @code{set
19236 write}, for your new setting to take effect.
19240 Display whether executable files and core files are opened for writing
19241 as well as reading.
19244 @node Compiling and Injecting Code
19245 @section Compiling and injecting code in @value{GDBN}
19246 @cindex injecting code
19247 @cindex writing into executables
19248 @cindex compiling code
19250 @value{GDBN} supports on-demand compilation and code injection into
19251 programs running under @value{GDBN}. GCC 5.0 or higher built with
19252 @file{libcc1.so} must be installed for this functionality to be enabled.
19253 This functionality is implemented with the following commands.
19256 @kindex compile code
19257 @item compile code @var{source-code}
19258 @itemx compile code -raw @var{--} @var{source-code}
19259 Compile @var{source-code} with the compiler language found as the current
19260 language in @value{GDBN} (@pxref{Languages}). If compilation and
19261 injection is not supported with the current language specified in
19262 @value{GDBN}, or the compiler does not support this feature, an error
19263 message will be printed. If @var{source-code} compiles and links
19264 successfully, @value{GDBN} will load the object-code emitted,
19265 and execute it within the context of the currently selected inferior.
19266 It is important to note that the compiled code is executed immediately.
19267 After execution, the compiled code is removed from @value{GDBN} and any
19268 new types or variables you have defined will be deleted.
19270 The command allows you to specify @var{source-code} in two ways.
19271 The simplest method is to provide a single line of code to the command.
19275 compile code printf ("hello world\n");
19278 If you specify options on the command line as well as source code, they
19279 may conflict. The @samp{--} delimiter can be used to separate options
19280 from actual source code. E.g.:
19283 compile code -r -- printf ("hello world\n");
19286 Alternatively you can enter source code as multiple lines of text. To
19287 enter this mode, invoke the @samp{compile code} command without any text
19288 following the command. This will start the multiple-line editor and
19289 allow you to type as many lines of source code as required. When you
19290 have completed typing, enter @samp{end} on its own line to exit the
19295 >printf ("hello\n");
19296 >printf ("world\n");
19300 Specifying @samp{-raw}, prohibits @value{GDBN} from wrapping the
19301 provided @var{source-code} in a callable scope. In this case, you must
19302 specify the entry point of the code by defining a function named
19303 @code{_gdb_expr_}. The @samp{-raw} code cannot access variables of the
19304 inferior. Using @samp{-raw} option may be needed for example when
19305 @var{source-code} requires @samp{#include} lines which may conflict with
19306 inferior symbols otherwise.
19308 @kindex compile file
19309 @item compile file @var{filename}
19310 @itemx compile file -raw @var{filename}
19311 Like @code{compile code}, but take the source code from @var{filename}.
19314 compile file /home/user/example.c
19319 @item compile print [[@var{options}] --] @var{expr}
19320 @itemx compile print [[@var{options}] --] /@var{f} @var{expr}
19321 Compile and execute @var{expr} with the compiler language found as the
19322 current language in @value{GDBN} (@pxref{Languages}). By default the
19323 value of @var{expr} is printed in a format appropriate to its data type;
19324 you can choose a different format by specifying @samp{/@var{f}}, where
19325 @var{f} is a letter specifying the format; see @ref{Output Formats,,Output
19326 Formats}. The @code{compile print} command accepts the same options
19327 as the @code{print} command; see @ref{print options}.
19329 @item compile print [[@var{options}] --]
19330 @itemx compile print [[@var{options}] --] /@var{f}
19331 @cindex reprint the last value
19332 Alternatively you can enter the expression (source code producing it) as
19333 multiple lines of text. To enter this mode, invoke the @samp{compile print}
19334 command without any text following the command. This will start the
19335 multiple-line editor.
19339 The process of compiling and injecting the code can be inspected using:
19342 @anchor{set debug compile}
19343 @item set debug compile
19344 @cindex compile command debugging info
19345 Turns on or off display of @value{GDBN} process of compiling and
19346 injecting the code. The default is off.
19348 @item show debug compile
19349 Displays the current state of displaying @value{GDBN} process of
19350 compiling and injecting the code.
19352 @anchor{set debug compile-cplus-types}
19353 @item set debug compile-cplus-types
19354 @cindex compile C@t{++} type conversion
19355 Turns on or off the display of C@t{++} type conversion debugging information.
19356 The default is off.
19358 @item show debug compile-cplus-types
19359 Displays the current state of displaying debugging information for
19360 C@t{++} type conversion.
19363 @subsection Compilation options for the @code{compile} command
19365 @value{GDBN} needs to specify the right compilation options for the code
19366 to be injected, in part to make its ABI compatible with the inferior
19367 and in part to make the injected code compatible with @value{GDBN}'s
19371 The options used, in increasing precedence:
19374 @item target architecture and OS options (@code{gdbarch})
19375 These options depend on target processor type and target operating
19376 system, usually they specify at least 32-bit (@code{-m32}) or 64-bit
19377 (@code{-m64}) compilation option.
19379 @item compilation options recorded in the target
19380 @value{NGCC} (since version 4.7) stores the options used for compilation
19381 into @code{DW_AT_producer} part of DWARF debugging information according
19382 to the @value{NGCC} option @code{-grecord-gcc-switches}. One has to
19383 explicitly specify @code{-g} during inferior compilation otherwise
19384 @value{NGCC} produces no DWARF. This feature is only relevant for
19385 platforms where @code{-g} produces DWARF by default, otherwise one may
19386 try to enforce DWARF by using @code{-gdwarf-4}.
19388 @item compilation options set by @code{set compile-args}
19392 You can override compilation options using the following command:
19395 @item set compile-args
19396 @cindex compile command options override
19397 Set compilation options used for compiling and injecting code with the
19398 @code{compile} commands. These options override any conflicting ones
19399 from the target architecture and/or options stored during inferior
19402 @item show compile-args
19403 Displays the current state of compilation options override.
19404 This does not show all the options actually used during compilation,
19405 use @ref{set debug compile} for that.
19408 @subsection Caveats when using the @code{compile} command
19410 There are a few caveats to keep in mind when using the @code{compile}
19411 command. As the caveats are different per language, the table below
19412 highlights specific issues on a per language basis.
19415 @item C code examples and caveats
19416 When the language in @value{GDBN} is set to @samp{C}, the compiler will
19417 attempt to compile the source code with a @samp{C} compiler. The source
19418 code provided to the @code{compile} command will have much the same
19419 access to variables and types as it normally would if it were part of
19420 the program currently being debugged in @value{GDBN}.
19422 Below is a sample program that forms the basis of the examples that
19423 follow. This program has been compiled and loaded into @value{GDBN},
19424 much like any other normal debugging session.
19427 void function1 (void)
19430 printf ("function 1\n");
19433 void function2 (void)
19448 For the purposes of the examples in this section, the program above has
19449 been compiled, loaded into @value{GDBN}, stopped at the function
19450 @code{main}, and @value{GDBN} is awaiting input from the user.
19452 To access variables and types for any program in @value{GDBN}, the
19453 program must be compiled and packaged with debug information. The
19454 @code{compile} command is not an exception to this rule. Without debug
19455 information, you can still use the @code{compile} command, but you will
19456 be very limited in what variables and types you can access.
19458 So with that in mind, the example above has been compiled with debug
19459 information enabled. The @code{compile} command will have access to
19460 all variables and types (except those that may have been optimized
19461 out). Currently, as @value{GDBN} has stopped the program in the
19462 @code{main} function, the @code{compile} command would have access to
19463 the variable @code{k}. You could invoke the @code{compile} command
19464 and type some source code to set the value of @code{k}. You can also
19465 read it, or do anything with that variable you would normally do in
19466 @code{C}. Be aware that changes to inferior variables in the
19467 @code{compile} command are persistent. In the following example:
19470 compile code k = 3;
19474 the variable @code{k} is now 3. It will retain that value until
19475 something else in the example program changes it, or another
19476 @code{compile} command changes it.
19478 Normal scope and access rules apply to source code compiled and
19479 injected by the @code{compile} command. In the example, the variables
19480 @code{j} and @code{k} are not accessible yet, because the program is
19481 currently stopped in the @code{main} function, where these variables
19482 are not in scope. Therefore, the following command
19485 compile code j = 3;
19489 will result in a compilation error message.
19491 Once the program is continued, execution will bring these variables in
19492 scope, and they will become accessible; then the code you specify via
19493 the @code{compile} command will be able to access them.
19495 You can create variables and types with the @code{compile} command as
19496 part of your source code. Variables and types that are created as part
19497 of the @code{compile} command are not visible to the rest of the program for
19498 the duration of its run. This example is valid:
19501 compile code int ff = 5; printf ("ff is %d\n", ff);
19504 However, if you were to type the following into @value{GDBN} after that
19505 command has completed:
19508 compile code printf ("ff is %d\n'', ff);
19512 a compiler error would be raised as the variable @code{ff} no longer
19513 exists. Object code generated and injected by the @code{compile}
19514 command is removed when its execution ends. Caution is advised
19515 when assigning to program variables values of variables created by the
19516 code submitted to the @code{compile} command. This example is valid:
19519 compile code int ff = 5; k = ff;
19522 The value of the variable @code{ff} is assigned to @code{k}. The variable
19523 @code{k} does not require the existence of @code{ff} to maintain the value
19524 it has been assigned. However, pointers require particular care in
19525 assignment. If the source code compiled with the @code{compile} command
19526 changed the address of a pointer in the example program, perhaps to a
19527 variable created in the @code{compile} command, that pointer would point
19528 to an invalid location when the command exits. The following example
19529 would likely cause issues with your debugged program:
19532 compile code int ff = 5; p = &ff;
19535 In this example, @code{p} would point to @code{ff} when the
19536 @code{compile} command is executing the source code provided to it.
19537 However, as variables in the (example) program persist with their
19538 assigned values, the variable @code{p} would point to an invalid
19539 location when the command exists. A general rule should be followed
19540 in that you should either assign @code{NULL} to any assigned pointers,
19541 or restore a valid location to the pointer before the command exits.
19543 Similar caution must be exercised with any structs, unions, and typedefs
19544 defined in @code{compile} command. Types defined in the @code{compile}
19545 command will no longer be available in the next @code{compile} command.
19546 Therefore, if you cast a variable to a type defined in the
19547 @code{compile} command, care must be taken to ensure that any future
19548 need to resolve the type can be achieved.
19551 (gdb) compile code static struct a @{ int a; @} v = @{ 42 @}; argv = &v;
19552 (gdb) compile code printf ("%d\n", ((struct a *) argv)->a);
19553 gdb command line:1:36: error: dereferencing pointer to incomplete type ‘struct a’
19554 Compilation failed.
19555 (gdb) compile code struct a @{ int a; @}; printf ("%d\n", ((struct a *) argv)->a);
19559 Variables that have been optimized away by the compiler are not
19560 accessible to the code submitted to the @code{compile} command.
19561 Access to those variables will generate a compiler error which @value{GDBN}
19562 will print to the console.
19565 @subsection Compiler search for the @code{compile} command
19567 @value{GDBN} needs to find @value{NGCC} for the inferior being debugged
19568 which may not be obvious for remote targets of different architecture
19569 than where @value{GDBN} is running. Environment variable @code{PATH} on
19570 @value{GDBN} host is searched for @value{NGCC} binary matching the
19571 target architecture and operating system. This search can be overriden
19572 by @code{set compile-gcc} @value{GDBN} command below. @code{PATH} is
19573 taken from shell that executed @value{GDBN}, it is not the value set by
19574 @value{GDBN} command @code{set environment}). @xref{Environment}.
19577 Specifically @code{PATH} is searched for binaries matching regular expression
19578 @code{@var{arch}(-[^-]*)?-@var{os}-gcc} according to the inferior target being
19579 debugged. @var{arch} is processor name --- multiarch is supported, so for
19580 example both @code{i386} and @code{x86_64} targets look for pattern
19581 @code{(x86_64|i.86)} and both @code{s390} and @code{s390x} targets look
19582 for pattern @code{s390x?}. @var{os} is currently supported only for
19583 pattern @code{linux(-gnu)?}.
19585 On Posix hosts the compiler driver @value{GDBN} needs to find also
19586 shared library @file{libcc1.so} from the compiler. It is searched in
19587 default shared library search path (overridable with usual environment
19588 variable @code{LD_LIBRARY_PATH}), unrelated to @code{PATH} or @code{set
19589 compile-gcc} settings. Contrary to it @file{libcc1plugin.so} is found
19590 according to the installation of the found compiler --- as possibly
19591 specified by the @code{set compile-gcc} command.
19594 @item set compile-gcc
19595 @cindex compile command driver filename override
19596 Set compilation command used for compiling and injecting code with the
19597 @code{compile} commands. If this option is not set (it is set to
19598 an empty string), the search described above will occur --- that is the
19601 @item show compile-gcc
19602 Displays the current compile command @value{NGCC} driver filename.
19603 If set, it is the main command @command{gcc}, found usually for example
19604 under name @file{x86_64-linux-gnu-gcc}.
19608 @chapter @value{GDBN} Files
19610 @value{GDBN} needs to know the file name of the program to be debugged,
19611 both in order to read its symbol table and in order to start your
19612 program. To debug a core dump of a previous run, you must also tell
19613 @value{GDBN} the name of the core dump file.
19616 * Files:: Commands to specify files
19617 * File Caching:: Information about @value{GDBN}'s file caching
19618 * Separate Debug Files:: Debugging information in separate files
19619 * MiniDebugInfo:: Debugging information in a special section
19620 * Index Files:: Index files speed up GDB
19621 * Symbol Errors:: Errors reading symbol files
19622 * Data Files:: GDB data files
19626 @section Commands to Specify Files
19628 @cindex symbol table
19629 @cindex core dump file
19631 You may want to specify executable and core dump file names. The usual
19632 way to do this is at start-up time, using the arguments to
19633 @value{GDBN}'s start-up commands (@pxref{Invocation, , Getting In and
19634 Out of @value{GDBN}}).
19636 Occasionally it is necessary to change to a different file during a
19637 @value{GDBN} session. Or you may run @value{GDBN} and forget to
19638 specify a file you want to use. Or you are debugging a remote target
19639 via @code{gdbserver} (@pxref{Server, file, Using the @code{gdbserver}
19640 Program}). In these situations the @value{GDBN} commands to specify
19641 new files are useful.
19644 @cindex executable file
19646 @item file @var{filename}
19647 Use @var{filename} as the program to be debugged. It is read for its
19648 symbols and for the contents of pure memory. It is also the program
19649 executed when you use the @code{run} command. If you do not specify a
19650 directory and the file is not found in the @value{GDBN} working directory,
19651 @value{GDBN} uses the environment variable @code{PATH} as a list of
19652 directories to search, just as the shell does when looking for a program
19653 to run. You can change the value of this variable, for both @value{GDBN}
19654 and your program, using the @code{path} command.
19656 @cindex unlinked object files
19657 @cindex patching object files
19658 You can load unlinked object @file{.o} files into @value{GDBN} using
19659 the @code{file} command. You will not be able to ``run'' an object
19660 file, but you can disassemble functions and inspect variables. Also,
19661 if the underlying BFD functionality supports it, you could use
19662 @kbd{gdb -write} to patch object files using this technique. Note
19663 that @value{GDBN} can neither interpret nor modify relocations in this
19664 case, so branches and some initialized variables will appear to go to
19665 the wrong place. But this feature is still handy from time to time.
19668 @code{file} with no argument makes @value{GDBN} discard any information it
19669 has on both executable file and the symbol table.
19672 @item exec-file @r{[} @var{filename} @r{]}
19673 Specify that the program to be run (but not the symbol table) is found
19674 in @var{filename}. @value{GDBN} searches the environment variable @code{PATH}
19675 if necessary to locate your program. Omitting @var{filename} means to
19676 discard information on the executable file.
19678 @kindex symbol-file
19679 @item symbol-file @r{[} @var{filename} @r{[} -o @var{offset} @r{]]}
19680 Read symbol table information from file @var{filename}. @code{PATH} is
19681 searched when necessary. Use the @code{file} command to get both symbol
19682 table and program to run from the same file.
19684 If an optional @var{offset} is specified, it is added to the start
19685 address of each section in the symbol file. This is useful if the
19686 program is relocated at runtime, such as the Linux kernel with kASLR
19689 @code{symbol-file} with no argument clears out @value{GDBN} information on your
19690 program's symbol table.
19692 The @code{symbol-file} command causes @value{GDBN} to forget the contents of
19693 some breakpoints and auto-display expressions. This is because they may
19694 contain pointers to the internal data recording symbols and data types,
19695 which are part of the old symbol table data being discarded inside
19698 @code{symbol-file} does not repeat if you press @key{RET} again after
19701 When @value{GDBN} is configured for a particular environment, it
19702 understands debugging information in whatever format is the standard
19703 generated for that environment; you may use either a @sc{gnu} compiler, or
19704 other compilers that adhere to the local conventions.
19705 Best results are usually obtained from @sc{gnu} compilers; for example,
19706 using @code{@value{NGCC}} you can generate debugging information for
19709 For most kinds of object files, with the exception of old SVR3 systems
19710 using COFF, the @code{symbol-file} command does not normally read the
19711 symbol table in full right away. Instead, it scans the symbol table
19712 quickly to find which source files and which symbols are present. The
19713 details are read later, one source file at a time, as they are needed.
19715 The purpose of this two-stage reading strategy is to make @value{GDBN}
19716 start up faster. For the most part, it is invisible except for
19717 occasional pauses while the symbol table details for a particular source
19718 file are being read. (The @code{set verbose} command can turn these
19719 pauses into messages if desired. @xref{Messages/Warnings, ,Optional
19720 Warnings and Messages}.)
19722 We have not implemented the two-stage strategy for COFF yet. When the
19723 symbol table is stored in COFF format, @code{symbol-file} reads the
19724 symbol table data in full right away. Note that ``stabs-in-COFF''
19725 still does the two-stage strategy, since the debug info is actually
19729 @cindex reading symbols immediately
19730 @cindex symbols, reading immediately
19731 @item symbol-file @r{[} -readnow @r{]} @var{filename}
19732 @itemx file @r{[} -readnow @r{]} @var{filename}
19733 You can override the @value{GDBN} two-stage strategy for reading symbol
19734 tables by using the @samp{-readnow} option with any of the commands that
19735 load symbol table information, if you want to be sure @value{GDBN} has the
19736 entire symbol table available.
19738 @cindex @code{-readnever}, option for symbol-file command
19739 @cindex never read symbols
19740 @cindex symbols, never read
19741 @item symbol-file @r{[} -readnever @r{]} @var{filename}
19742 @itemx file @r{[} -readnever @r{]} @var{filename}
19743 You can instruct @value{GDBN} to never read the symbolic information
19744 contained in @var{filename} by using the @samp{-readnever} option.
19745 @xref{--readnever}.
19747 @c FIXME: for now no mention of directories, since this seems to be in
19748 @c flux. 13mar1992 status is that in theory GDB would look either in
19749 @c current dir or in same dir as myprog; but issues like competing
19750 @c GDB's, or clutter in system dirs, mean that in practice right now
19751 @c only current dir is used. FFish says maybe a special GDB hierarchy
19752 @c (eg rooted in val of env var GDBSYMS) could exist for mappable symbol
19756 @item core-file @r{[}@var{filename}@r{]}
19758 Specify the whereabouts of a core dump file to be used as the ``contents
19759 of memory''. Traditionally, core files contain only some parts of the
19760 address space of the process that generated them; @value{GDBN} can access the
19761 executable file itself for other parts.
19763 @code{core-file} with no argument specifies that no core file is
19766 Note that the core file is ignored when your program is actually running
19767 under @value{GDBN}. So, if you have been running your program and you
19768 wish to debug a core file instead, you must kill the subprocess in which
19769 the program is running. To do this, use the @code{kill} command
19770 (@pxref{Kill Process, ,Killing the Child Process}).
19772 @kindex add-symbol-file
19773 @cindex dynamic linking
19774 @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{]}
19775 The @code{add-symbol-file} command reads additional symbol table
19776 information from the file @var{filename}. You would use this command
19777 when @var{filename} has been dynamically loaded (by some other means)
19778 into the program that is running. The @var{textaddress} parameter gives
19779 the memory address at which the file's text section has been loaded.
19780 You can additionally specify the base address of other sections using
19781 an arbitrary number of @samp{-s @var{section} @var{address}} pairs.
19782 If a section is omitted, @value{GDBN} will use its default addresses
19783 as found in @var{filename}. Any @var{address} or @var{textaddress}
19784 can be given as an expression.
19786 If an optional @var{offset} is specified, it is added to the start
19787 address of each section, except those for which the address was
19788 specified explicitly.
19790 The symbol table of the file @var{filename} is added to the symbol table
19791 originally read with the @code{symbol-file} command. You can use the
19792 @code{add-symbol-file} command any number of times; the new symbol data
19793 thus read is kept in addition to the old.
19795 Changes can be reverted using the command @code{remove-symbol-file}.
19797 @cindex relocatable object files, reading symbols from
19798 @cindex object files, relocatable, reading symbols from
19799 @cindex reading symbols from relocatable object files
19800 @cindex symbols, reading from relocatable object files
19801 @cindex @file{.o} files, reading symbols from
19802 Although @var{filename} is typically a shared library file, an
19803 executable file, or some other object file which has been fully
19804 relocated for loading into a process, you can also load symbolic
19805 information from relocatable @file{.o} files, as long as:
19809 the file's symbolic information refers only to linker symbols defined in
19810 that file, not to symbols defined by other object files,
19812 every section the file's symbolic information refers to has actually
19813 been loaded into the inferior, as it appears in the file, and
19815 you can determine the address at which every section was loaded, and
19816 provide these to the @code{add-symbol-file} command.
19820 Some embedded operating systems, like Sun Chorus and VxWorks, can load
19821 relocatable files into an already running program; such systems
19822 typically make the requirements above easy to meet. However, it's
19823 important to recognize that many native systems use complex link
19824 procedures (@code{.linkonce} section factoring and C@t{++} constructor table
19825 assembly, for example) that make the requirements difficult to meet. In
19826 general, one cannot assume that using @code{add-symbol-file} to read a
19827 relocatable object file's symbolic information will have the same effect
19828 as linking the relocatable object file into the program in the normal
19831 @code{add-symbol-file} does not repeat if you press @key{RET} after using it.
19833 @kindex remove-symbol-file
19834 @item remove-symbol-file @var{filename}
19835 @item remove-symbol-file -a @var{address}
19836 Remove a symbol file added via the @code{add-symbol-file} command. The
19837 file to remove can be identified by its @var{filename} or by an @var{address}
19838 that lies within the boundaries of this symbol file in memory. Example:
19841 (gdb) add-symbol-file /home/user/gdb/mylib.so 0x7ffff7ff9480
19842 add symbol table from file "/home/user/gdb/mylib.so" at
19843 .text_addr = 0x7ffff7ff9480
19845 Reading symbols from /home/user/gdb/mylib.so...done.
19846 (gdb) remove-symbol-file -a 0x7ffff7ff9480
19847 Remove symbol table from file "/home/user/gdb/mylib.so"? (y or n) y
19852 @code{remove-symbol-file} does not repeat if you press @key{RET} after using it.
19854 @kindex add-symbol-file-from-memory
19855 @cindex @code{syscall DSO}
19856 @cindex load symbols from memory
19857 @item add-symbol-file-from-memory @var{address}
19858 Load symbols from the given @var{address} in a dynamically loaded
19859 object file whose image is mapped directly into the inferior's memory.
19860 For example, the Linux kernel maps a @code{syscall DSO} into each
19861 process's address space; this DSO provides kernel-specific code for
19862 some system calls. The argument can be any expression whose
19863 evaluation yields the address of the file's shared object file header.
19864 For this command to work, you must have used @code{symbol-file} or
19865 @code{exec-file} commands in advance.
19868 @item section @var{section} @var{addr}
19869 The @code{section} command changes the base address of the named
19870 @var{section} of the exec file to @var{addr}. This can be used if the
19871 exec file does not contain section addresses, (such as in the
19872 @code{a.out} format), or when the addresses specified in the file
19873 itself are wrong. Each section must be changed separately. The
19874 @code{info files} command, described below, lists all the sections and
19878 @kindex info target
19881 @code{info files} and @code{info target} are synonymous; both print the
19882 current target (@pxref{Targets, ,Specifying a Debugging Target}),
19883 including the names of the executable and core dump files currently in
19884 use by @value{GDBN}, and the files from which symbols were loaded. The
19885 command @code{help target} lists all possible targets rather than
19888 @kindex maint info sections
19889 @item maint info sections
19890 Another command that can give you extra information about program sections
19891 is @code{maint info sections}. In addition to the section information
19892 displayed by @code{info files}, this command displays the flags and file
19893 offset of each section in the executable and core dump files. In addition,
19894 @code{maint info sections} provides the following command options (which
19895 may be arbitrarily combined):
19899 Display sections for all loaded object files, including shared libraries.
19900 @item @var{sections}
19901 Display info only for named @var{sections}.
19902 @item @var{section-flags}
19903 Display info only for sections for which @var{section-flags} are true.
19904 The section flags that @value{GDBN} currently knows about are:
19907 Section will have space allocated in the process when loaded.
19908 Set for all sections except those containing debug information.
19910 Section will be loaded from the file into the child process memory.
19911 Set for pre-initialized code and data, clear for @code{.bss} sections.
19913 Section needs to be relocated before loading.
19915 Section cannot be modified by the child process.
19917 Section contains executable code only.
19919 Section contains data only (no executable code).
19921 Section will reside in ROM.
19923 Section contains data for constructor/destructor lists.
19925 Section is not empty.
19927 An instruction to the linker to not output the section.
19928 @item COFF_SHARED_LIBRARY
19929 A notification to the linker that the section contains
19930 COFF shared library information.
19932 Section contains common symbols.
19935 @kindex set trust-readonly-sections
19936 @cindex read-only sections
19937 @item set trust-readonly-sections on
19938 Tell @value{GDBN} that readonly sections in your object file
19939 really are read-only (i.e.@: that their contents will not change).
19940 In that case, @value{GDBN} can fetch values from these sections
19941 out of the object file, rather than from the target program.
19942 For some targets (notably embedded ones), this can be a significant
19943 enhancement to debugging performance.
19945 The default is off.
19947 @item set trust-readonly-sections off
19948 Tell @value{GDBN} not to trust readonly sections. This means that
19949 the contents of the section might change while the program is running,
19950 and must therefore be fetched from the target when needed.
19952 @item show trust-readonly-sections
19953 Show the current setting of trusting readonly sections.
19956 All file-specifying commands allow both absolute and relative file names
19957 as arguments. @value{GDBN} always converts the file name to an absolute file
19958 name and remembers it that way.
19960 @cindex shared libraries
19961 @anchor{Shared Libraries}
19962 @value{GDBN} supports @sc{gnu}/Linux, MS-Windows, SunOS,
19963 Darwin/Mach-O, SVr4, IBM RS/6000 AIX, QNX Neutrino, FDPIC (FR-V), and
19964 DSBT (TIC6X) shared libraries.
19966 On MS-Windows @value{GDBN} must be linked with the Expat library to support
19967 shared libraries. @xref{Expat}.
19969 @value{GDBN} automatically loads symbol definitions from shared libraries
19970 when you use the @code{run} command, or when you examine a core file.
19971 (Before you issue the @code{run} command, @value{GDBN} does not understand
19972 references to a function in a shared library, however---unless you are
19973 debugging a core file).
19975 @c FIXME: some @value{GDBN} release may permit some refs to undef
19976 @c FIXME...symbols---eg in a break cmd---assuming they are from a shared
19977 @c FIXME...lib; check this from time to time when updating manual
19979 There are times, however, when you may wish to not automatically load
19980 symbol definitions from shared libraries, such as when they are
19981 particularly large or there are many of them.
19983 To control the automatic loading of shared library symbols, use the
19987 @kindex set auto-solib-add
19988 @item set auto-solib-add @var{mode}
19989 If @var{mode} is @code{on}, symbols from all shared object libraries
19990 will be loaded automatically when the inferior begins execution, you
19991 attach to an independently started inferior, or when the dynamic linker
19992 informs @value{GDBN} that a new library has been loaded. If @var{mode}
19993 is @code{off}, symbols must be loaded manually, using the
19994 @code{sharedlibrary} command. The default value is @code{on}.
19996 @cindex memory used for symbol tables
19997 If your program uses lots of shared libraries with debug info that
19998 takes large amounts of memory, you can decrease the @value{GDBN}
19999 memory footprint by preventing it from automatically loading the
20000 symbols from shared libraries. To that end, type @kbd{set
20001 auto-solib-add off} before running the inferior, then load each
20002 library whose debug symbols you do need with @kbd{sharedlibrary
20003 @var{regexp}}, where @var{regexp} is a regular expression that matches
20004 the libraries whose symbols you want to be loaded.
20006 @kindex show auto-solib-add
20007 @item show auto-solib-add
20008 Display the current autoloading mode.
20011 @cindex load shared library
20012 To explicitly load shared library symbols, use the @code{sharedlibrary}
20016 @kindex info sharedlibrary
20018 @item info share @var{regex}
20019 @itemx info sharedlibrary @var{regex}
20020 Print the names of the shared libraries which are currently loaded
20021 that match @var{regex}. If @var{regex} is omitted then print
20022 all shared libraries that are loaded.
20025 @item info dll @var{regex}
20026 This is an alias of @code{info sharedlibrary}.
20028 @kindex sharedlibrary
20030 @item sharedlibrary @var{regex}
20031 @itemx share @var{regex}
20032 Load shared object library symbols for files matching a
20033 Unix regular expression.
20034 As with files loaded automatically, it only loads shared libraries
20035 required by your program for a core file or after typing @code{run}. If
20036 @var{regex} is omitted all shared libraries required by your program are
20039 @item nosharedlibrary
20040 @kindex nosharedlibrary
20041 @cindex unload symbols from shared libraries
20042 Unload all shared object library symbols. This discards all symbols
20043 that have been loaded from all shared libraries. Symbols from shared
20044 libraries that were loaded by explicit user requests are not
20048 Sometimes you may wish that @value{GDBN} stops and gives you control
20049 when any of shared library events happen. The best way to do this is
20050 to use @code{catch load} and @code{catch unload} (@pxref{Set
20053 @value{GDBN} also supports the the @code{set stop-on-solib-events}
20054 command for this. This command exists for historical reasons. It is
20055 less useful than setting a catchpoint, because it does not allow for
20056 conditions or commands as a catchpoint does.
20059 @item set stop-on-solib-events
20060 @kindex set stop-on-solib-events
20061 This command controls whether @value{GDBN} should give you control
20062 when the dynamic linker notifies it about some shared library event.
20063 The most common event of interest is loading or unloading of a new
20066 @item show stop-on-solib-events
20067 @kindex show stop-on-solib-events
20068 Show whether @value{GDBN} stops and gives you control when shared
20069 library events happen.
20072 Shared libraries are also supported in many cross or remote debugging
20073 configurations. @value{GDBN} needs to have access to the target's libraries;
20074 this can be accomplished either by providing copies of the libraries
20075 on the host system, or by asking @value{GDBN} to automatically retrieve the
20076 libraries from the target. If copies of the target libraries are
20077 provided, they need to be the same as the target libraries, although the
20078 copies on the target can be stripped as long as the copies on the host are
20081 @cindex where to look for shared libraries
20082 For remote debugging, you need to tell @value{GDBN} where the target
20083 libraries are, so that it can load the correct copies---otherwise, it
20084 may try to load the host's libraries. @value{GDBN} has two variables
20085 to specify the search directories for target libraries.
20088 @cindex prefix for executable and shared library file names
20089 @cindex system root, alternate
20090 @kindex set solib-absolute-prefix
20091 @kindex set sysroot
20092 @item set sysroot @var{path}
20093 Use @var{path} as the system root for the program being debugged. Any
20094 absolute shared library paths will be prefixed with @var{path}; many
20095 runtime loaders store the absolute paths to the shared library in the
20096 target program's memory. When starting processes remotely, and when
20097 attaching to already-running processes (local or remote), their
20098 executable filenames will be prefixed with @var{path} if reported to
20099 @value{GDBN} as absolute by the operating system. If you use
20100 @code{set sysroot} to find executables and shared libraries, they need
20101 to be laid out in the same way that they are on the target, with
20102 e.g.@: a @file{/bin}, @file{/lib} and @file{/usr/lib} hierarchy under
20105 If @var{path} starts with the sequence @file{target:} and the target
20106 system is remote then @value{GDBN} will retrieve the target binaries
20107 from the remote system. This is only supported when using a remote
20108 target that supports the @code{remote get} command (@pxref{File
20109 Transfer,,Sending files to a remote system}). The part of @var{path}
20110 following the initial @file{target:} (if present) is used as system
20111 root prefix on the remote file system. If @var{path} starts with the
20112 sequence @file{remote:} this is converted to the sequence
20113 @file{target:} by @code{set sysroot}@footnote{Historically the
20114 functionality to retrieve binaries from the remote system was
20115 provided by prefixing @var{path} with @file{remote:}}. If you want
20116 to specify a local system root using a directory that happens to be
20117 named @file{target:} or @file{remote:}, you need to use some
20118 equivalent variant of the name like @file{./target:}.
20120 For targets with an MS-DOS based filesystem, such as MS-Windows and
20121 SymbianOS, @value{GDBN} tries prefixing a few variants of the target
20122 absolute file name with @var{path}. But first, on Unix hosts,
20123 @value{GDBN} converts all backslash directory separators into forward
20124 slashes, because the backslash is not a directory separator on Unix:
20127 c:\foo\bar.dll @result{} c:/foo/bar.dll
20130 Then, @value{GDBN} attempts prefixing the target file name with
20131 @var{path}, and looks for the resulting file name in the host file
20135 c:/foo/bar.dll @result{} /path/to/sysroot/c:/foo/bar.dll
20138 If that does not find the binary, @value{GDBN} tries removing
20139 the @samp{:} character from the drive spec, both for convenience, and,
20140 for the case of the host file system not supporting file names with
20144 c:/foo/bar.dll @result{} /path/to/sysroot/c/foo/bar.dll
20147 This makes it possible to have a system root that mirrors a target
20148 with more than one drive. E.g., you may want to setup your local
20149 copies of the target system shared libraries like so (note @samp{c} vs
20153 @file{/path/to/sysroot/c/sys/bin/foo.dll}
20154 @file{/path/to/sysroot/c/sys/bin/bar.dll}
20155 @file{/path/to/sysroot/z/sys/bin/bar.dll}
20159 and point the system root at @file{/path/to/sysroot}, so that
20160 @value{GDBN} can find the correct copies of both
20161 @file{c:\sys\bin\foo.dll}, and @file{z:\sys\bin\bar.dll}.
20163 If that still does not find the binary, @value{GDBN} tries
20164 removing the whole drive spec from the target file name:
20167 c:/foo/bar.dll @result{} /path/to/sysroot/foo/bar.dll
20170 This last lookup makes it possible to not care about the drive name,
20171 if you don't want or need to.
20173 The @code{set solib-absolute-prefix} command is an alias for @code{set
20176 @cindex default system root
20177 @cindex @samp{--with-sysroot}
20178 You can set the default system root by using the configure-time
20179 @samp{--with-sysroot} option. If the system root is inside
20180 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
20181 @samp{--exec-prefix}), then the default system root will be updated
20182 automatically if the installed @value{GDBN} is moved to a new
20185 @kindex show sysroot
20187 Display the current executable and shared library prefix.
20189 @kindex set solib-search-path
20190 @item set solib-search-path @var{path}
20191 If this variable is set, @var{path} is a colon-separated list of
20192 directories to search for shared libraries. @samp{solib-search-path}
20193 is used after @samp{sysroot} fails to locate the library, or if the
20194 path to the library is relative instead of absolute. If you want to
20195 use @samp{solib-search-path} instead of @samp{sysroot}, be sure to set
20196 @samp{sysroot} to a nonexistent directory to prevent @value{GDBN} from
20197 finding your host's libraries. @samp{sysroot} is preferred; setting
20198 it to a nonexistent directory may interfere with automatic loading
20199 of shared library symbols.
20201 @kindex show solib-search-path
20202 @item show solib-search-path
20203 Display the current shared library search path.
20205 @cindex DOS file-name semantics of file names.
20206 @kindex set target-file-system-kind (unix|dos-based|auto)
20207 @kindex show target-file-system-kind
20208 @item set target-file-system-kind @var{kind}
20209 Set assumed file system kind for target reported file names.
20211 Shared library file names as reported by the target system may not
20212 make sense as is on the system @value{GDBN} is running on. For
20213 example, when remote debugging a target that has MS-DOS based file
20214 system semantics, from a Unix host, the target may be reporting to
20215 @value{GDBN} a list of loaded shared libraries with file names such as
20216 @file{c:\Windows\kernel32.dll}. On Unix hosts, there's no concept of
20217 drive letters, so the @samp{c:\} prefix is not normally understood as
20218 indicating an absolute file name, and neither is the backslash
20219 normally considered a directory separator character. In that case,
20220 the native file system would interpret this whole absolute file name
20221 as a relative file name with no directory components. This would make
20222 it impossible to point @value{GDBN} at a copy of the remote target's
20223 shared libraries on the host using @code{set sysroot}, and impractical
20224 with @code{set solib-search-path}. Setting
20225 @code{target-file-system-kind} to @code{dos-based} tells @value{GDBN}
20226 to interpret such file names similarly to how the target would, and to
20227 map them to file names valid on @value{GDBN}'s native file system
20228 semantics. The value of @var{kind} can be @code{"auto"}, in addition
20229 to one of the supported file system kinds. In that case, @value{GDBN}
20230 tries to determine the appropriate file system variant based on the
20231 current target's operating system (@pxref{ABI, ,Configuring the
20232 Current ABI}). The supported file system settings are:
20236 Instruct @value{GDBN} to assume the target file system is of Unix
20237 kind. Only file names starting the forward slash (@samp{/}) character
20238 are considered absolute, and the directory separator character is also
20242 Instruct @value{GDBN} to assume the target file system is DOS based.
20243 File names starting with either a forward slash, or a drive letter
20244 followed by a colon (e.g., @samp{c:}), are considered absolute, and
20245 both the slash (@samp{/}) and the backslash (@samp{\\}) characters are
20246 considered directory separators.
20249 Instruct @value{GDBN} to use the file system kind associated with the
20250 target operating system (@pxref{ABI, ,Configuring the Current ABI}).
20251 This is the default.
20255 @cindex file name canonicalization
20256 @cindex base name differences
20257 When processing file names provided by the user, @value{GDBN}
20258 frequently needs to compare them to the file names recorded in the
20259 program's debug info. Normally, @value{GDBN} compares just the
20260 @dfn{base names} of the files as strings, which is reasonably fast
20261 even for very large programs. (The base name of a file is the last
20262 portion of its name, after stripping all the leading directories.)
20263 This shortcut in comparison is based upon the assumption that files
20264 cannot have more than one base name. This is usually true, but
20265 references to files that use symlinks or similar filesystem
20266 facilities violate that assumption. If your program records files
20267 using such facilities, or if you provide file names to @value{GDBN}
20268 using symlinks etc., you can set @code{basenames-may-differ} to
20269 @code{true} to instruct @value{GDBN} to completely canonicalize each
20270 pair of file names it needs to compare. This will make file-name
20271 comparisons accurate, but at a price of a significant slowdown.
20274 @item set basenames-may-differ
20275 @kindex set basenames-may-differ
20276 Set whether a source file may have multiple base names.
20278 @item show basenames-may-differ
20279 @kindex show basenames-may-differ
20280 Show whether a source file may have multiple base names.
20284 @section File Caching
20285 @cindex caching of opened files
20286 @cindex caching of bfd objects
20288 To speed up file loading, and reduce memory usage, @value{GDBN} will
20289 reuse the @code{bfd} objects used to track open files. @xref{Top, ,
20290 BFD, bfd, The Binary File Descriptor Library}. The following commands
20291 allow visibility and control of the caching behavior.
20294 @kindex maint info bfds
20295 @item maint info bfds
20296 This prints information about each @code{bfd} object that is known to
20299 @kindex maint set bfd-sharing
20300 @kindex maint show bfd-sharing
20301 @kindex bfd caching
20302 @item maint set bfd-sharing
20303 @item maint show bfd-sharing
20304 Control whether @code{bfd} objects can be shared. When sharing is
20305 enabled @value{GDBN} reuses already open @code{bfd} objects rather
20306 than reopening the same file. Turning sharing off does not cause
20307 already shared @code{bfd} objects to be unshared, but all future files
20308 that are opened will create a new @code{bfd} object. Similarly,
20309 re-enabling sharing does not cause multiple existing @code{bfd}
20310 objects to be collapsed into a single shared @code{bfd} object.
20312 @kindex set debug bfd-cache @var{level}
20313 @kindex bfd caching
20314 @item set debug bfd-cache @var{level}
20315 Turns on debugging of the bfd cache, setting the level to @var{level}.
20317 @kindex show debug bfd-cache
20318 @kindex bfd caching
20319 @item show debug bfd-cache
20320 Show the current debugging level of the bfd cache.
20323 @node Separate Debug Files
20324 @section Debugging Information in Separate Files
20325 @cindex separate debugging information files
20326 @cindex debugging information in separate files
20327 @cindex @file{.debug} subdirectories
20328 @cindex debugging information directory, global
20329 @cindex global debugging information directories
20330 @cindex build ID, and separate debugging files
20331 @cindex @file{.build-id} directory
20333 @value{GDBN} allows you to put a program's debugging information in a
20334 file separate from the executable itself, in a way that allows
20335 @value{GDBN} to find and load the debugging information automatically.
20336 Since debugging information can be very large---sometimes larger
20337 than the executable code itself---some systems distribute debugging
20338 information for their executables in separate files, which users can
20339 install only when they need to debug a problem.
20341 @value{GDBN} supports two ways of specifying the separate debug info
20346 The executable contains a @dfn{debug link} that specifies the name of
20347 the separate debug info file. The separate debug file's name is
20348 usually @file{@var{executable}.debug}, where @var{executable} is the
20349 name of the corresponding executable file without leading directories
20350 (e.g., @file{ls.debug} for @file{/usr/bin/ls}). In addition, the
20351 debug link specifies a 32-bit @dfn{Cyclic Redundancy Check} (CRC)
20352 checksum for the debug file, which @value{GDBN} uses to validate that
20353 the executable and the debug file came from the same build.
20356 The executable contains a @dfn{build ID}, a unique bit string that is
20357 also present in the corresponding debug info file. (This is supported
20358 only on some operating systems, when using the ELF or PE file formats
20359 for binary files and the @sc{gnu} Binutils.) For more details about
20360 this feature, see the description of the @option{--build-id}
20361 command-line option in @ref{Options, , Command Line Options, ld,
20362 The GNU Linker}. The debug info file's name is not specified
20363 explicitly by the build ID, but can be computed from the build ID, see
20367 Depending on the way the debug info file is specified, @value{GDBN}
20368 uses two different methods of looking for the debug file:
20372 For the ``debug link'' method, @value{GDBN} looks up the named file in
20373 the directory of the executable file, then in a subdirectory of that
20374 directory named @file{.debug}, and finally under each one of the
20375 global debug directories, in a subdirectory whose name is identical to
20376 the leading directories of the executable's absolute file name. (On
20377 MS-Windows/MS-DOS, the drive letter of the executable's leading
20378 directories is converted to a one-letter subdirectory, i.e.@:
20379 @file{d:/usr/bin/} is converted to @file{/d/usr/bin/}, because Windows
20380 filesystems disallow colons in file names.)
20383 For the ``build ID'' method, @value{GDBN} looks in the
20384 @file{.build-id} subdirectory of each one of the global debug directories for
20385 a file named @file{@var{nn}/@var{nnnnnnnn}.debug}, where @var{nn} are the
20386 first 2 hex characters of the build ID bit string, and @var{nnnnnnnn}
20387 are the rest of the bit string. (Real build ID strings are 32 or more
20388 hex characters, not 10.)
20391 So, for example, suppose you ask @value{GDBN} to debug
20392 @file{/usr/bin/ls}, which has a debug link that specifies the
20393 file @file{ls.debug}, and a build ID whose value in hex is
20394 @code{abcdef1234}. If the list of the global debug directories includes
20395 @file{/usr/lib/debug}, then @value{GDBN} will look for the following
20396 debug information files, in the indicated order:
20400 @file{/usr/lib/debug/.build-id/ab/cdef1234.debug}
20402 @file{/usr/bin/ls.debug}
20404 @file{/usr/bin/.debug/ls.debug}
20406 @file{/usr/lib/debug/usr/bin/ls.debug}.
20409 @anchor{debug-file-directory}
20410 Global debugging info directories default to what is set by @value{GDBN}
20411 configure option @option{--with-separate-debug-dir}. During @value{GDBN} run
20412 you can also set the global debugging info directories, and view the list
20413 @value{GDBN} is currently using.
20417 @kindex set debug-file-directory
20418 @item set debug-file-directory @var{directories}
20419 Set the directories which @value{GDBN} searches for separate debugging
20420 information files to @var{directory}. Multiple path components can be set
20421 concatenating them by a path separator.
20423 @kindex show debug-file-directory
20424 @item show debug-file-directory
20425 Show the directories @value{GDBN} searches for separate debugging
20430 @cindex @code{.gnu_debuglink} sections
20431 @cindex debug link sections
20432 A debug link is a special section of the executable file named
20433 @code{.gnu_debuglink}. The section must contain:
20437 A filename, with any leading directory components removed, followed by
20440 zero to three bytes of padding, as needed to reach the next four-byte
20441 boundary within the section, and
20443 a four-byte CRC checksum, stored in the same endianness used for the
20444 executable file itself. The checksum is computed on the debugging
20445 information file's full contents by the function given below, passing
20446 zero as the @var{crc} argument.
20449 Any executable file format can carry a debug link, as long as it can
20450 contain a section named @code{.gnu_debuglink} with the contents
20453 @cindex @code{.note.gnu.build-id} sections
20454 @cindex build ID sections
20455 The build ID is a special section in the executable file (and in other
20456 ELF binary files that @value{GDBN} may consider). This section is
20457 often named @code{.note.gnu.build-id}, but that name is not mandatory.
20458 It contains unique identification for the built files---the ID remains
20459 the same across multiple builds of the same build tree. The default
20460 algorithm SHA1 produces 160 bits (40 hexadecimal characters) of the
20461 content for the build ID string. The same section with an identical
20462 value is present in the original built binary with symbols, in its
20463 stripped variant, and in the separate debugging information file.
20465 The debugging information file itself should be an ordinary
20466 executable, containing a full set of linker symbols, sections, and
20467 debugging information. The sections of the debugging information file
20468 should have the same names, addresses, and sizes as the original file,
20469 but they need not contain any data---much like a @code{.bss} section
20470 in an ordinary executable.
20472 The @sc{gnu} binary utilities (Binutils) package includes the
20473 @samp{objcopy} utility that can produce
20474 the separated executable / debugging information file pairs using the
20475 following commands:
20478 @kbd{objcopy --only-keep-debug foo foo.debug}
20483 These commands remove the debugging
20484 information from the executable file @file{foo} and place it in the file
20485 @file{foo.debug}. You can use the first, second or both methods to link the
20490 The debug link method needs the following additional command to also leave
20491 behind a debug link in @file{foo}:
20494 @kbd{objcopy --add-gnu-debuglink=foo.debug foo}
20497 Ulrich Drepper's @file{elfutils} package, starting with version 0.53, contains
20498 a version of the @code{strip} command such that the command @kbd{strip foo -f
20499 foo.debug} has the same functionality as the two @code{objcopy} commands and
20500 the @code{ln -s} command above, together.
20503 Build ID gets embedded into the main executable using @code{ld --build-id} or
20504 the @value{NGCC} counterpart @code{gcc -Wl,--build-id}. Build ID support plus
20505 compatibility fixes for debug files separation are present in @sc{gnu} binary
20506 utilities (Binutils) package since version 2.18.
20511 @cindex CRC algorithm definition
20512 The CRC used in @code{.gnu_debuglink} is the CRC-32 defined in
20513 IEEE 802.3 using the polynomial:
20515 @c TexInfo requires naked braces for multi-digit exponents for Tex
20516 @c output, but this causes HTML output to barf. HTML has to be set using
20517 @c raw commands. So we end up having to specify this equation in 2
20522 <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>
20523 + <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
20529 @math{x^{32} + x^{26} + x^{23} + x^{22} + x^{16} + x^{12} + x^{11}}
20530 @math{+ x^{10} + x^8 + x^7 + x^5 + x^4 + x^2 + x + 1}
20534 The function is computed byte at a time, taking the least
20535 significant bit of each byte first. The initial pattern
20536 @code{0xffffffff} is used, to ensure leading zeros affect the CRC and
20537 the final result is inverted to ensure trailing zeros also affect the
20540 @emph{Note:} This is the same CRC polynomial as used in handling the
20541 @dfn{Remote Serial Protocol} @code{qCRC} packet (@pxref{qCRC packet}).
20542 However in the case of the Remote Serial Protocol, the CRC is computed
20543 @emph{most} significant bit first, and the result is not inverted, so
20544 trailing zeros have no effect on the CRC value.
20546 To complete the description, we show below the code of the function
20547 which produces the CRC used in @code{.gnu_debuglink}. Inverting the
20548 initially supplied @code{crc} argument means that an initial call to
20549 this function passing in zero will start computing the CRC using
20552 @kindex gnu_debuglink_crc32
20555 gnu_debuglink_crc32 (unsigned long crc,
20556 unsigned char *buf, size_t len)
20558 static const unsigned long crc32_table[256] =
20560 0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419,
20561 0x706af48f, 0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4,
20562 0xe0d5e91e, 0x97d2d988, 0x09b64c2b, 0x7eb17cbd, 0xe7b82d07,
20563 0x90bf1d91, 0x1db71064, 0x6ab020f2, 0xf3b97148, 0x84be41de,
20564 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7, 0x136c9856,
20565 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
20566 0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4,
20567 0xa2677172, 0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b,
20568 0x35b5a8fa, 0x42b2986c, 0xdbbbc9d6, 0xacbcf940, 0x32d86ce3,
20569 0x45df5c75, 0xdcd60dcf, 0xabd13d59, 0x26d930ac, 0x51de003a,
20570 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423, 0xcfba9599,
20571 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
20572 0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190,
20573 0x01db7106, 0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f,
20574 0x9fbfe4a5, 0xe8b8d433, 0x7807c9a2, 0x0f00f934, 0x9609a88e,
20575 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d, 0x91646c97, 0xe6635c01,
20576 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e, 0x6c0695ed,
20577 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
20578 0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3,
20579 0xfbd44c65, 0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2,
20580 0x4adfa541, 0x3dd895d7, 0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a,
20581 0x346ed9fc, 0xad678846, 0xda60b8d0, 0x44042d73, 0x33031de5,
20582 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa, 0xbe0b1010,
20583 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
20584 0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17,
20585 0x2eb40d81, 0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6,
20586 0x03b6e20c, 0x74b1d29a, 0xead54739, 0x9dd277af, 0x04db2615,
20587 0x73dc1683, 0xe3630b12, 0x94643b84, 0x0d6d6a3e, 0x7a6a5aa8,
20588 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1, 0xf00f9344,
20589 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
20590 0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a,
20591 0x67dd4acc, 0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5,
20592 0xd6d6a3e8, 0xa1d1937e, 0x38d8c2c4, 0x4fdff252, 0xd1bb67f1,
20593 0xa6bc5767, 0x3fb506dd, 0x48b2364b, 0xd80d2bda, 0xaf0a1b4c,
20594 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55, 0x316e8eef,
20595 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
20596 0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe,
20597 0xb2bd0b28, 0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31,
20598 0x2cd99e8b, 0x5bdeae1d, 0x9b64c2b0, 0xec63f226, 0x756aa39c,
20599 0x026d930a, 0x9c0906a9, 0xeb0e363f, 0x72076785, 0x05005713,
20600 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38, 0x92d28e9b,
20601 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
20602 0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1,
20603 0x18b74777, 0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c,
20604 0x8f659eff, 0xf862ae69, 0x616bffd3, 0x166ccf45, 0xa00ae278,
20605 0xd70dd2ee, 0x4e048354, 0x3903b3c2, 0xa7672661, 0xd06016f7,
20606 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc, 0x40df0b66,
20607 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
20608 0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605,
20609 0xcdd70693, 0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8,
20610 0x5d681b02, 0x2a6f2b94, 0xb40bbe37, 0xc30c8ea1, 0x5a05df1b,
20613 unsigned char *end;
20615 crc = ~crc & 0xffffffff;
20616 for (end = buf + len; buf < end; ++buf)
20617 crc = crc32_table[(crc ^ *buf) & 0xff] ^ (crc >> 8);
20618 return ~crc & 0xffffffff;
20623 This computation does not apply to the ``build ID'' method.
20625 @node MiniDebugInfo
20626 @section Debugging information in a special section
20627 @cindex separate debug sections
20628 @cindex @samp{.gnu_debugdata} section
20630 Some systems ship pre-built executables and libraries that have a
20631 special @samp{.gnu_debugdata} section. This feature is called
20632 @dfn{MiniDebugInfo}. This section holds an LZMA-compressed object and
20633 is used to supply extra symbols for backtraces.
20635 The intent of this section is to provide extra minimal debugging
20636 information for use in simple backtraces. It is not intended to be a
20637 replacement for full separate debugging information (@pxref{Separate
20638 Debug Files}). The example below shows the intended use; however,
20639 @value{GDBN} does not currently put restrictions on what sort of
20640 debugging information might be included in the section.
20642 @value{GDBN} has support for this extension. If the section exists,
20643 then it is used provided that no other source of debugging information
20644 can be found, and that @value{GDBN} was configured with LZMA support.
20646 This section can be easily created using @command{objcopy} and other
20647 standard utilities:
20650 # Extract the dynamic symbols from the main binary, there is no need
20651 # to also have these in the normal symbol table.
20652 nm -D @var{binary} --format=posix --defined-only \
20653 | awk '@{ print $1 @}' | sort > dynsyms
20655 # Extract all the text (i.e. function) symbols from the debuginfo.
20656 # (Note that we actually also accept "D" symbols, for the benefit
20657 # of platforms like PowerPC64 that use function descriptors.)
20658 nm @var{binary} --format=posix --defined-only \
20659 | awk '@{ if ($2 == "T" || $2 == "t" || $2 == "D") print $1 @}' \
20662 # Keep all the function symbols not already in the dynamic symbol
20664 comm -13 dynsyms funcsyms > keep_symbols
20666 # Separate full debug info into debug binary.
20667 objcopy --only-keep-debug @var{binary} debug
20669 # Copy the full debuginfo, keeping only a minimal set of symbols and
20670 # removing some unnecessary sections.
20671 objcopy -S --remove-section .gdb_index --remove-section .comment \
20672 --keep-symbols=keep_symbols debug mini_debuginfo
20674 # Drop the full debug info from the original binary.
20675 strip --strip-all -R .comment @var{binary}
20677 # Inject the compressed data into the .gnu_debugdata section of the
20680 objcopy --add-section .gnu_debugdata=mini_debuginfo.xz @var{binary}
20684 @section Index Files Speed Up @value{GDBN}
20685 @cindex index files
20686 @cindex @samp{.gdb_index} section
20688 When @value{GDBN} finds a symbol file, it scans the symbols in the
20689 file in order to construct an internal symbol table. This lets most
20690 @value{GDBN} operations work quickly---at the cost of a delay early
20691 on. For large programs, this delay can be quite lengthy, so
20692 @value{GDBN} provides a way to build an index, which speeds up
20695 For convenience, @value{GDBN} comes with a program,
20696 @command{gdb-add-index}, which can be used to add the index to a
20697 symbol file. It takes the symbol file as its only argument:
20700 $ gdb-add-index symfile
20703 @xref{gdb-add-index}.
20705 It is also possible to do the work manually. Here is what
20706 @command{gdb-add-index} does behind the curtains.
20708 The index is stored as a section in the symbol file. @value{GDBN} can
20709 write the index to a file, then you can put it into the symbol file
20710 using @command{objcopy}.
20712 To create an index file, use the @code{save gdb-index} command:
20715 @item save gdb-index [-dwarf-5] @var{directory}
20716 @kindex save gdb-index
20717 Create index files for all symbol files currently known by
20718 @value{GDBN}. For each known @var{symbol-file}, this command by
20719 default creates it produces a single file
20720 @file{@var{symbol-file}.gdb-index}. If you invoke this command with
20721 the @option{-dwarf-5} option, it produces 2 files:
20722 @file{@var{symbol-file}.debug_names} and
20723 @file{@var{symbol-file}.debug_str}. The files are created in the
20724 given @var{directory}.
20727 Once you have created an index file you can merge it into your symbol
20728 file, here named @file{symfile}, using @command{objcopy}:
20731 $ objcopy --add-section .gdb_index=symfile.gdb-index \
20732 --set-section-flags .gdb_index=readonly symfile symfile
20735 Or for @code{-dwarf-5}:
20738 $ objcopy --dump-section .debug_str=symfile.debug_str.new symfile
20739 $ cat symfile.debug_str >>symfile.debug_str.new
20740 $ objcopy --add-section .debug_names=symfile.gdb-index \
20741 --set-section-flags .debug_names=readonly \
20742 --update-section .debug_str=symfile.debug_str.new symfile symfile
20745 @value{GDBN} will normally ignore older versions of @file{.gdb_index}
20746 sections that have been deprecated. Usually they are deprecated because
20747 they are missing a new feature or have performance issues.
20748 To tell @value{GDBN} to use a deprecated index section anyway
20749 specify @code{set use-deprecated-index-sections on}.
20750 The default is @code{off}.
20751 This can speed up startup, but may result in some functionality being lost.
20752 @xref{Index Section Format}.
20754 @emph{Warning:} Setting @code{use-deprecated-index-sections} to @code{on}
20755 must be done before gdb reads the file. The following will not work:
20758 $ gdb -ex "set use-deprecated-index-sections on" <program>
20761 Instead you must do, for example,
20764 $ gdb -iex "set use-deprecated-index-sections on" <program>
20767 There are currently some limitation on indices. They only work when
20768 for DWARF debugging information, not stabs. And, they do not
20769 currently work for programs using Ada.
20771 @subsection Automatic symbol index cache
20773 @cindex automatic symbol index cache
20774 It is possible for @value{GDBN} to automatically save a copy of this index in a
20775 cache on disk and retrieve it from there when loading the same binary in the
20776 future. This feature can be turned on with @kbd{set index-cache on}. The
20777 following commands can be used to tweak the behavior of the index cache.
20781 @kindex set index-cache
20782 @item set index-cache on
20783 @itemx set index-cache off
20784 Enable or disable the use of the symbol index cache.
20786 @item set index-cache directory @var{directory}
20787 @kindex show index-cache
20788 @itemx show index-cache directory
20789 Set/show the directory where index files will be saved.
20791 The default value for this directory depends on the host platform. On
20792 most systems, the index is cached in the @file{gdb} subdirectory of
20793 the directory pointed to by the @env{XDG_CACHE_HOME} environment
20794 variable, if it is defined, else in the @file{.cache/gdb} subdirectory
20795 of your home directory. However, on some systems, the default may
20796 differ according to local convention.
20798 There is no limit on the disk space used by index cache. It is perfectly safe
20799 to delete the content of that directory to free up disk space.
20801 @item show index-cache stats
20802 Print the number of cache hits and misses since the launch of @value{GDBN}.
20806 @node Symbol Errors
20807 @section Errors Reading Symbol Files
20809 While reading a symbol file, @value{GDBN} occasionally encounters problems,
20810 such as symbol types it does not recognize, or known bugs in compiler
20811 output. By default, @value{GDBN} does not notify you of such problems, since
20812 they are relatively common and primarily of interest to people
20813 debugging compilers. If you are interested in seeing information
20814 about ill-constructed symbol tables, you can either ask @value{GDBN} to print
20815 only one message about each such type of problem, no matter how many
20816 times the problem occurs; or you can ask @value{GDBN} to print more messages,
20817 to see how many times the problems occur, with the @code{set
20818 complaints} command (@pxref{Messages/Warnings, ,Optional Warnings and
20821 The messages currently printed, and their meanings, include:
20824 @item inner block not inside outer block in @var{symbol}
20826 The symbol information shows where symbol scopes begin and end
20827 (such as at the start of a function or a block of statements). This
20828 error indicates that an inner scope block is not fully contained
20829 in its outer scope blocks.
20831 @value{GDBN} circumvents the problem by treating the inner block as if it had
20832 the same scope as the outer block. In the error message, @var{symbol}
20833 may be shown as ``@code{(don't know)}'' if the outer block is not a
20836 @item block at @var{address} out of order
20838 The symbol information for symbol scope blocks should occur in
20839 order of increasing addresses. This error indicates that it does not
20842 @value{GDBN} does not circumvent this problem, and has trouble
20843 locating symbols in the source file whose symbols it is reading. (You
20844 can often determine what source file is affected by specifying
20845 @code{set verbose on}. @xref{Messages/Warnings, ,Optional Warnings and
20848 @item bad block start address patched
20850 The symbol information for a symbol scope block has a start address
20851 smaller than the address of the preceding source line. This is known
20852 to occur in the SunOS 4.1.1 (and earlier) C compiler.
20854 @value{GDBN} circumvents the problem by treating the symbol scope block as
20855 starting on the previous source line.
20857 @item bad string table offset in symbol @var{n}
20860 Symbol number @var{n} contains a pointer into the string table which is
20861 larger than the size of the string table.
20863 @value{GDBN} circumvents the problem by considering the symbol to have the
20864 name @code{foo}, which may cause other problems if many symbols end up
20867 @item unknown symbol type @code{0x@var{nn}}
20869 The symbol information contains new data types that @value{GDBN} does
20870 not yet know how to read. @code{0x@var{nn}} is the symbol type of the
20871 uncomprehended information, in hexadecimal.
20873 @value{GDBN} circumvents the error by ignoring this symbol information.
20874 This usually allows you to debug your program, though certain symbols
20875 are not accessible. If you encounter such a problem and feel like
20876 debugging it, you can debug @code{@value{GDBP}} with itself, breakpoint
20877 on @code{complain}, then go up to the function @code{read_dbx_symtab}
20878 and examine @code{*bufp} to see the symbol.
20880 @item stub type has NULL name
20882 @value{GDBN} could not find the full definition for a struct or class.
20884 @item const/volatile indicator missing (ok if using g++ v1.x), got@dots{}
20885 The symbol information for a C@t{++} member function is missing some
20886 information that recent versions of the compiler should have output for
20889 @item info mismatch between compiler and debugger
20891 @value{GDBN} could not parse a type specification output by the compiler.
20896 @section GDB Data Files
20898 @cindex prefix for data files
20899 @value{GDBN} will sometimes read an auxiliary data file. These files
20900 are kept in a directory known as the @dfn{data directory}.
20902 You can set the data directory's name, and view the name @value{GDBN}
20903 is currently using.
20906 @kindex set data-directory
20907 @item set data-directory @var{directory}
20908 Set the directory which @value{GDBN} searches for auxiliary data files
20909 to @var{directory}.
20911 @kindex show data-directory
20912 @item show data-directory
20913 Show the directory @value{GDBN} searches for auxiliary data files.
20916 @cindex default data directory
20917 @cindex @samp{--with-gdb-datadir}
20918 You can set the default data directory by using the configure-time
20919 @samp{--with-gdb-datadir} option. If the data directory is inside
20920 @value{GDBN}'s configured binary prefix (set with @samp{--prefix} or
20921 @samp{--exec-prefix}), then the default data directory will be updated
20922 automatically if the installed @value{GDBN} is moved to a new
20925 The data directory may also be specified with the
20926 @code{--data-directory} command line option.
20927 @xref{Mode Options}.
20930 @chapter Specifying a Debugging Target
20932 @cindex debugging target
20933 A @dfn{target} is the execution environment occupied by your program.
20935 Often, @value{GDBN} runs in the same host environment as your program;
20936 in that case, the debugging target is specified as a side effect when
20937 you use the @code{file} or @code{core} commands. When you need more
20938 flexibility---for example, running @value{GDBN} on a physically separate
20939 host, or controlling a standalone system over a serial port or a
20940 realtime system over a TCP/IP connection---you can use the @code{target}
20941 command to specify one of the target types configured for @value{GDBN}
20942 (@pxref{Target Commands, ,Commands for Managing Targets}).
20944 @cindex target architecture
20945 It is possible to build @value{GDBN} for several different @dfn{target
20946 architectures}. When @value{GDBN} is built like that, you can choose
20947 one of the available architectures with the @kbd{set architecture}
20951 @kindex set architecture
20952 @kindex show architecture
20953 @item set architecture @var{arch}
20954 This command sets the current target architecture to @var{arch}. The
20955 value of @var{arch} can be @code{"auto"}, in addition to one of the
20956 supported architectures.
20958 @item show architecture
20959 Show the current target architecture.
20961 @item set processor
20963 @kindex set processor
20964 @kindex show processor
20965 These are alias commands for, respectively, @code{set architecture}
20966 and @code{show architecture}.
20970 * Active Targets:: Active targets
20971 * Target Commands:: Commands for managing targets
20972 * Byte Order:: Choosing target byte order
20975 @node Active Targets
20976 @section Active Targets
20978 @cindex stacking targets
20979 @cindex active targets
20980 @cindex multiple targets
20982 There are multiple classes of targets such as: processes, executable files or
20983 recording sessions. Core files belong to the process class, making core file
20984 and process mutually exclusive. Otherwise, @value{GDBN} can work concurrently
20985 on multiple active targets, one in each class. This allows you to (for
20986 example) start a process and inspect its activity, while still having access to
20987 the executable file after the process finishes. Or if you start process
20988 recording (@pxref{Reverse Execution}) and @code{reverse-step} there, you are
20989 presented a virtual layer of the recording target, while the process target
20990 remains stopped at the chronologically last point of the process execution.
20992 Use the @code{core-file} and @code{exec-file} commands to select a new core
20993 file or executable target (@pxref{Files, ,Commands to Specify Files}). To
20994 specify as a target a process that is already running, use the @code{attach}
20995 command (@pxref{Attach, ,Debugging an Already-running Process}).
20997 @node Target Commands
20998 @section Commands for Managing Targets
21001 @item target @var{type} @var{parameters}
21002 Connects the @value{GDBN} host environment to a target machine or
21003 process. A target is typically a protocol for talking to debugging
21004 facilities. You use the argument @var{type} to specify the type or
21005 protocol of the target machine.
21007 Further @var{parameters} are interpreted by the target protocol, but
21008 typically include things like device names or host names to connect
21009 with, process numbers, and baud rates.
21011 The @code{target} command does not repeat if you press @key{RET} again
21012 after executing the command.
21014 @kindex help target
21016 Displays the names of all targets available. To display targets
21017 currently selected, use either @code{info target} or @code{info files}
21018 (@pxref{Files, ,Commands to Specify Files}).
21020 @item help target @var{name}
21021 Describe a particular target, including any parameters necessary to
21024 @kindex set gnutarget
21025 @item set gnutarget @var{args}
21026 @value{GDBN} uses its own library BFD to read your files. @value{GDBN}
21027 knows whether it is reading an @dfn{executable},
21028 a @dfn{core}, or a @dfn{.o} file; however, you can specify the file format
21029 with the @code{set gnutarget} command. Unlike most @code{target} commands,
21030 with @code{gnutarget} the @code{target} refers to a program, not a machine.
21033 @emph{Warning:} To specify a file format with @code{set gnutarget},
21034 you must know the actual BFD name.
21038 @xref{Files, , Commands to Specify Files}.
21040 @kindex show gnutarget
21041 @item show gnutarget
21042 Use the @code{show gnutarget} command to display what file format
21043 @code{gnutarget} is set to read. If you have not set @code{gnutarget},
21044 @value{GDBN} will determine the file format for each file automatically,
21045 and @code{show gnutarget} displays @samp{The current BFD target is "auto"}.
21048 @cindex common targets
21049 Here are some common targets (available, or not, depending on the GDB
21054 @item target exec @var{program}
21055 @cindex executable file target
21056 An executable file. @samp{target exec @var{program}} is the same as
21057 @samp{exec-file @var{program}}.
21059 @item target core @var{filename}
21060 @cindex core dump file target
21061 A core dump file. @samp{target core @var{filename}} is the same as
21062 @samp{core-file @var{filename}}.
21064 @item target remote @var{medium}
21065 @cindex remote target
21066 A remote system connected to @value{GDBN} via a serial line or network
21067 connection. This command tells @value{GDBN} to use its own remote
21068 protocol over @var{medium} for debugging. @xref{Remote Debugging}.
21070 For example, if you have a board connected to @file{/dev/ttya} on the
21071 machine running @value{GDBN}, you could say:
21074 target remote /dev/ttya
21077 @code{target remote} supports the @code{load} command. This is only
21078 useful if you have some other way of getting the stub to the target
21079 system, and you can put it somewhere in memory where it won't get
21080 clobbered by the download.
21082 @item target sim @r{[}@var{simargs}@r{]} @dots{}
21083 @cindex built-in simulator target
21084 Builtin CPU simulator. @value{GDBN} includes simulators for most architectures.
21092 works; however, you cannot assume that a specific memory map, device
21093 drivers, or even basic I/O is available, although some simulators do
21094 provide these. For info about any processor-specific simulator details,
21095 see the appropriate section in @ref{Embedded Processors, ,Embedded
21098 @item target native
21099 @cindex native target
21100 Setup for local/native process debugging. Useful to make the
21101 @code{run} command spawn native processes (likewise @code{attach},
21102 etc.@:) even when @code{set auto-connect-native-target} is @code{off}
21103 (@pxref{set auto-connect-native-target}).
21107 Different targets are available on different configurations of @value{GDBN};
21108 your configuration may have more or fewer targets.
21110 Many remote targets require you to download the executable's code once
21111 you've successfully established a connection. You may wish to control
21112 various aspects of this process.
21117 @kindex set hash@r{, for remote monitors}
21118 @cindex hash mark while downloading
21119 This command controls whether a hash mark @samp{#} is displayed while
21120 downloading a file to the remote monitor. If on, a hash mark is
21121 displayed after each S-record is successfully downloaded to the
21125 @kindex show hash@r{, for remote monitors}
21126 Show the current status of displaying the hash mark.
21128 @item set debug monitor
21129 @kindex set debug monitor
21130 @cindex display remote monitor communications
21131 Enable or disable display of communications messages between
21132 @value{GDBN} and the remote monitor.
21134 @item show debug monitor
21135 @kindex show debug monitor
21136 Show the current status of displaying communications between
21137 @value{GDBN} and the remote monitor.
21142 @kindex load @var{filename} @var{offset}
21143 @item load @var{filename} @var{offset}
21145 Depending on what remote debugging facilities are configured into
21146 @value{GDBN}, the @code{load} command may be available. Where it exists, it
21147 is meant to make @var{filename} (an executable) available for debugging
21148 on the remote system---by downloading, or dynamic linking, for example.
21149 @code{load} also records the @var{filename} symbol table in @value{GDBN}, like
21150 the @code{add-symbol-file} command.
21152 If your @value{GDBN} does not have a @code{load} command, attempting to
21153 execute it gets the error message ``@code{You can't do that when your
21154 target is @dots{}}''
21156 The file is loaded at whatever address is specified in the executable.
21157 For some object file formats, you can specify the load address when you
21158 link the program; for other formats, like a.out, the object file format
21159 specifies a fixed address.
21160 @c FIXME! This would be a good place for an xref to the GNU linker doc.
21162 It is also possible to tell @value{GDBN} to load the executable file at a
21163 specific offset described by the optional argument @var{offset}. When
21164 @var{offset} is provided, @var{filename} must also be provided.
21166 Depending on the remote side capabilities, @value{GDBN} may be able to
21167 load programs into flash memory.
21169 @code{load} does not repeat if you press @key{RET} again after using it.
21174 @kindex flash-erase
21176 @anchor{flash-erase}
21178 Erases all known flash memory regions on the target.
21183 @section Choosing Target Byte Order
21185 @cindex choosing target byte order
21186 @cindex target byte order
21188 Some types of processors, such as the @acronym{MIPS}, PowerPC, and Renesas SH,
21189 offer the ability to run either big-endian or little-endian byte
21190 orders. Usually the executable or symbol will include a bit to
21191 designate the endian-ness, and you will not need to worry about
21192 which to use. However, you may still find it useful to adjust
21193 @value{GDBN}'s idea of processor endian-ness manually.
21197 @item set endian big
21198 Instruct @value{GDBN} to assume the target is big-endian.
21200 @item set endian little
21201 Instruct @value{GDBN} to assume the target is little-endian.
21203 @item set endian auto
21204 Instruct @value{GDBN} to use the byte order associated with the
21208 Display @value{GDBN}'s current idea of the target byte order.
21212 If the @code{set endian auto} mode is in effect and no executable has
21213 been selected, then the endianness used is the last one chosen either
21214 by one of the @code{set endian big} and @code{set endian little}
21215 commands or by inferring from the last executable used. If no
21216 endianness has been previously chosen, then the default for this mode
21217 is inferred from the target @value{GDBN} has been built for, and is
21218 @code{little} if the name of the target CPU has an @code{el} suffix
21219 and @code{big} otherwise.
21221 Note that these commands merely adjust interpretation of symbolic
21222 data on the host, and that they have absolutely no effect on the
21226 @node Remote Debugging
21227 @chapter Debugging Remote Programs
21228 @cindex remote debugging
21230 If you are trying to debug a program running on a machine that cannot run
21231 @value{GDBN} in the usual way, it is often useful to use remote debugging.
21232 For example, you might use remote debugging on an operating system kernel,
21233 or on a small system which does not have a general purpose operating system
21234 powerful enough to run a full-featured debugger.
21236 Some configurations of @value{GDBN} have special serial or TCP/IP interfaces
21237 to make this work with particular debugging targets. In addition,
21238 @value{GDBN} comes with a generic serial protocol (specific to @value{GDBN},
21239 but not specific to any particular target system) which you can use if you
21240 write the remote stubs---the code that runs on the remote system to
21241 communicate with @value{GDBN}.
21243 Other remote targets may be available in your
21244 configuration of @value{GDBN}; use @code{help target} to list them.
21247 * Connecting:: Connecting to a remote target
21248 * File Transfer:: Sending files to a remote system
21249 * Server:: Using the gdbserver program
21250 * Remote Configuration:: Remote configuration
21251 * Remote Stub:: Implementing a remote stub
21255 @section Connecting to a Remote Target
21256 @cindex remote debugging, connecting
21257 @cindex @code{gdbserver}, connecting
21258 @cindex remote debugging, types of connections
21259 @cindex @code{gdbserver}, types of connections
21260 @cindex @code{gdbserver}, @code{target remote} mode
21261 @cindex @code{gdbserver}, @code{target extended-remote} mode
21263 This section describes how to connect to a remote target, including the
21264 types of connections and their differences, how to set up executable and
21265 symbol files on the host and target, and the commands used for
21266 connecting to and disconnecting from the remote target.
21268 @subsection Types of Remote Connections
21270 @value{GDBN} supports two types of remote connections, @code{target remote}
21271 mode and @code{target extended-remote} mode. Note that many remote targets
21272 support only @code{target remote} mode. There are several major
21273 differences between the two types of connections, enumerated here:
21277 @cindex remote debugging, detach and program exit
21278 @item Result of detach or program exit
21279 @strong{With target remote mode:} When the debugged program exits or you
21280 detach from it, @value{GDBN} disconnects from the target. When using
21281 @code{gdbserver}, @code{gdbserver} will exit.
21283 @strong{With target extended-remote mode:} When the debugged program exits or
21284 you detach from it, @value{GDBN} remains connected to the target, even
21285 though no program is running. You can rerun the program, attach to a
21286 running program, or use @code{monitor} commands specific to the target.
21288 When using @code{gdbserver} in this case, it does not exit unless it was
21289 invoked using the @option{--once} option. If the @option{--once} option
21290 was not used, you can ask @code{gdbserver} to exit using the
21291 @code{monitor exit} command (@pxref{Monitor Commands for gdbserver}).
21293 @item Specifying the program to debug
21294 For both connection types you use the @code{file} command to specify the
21295 program on the host system. If you are using @code{gdbserver} there are
21296 some differences in how to specify the location of the program on the
21299 @strong{With target remote mode:} You must either specify the program to debug
21300 on the @code{gdbserver} command line or use the @option{--attach} option
21301 (@pxref{Attaching to a program,,Attaching to a Running Program}).
21303 @cindex @option{--multi}, @code{gdbserver} option
21304 @strong{With target extended-remote mode:} You may specify the program to debug
21305 on the @code{gdbserver} command line, or you can load the program or attach
21306 to it using @value{GDBN} commands after connecting to @code{gdbserver}.
21308 @anchor{--multi Option in Types of Remote Connnections}
21309 You can start @code{gdbserver} without supplying an initial command to run
21310 or process ID to attach. To do this, use the @option{--multi} command line
21311 option. Then you can connect using @code{target extended-remote} and start
21312 the program you want to debug (see below for details on using the
21313 @code{run} command in this scenario). Note that the conditions under which
21314 @code{gdbserver} terminates depend on how @value{GDBN} connects to it
21315 (@code{target remote} or @code{target extended-remote}). The
21316 @option{--multi} option to @code{gdbserver} has no influence on that.
21318 @item The @code{run} command
21319 @strong{With target remote mode:} The @code{run} command is not
21320 supported. Once a connection has been established, you can use all
21321 the usual @value{GDBN} commands to examine and change data. The
21322 remote program is already running, so you can use commands like
21323 @kbd{step} and @kbd{continue}.
21325 @strong{With target extended-remote mode:} The @code{run} command is
21326 supported. The @code{run} command uses the value set by
21327 @code{set remote exec-file} (@pxref{set remote exec-file}) to select
21328 the program to run. Command line arguments are supported, except for
21329 wildcard expansion and I/O redirection (@pxref{Arguments}).
21331 If you specify the program to debug on the command line, then the
21332 @code{run} command is not required to start execution, and you can
21333 resume using commands like @kbd{step} and @kbd{continue} as with
21334 @code{target remote} mode.
21336 @anchor{Attaching in Types of Remote Connections}
21338 @strong{With target remote mode:} The @value{GDBN} command @code{attach} is
21339 not supported. To attach to a running program using @code{gdbserver}, you
21340 must use the @option{--attach} option (@pxref{Running gdbserver}).
21342 @strong{With target extended-remote mode:} To attach to a running program,
21343 you may use the @code{attach} command after the connection has been
21344 established. If you are using @code{gdbserver}, you may also invoke
21345 @code{gdbserver} using the @option{--attach} option
21346 (@pxref{Running gdbserver}).
21350 @anchor{Host and target files}
21351 @subsection Host and Target Files
21352 @cindex remote debugging, symbol files
21353 @cindex symbol files, remote debugging
21355 @value{GDBN}, running on the host, needs access to symbol and debugging
21356 information for your program running on the target. This requires
21357 access to an unstripped copy of your program, and possibly any associated
21358 symbol files. Note that this section applies equally to both @code{target
21359 remote} mode and @code{target extended-remote} mode.
21361 Some remote targets (@pxref{qXfer executable filename read}, and
21362 @pxref{Host I/O Packets}) allow @value{GDBN} to access program files over
21363 the same connection used to communicate with @value{GDBN}. With such a
21364 target, if the remote program is unstripped, the only command you need is
21365 @code{target remote} (or @code{target extended-remote}).
21367 If the remote program is stripped, or the target does not support remote
21368 program file access, start up @value{GDBN} using the name of the local
21369 unstripped copy of your program as the first argument, or use the
21370 @code{file} command. Use @code{set sysroot} to specify the location (on
21371 the host) of target libraries (unless your @value{GDBN} was compiled with
21372 the correct sysroot using @code{--with-sysroot}). Alternatively, you
21373 may use @code{set solib-search-path} to specify how @value{GDBN} locates
21376 The symbol file and target libraries must exactly match the executable
21377 and libraries on the target, with one exception: the files on the host
21378 system should not be stripped, even if the files on the target system
21379 are. Mismatched or missing files will lead to confusing results
21380 during debugging. On @sc{gnu}/Linux targets, mismatched or missing
21381 files may also prevent @code{gdbserver} from debugging multi-threaded
21384 @subsection Remote Connection Commands
21385 @cindex remote connection commands
21386 @value{GDBN} can communicate with the target over a serial line, a
21387 local Unix domain socket, or
21388 over an @acronym{IP} network using @acronym{TCP} or @acronym{UDP}. In
21389 each case, @value{GDBN} uses the same protocol for debugging your
21390 program; only the medium carrying the debugging packets varies. The
21391 @code{target remote} and @code{target extended-remote} commands
21392 establish a connection to the target. Both commands accept the same
21393 arguments, which indicate the medium to use:
21397 @item target remote @var{serial-device}
21398 @itemx target extended-remote @var{serial-device}
21399 @cindex serial line, @code{target remote}
21400 Use @var{serial-device} to communicate with the target. For example,
21401 to use a serial line connected to the device named @file{/dev/ttyb}:
21404 target remote /dev/ttyb
21407 If you're using a serial line, you may want to give @value{GDBN} the
21408 @samp{--baud} option, or use the @code{set serial baud} command
21409 (@pxref{Remote Configuration, set serial baud}) before the
21410 @code{target} command.
21412 @item target remote @var{local-socket}
21413 @itemx target extended-remote @var{local-socket}
21414 @cindex local socket, @code{target remote}
21415 @cindex Unix domain socket
21416 Use @var{local-socket} to communicate with the target. For example,
21417 to use a local Unix domain socket bound to the file system entry @file{/tmp/gdb-socket0}:
21420 target remote /tmp/gdb-socket0
21423 Note that this command has the same form as the command to connect
21424 to a serial line. @value{GDBN} will automatically determine which
21425 kind of file you have specified and will make the appropriate kind
21427 This feature is not available if the host system does not support
21428 Unix domain sockets.
21430 @item target remote @code{@var{host}:@var{port}}
21431 @itemx target remote @code{@var{[host]}:@var{port}}
21432 @itemx target remote @code{tcp:@var{host}:@var{port}}
21433 @itemx target remote @code{tcp:@var{[host]}:@var{port}}
21434 @itemx target remote @code{tcp4:@var{host}:@var{port}}
21435 @itemx target remote @code{tcp6:@var{host}:@var{port}}
21436 @itemx target remote @code{tcp6:@var{[host]}:@var{port}}
21437 @itemx target extended-remote @code{@var{host}:@var{port}}
21438 @itemx target extended-remote @code{@var{[host]}:@var{port}}
21439 @itemx target extended-remote @code{tcp:@var{host}:@var{port}}
21440 @itemx target extended-remote @code{tcp:@var{[host]}:@var{port}}
21441 @itemx target extended-remote @code{tcp4:@var{host}:@var{port}}
21442 @itemx target extended-remote @code{tcp6:@var{host}:@var{port}}
21443 @itemx target extended-remote @code{tcp6:@var{[host]}:@var{port}}
21444 @cindex @acronym{TCP} port, @code{target remote}
21445 Debug using a @acronym{TCP} connection to @var{port} on @var{host}.
21446 The @var{host} may be either a host name, a numeric @acronym{IPv4}
21447 address, or a numeric @acronym{IPv6} address (with or without the
21448 square brackets to separate the address from the port); @var{port}
21449 must be a decimal number. The @var{host} could be the target machine
21450 itself, if it is directly connected to the net, or it might be a
21451 terminal server which in turn has a serial line to the target.
21453 For example, to connect to port 2828 on a terminal server named
21457 target remote manyfarms:2828
21460 To connect to port 2828 on a terminal server whose address is
21461 @code{2001:0db8:85a3:0000:0000:8a2e:0370:7334}, you can either use the
21462 square bracket syntax:
21465 target remote [2001:0db8:85a3:0000:0000:8a2e:0370:7334]:2828
21469 or explicitly specify the @acronym{IPv6} protocol:
21472 target remote tcp6:2001:0db8:85a3:0000:0000:8a2e:0370:7334:2828
21475 This last example may be confusing to the reader, because there is no
21476 visible separation between the hostname and the port number.
21477 Therefore, we recommend the user to provide @acronym{IPv6} addresses
21478 using square brackets for clarity. However, it is important to
21479 mention that for @value{GDBN} there is no ambiguity: the number after
21480 the last colon is considered to be the port number.
21482 If your remote target is actually running on the same machine as your
21483 debugger session (e.g.@: a simulator for your target running on the
21484 same host), you can omit the hostname. For example, to connect to
21485 port 1234 on your local machine:
21488 target remote :1234
21492 Note that the colon is still required here.
21494 @item target remote @code{udp:@var{host}:@var{port}}
21495 @itemx target remote @code{udp:@var{[host]}:@var{port}}
21496 @itemx target remote @code{udp4:@var{host}:@var{port}}
21497 @itemx target remote @code{udp6:@var{[host]}:@var{port}}
21498 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
21499 @itemx target extended-remote @code{udp:@var{host}:@var{port}}
21500 @itemx target extended-remote @code{udp:@var{[host]}:@var{port}}
21501 @itemx target extended-remote @code{udp4:@var{host}:@var{port}}
21502 @itemx target extended-remote @code{udp6:@var{host}:@var{port}}
21503 @itemx target extended-remote @code{udp6:@var{[host]}:@var{port}}
21504 @cindex @acronym{UDP} port, @code{target remote}
21505 Debug using @acronym{UDP} packets to @var{port} on @var{host}. For example, to
21506 connect to @acronym{UDP} port 2828 on a terminal server named @code{manyfarms}:
21509 target remote udp:manyfarms:2828
21512 When using a @acronym{UDP} connection for remote debugging, you should
21513 keep in mind that the `U' stands for ``Unreliable''. @acronym{UDP}
21514 can silently drop packets on busy or unreliable networks, which will
21515 cause havoc with your debugging session.
21517 @item target remote | @var{command}
21518 @itemx target extended-remote | @var{command}
21519 @cindex pipe, @code{target remote} to
21520 Run @var{command} in the background and communicate with it using a
21521 pipe. The @var{command} is a shell command, to be parsed and expanded
21522 by the system's command shell, @code{/bin/sh}; it should expect remote
21523 protocol packets on its standard input, and send replies on its
21524 standard output. You could use this to run a stand-alone simulator
21525 that speaks the remote debugging protocol, to make net connections
21526 using programs like @code{ssh}, or for other similar tricks.
21528 If @var{command} closes its standard output (perhaps by exiting),
21529 @value{GDBN} will try to send it a @code{SIGTERM} signal. (If the
21530 program has already exited, this will have no effect.)
21534 @cindex interrupting remote programs
21535 @cindex remote programs, interrupting
21536 Whenever @value{GDBN} is waiting for the remote program, if you type the
21537 interrupt character (often @kbd{Ctrl-c}), @value{GDBN} attempts to stop the
21538 program. This may or may not succeed, depending in part on the hardware
21539 and the serial drivers the remote system uses. If you type the
21540 interrupt character once again, @value{GDBN} displays this prompt:
21543 Interrupted while waiting for the program.
21544 Give up (and stop debugging it)? (y or n)
21547 In @code{target remote} mode, if you type @kbd{y}, @value{GDBN} abandons
21548 the remote debugging session. (If you decide you want to try again later,
21549 you can use @kbd{target remote} again to connect once more.) If you type
21550 @kbd{n}, @value{GDBN} goes back to waiting.
21552 In @code{target extended-remote} mode, typing @kbd{n} will leave
21553 @value{GDBN} connected to the target.
21556 @kindex detach (remote)
21558 When you have finished debugging the remote program, you can use the
21559 @code{detach} command to release it from @value{GDBN} control.
21560 Detaching from the target normally resumes its execution, but the results
21561 will depend on your particular remote stub. After the @code{detach}
21562 command in @code{target remote} mode, @value{GDBN} is free to connect to
21563 another target. In @code{target extended-remote} mode, @value{GDBN} is
21564 still connected to the target.
21568 The @code{disconnect} command closes the connection to the target, and
21569 the target is generally not resumed. It will wait for @value{GDBN}
21570 (this instance or another one) to connect and continue debugging. After
21571 the @code{disconnect} command, @value{GDBN} is again free to connect to
21574 @cindex send command to remote monitor
21575 @cindex extend @value{GDBN} for remote targets
21576 @cindex add new commands for external monitor
21578 @item monitor @var{cmd}
21579 This command allows you to send arbitrary commands directly to the
21580 remote monitor. Since @value{GDBN} doesn't care about the commands it
21581 sends like this, this command is the way to extend @value{GDBN}---you
21582 can add new commands that only the external monitor will understand
21586 @node File Transfer
21587 @section Sending files to a remote system
21588 @cindex remote target, file transfer
21589 @cindex file transfer
21590 @cindex sending files to remote systems
21592 Some remote targets offer the ability to transfer files over the same
21593 connection used to communicate with @value{GDBN}. This is convenient
21594 for targets accessible through other means, e.g.@: @sc{gnu}/Linux systems
21595 running @code{gdbserver} over a network interface. For other targets,
21596 e.g.@: embedded devices with only a single serial port, this may be
21597 the only way to upload or download files.
21599 Not all remote targets support these commands.
21603 @item remote put @var{hostfile} @var{targetfile}
21604 Copy file @var{hostfile} from the host system (the machine running
21605 @value{GDBN}) to @var{targetfile} on the target system.
21608 @item remote get @var{targetfile} @var{hostfile}
21609 Copy file @var{targetfile} from the target system to @var{hostfile}
21610 on the host system.
21612 @kindex remote delete
21613 @item remote delete @var{targetfile}
21614 Delete @var{targetfile} from the target system.
21619 @section Using the @code{gdbserver} Program
21622 @cindex remote connection without stubs
21623 @code{gdbserver} is a control program for Unix-like systems, which
21624 allows you to connect your program with a remote @value{GDBN} via
21625 @code{target remote} or @code{target extended-remote}---but without
21626 linking in the usual debugging stub.
21628 @code{gdbserver} is not a complete replacement for the debugging stubs,
21629 because it requires essentially the same operating-system facilities
21630 that @value{GDBN} itself does. In fact, a system that can run
21631 @code{gdbserver} to connect to a remote @value{GDBN} could also run
21632 @value{GDBN} locally! @code{gdbserver} is sometimes useful nevertheless,
21633 because it is a much smaller program than @value{GDBN} itself. It is
21634 also easier to port than all of @value{GDBN}, so you may be able to get
21635 started more quickly on a new system by using @code{gdbserver}.
21636 Finally, if you develop code for real-time systems, you may find that
21637 the tradeoffs involved in real-time operation make it more convenient to
21638 do as much development work as possible on another system, for example
21639 by cross-compiling. You can use @code{gdbserver} to make a similar
21640 choice for debugging.
21642 @value{GDBN} and @code{gdbserver} communicate via either a serial line
21643 or a TCP connection, using the standard @value{GDBN} remote serial
21647 @emph{Warning:} @code{gdbserver} does not have any built-in security.
21648 Do not run @code{gdbserver} connected to any public network; a
21649 @value{GDBN} connection to @code{gdbserver} provides access to the
21650 target system with the same privileges as the user running
21654 @anchor{Running gdbserver}
21655 @subsection Running @code{gdbserver}
21656 @cindex arguments, to @code{gdbserver}
21657 @cindex @code{gdbserver}, command-line arguments
21659 Run @code{gdbserver} on the target system. You need a copy of the
21660 program you want to debug, including any libraries it requires.
21661 @code{gdbserver} does not need your program's symbol table, so you can
21662 strip the program if necessary to save space. @value{GDBN} on the host
21663 system does all the symbol handling.
21665 To use the server, you must tell it how to communicate with @value{GDBN};
21666 the name of your program; and the arguments for your program. The usual
21670 target> gdbserver @var{comm} @var{program} [ @var{args} @dots{} ]
21673 @var{comm} is either a device name (to use a serial line), or a TCP
21674 hostname and portnumber, or @code{-} or @code{stdio} to use
21675 stdin/stdout of @code{gdbserver}.
21676 For example, to debug Emacs with the argument
21677 @samp{foo.txt} and communicate with @value{GDBN} over the serial port
21681 target> gdbserver /dev/com1 emacs foo.txt
21684 @code{gdbserver} waits passively for the host @value{GDBN} to communicate
21687 To use a TCP connection instead of a serial line:
21690 target> gdbserver host:2345 emacs foo.txt
21693 The only difference from the previous example is the first argument,
21694 specifying that you are communicating with the host @value{GDBN} via
21695 TCP. The @samp{host:2345} argument means that @code{gdbserver} is to
21696 expect a TCP connection from machine @samp{host} to local TCP port 2345.
21697 (Currently, the @samp{host} part is ignored.) You can choose any number
21698 you want for the port number as long as it does not conflict with any
21699 TCP ports already in use on the target system (for example, @code{23} is
21700 reserved for @code{telnet}).@footnote{If you choose a port number that
21701 conflicts with another service, @code{gdbserver} prints an error message
21702 and exits.} You must use the same port number with the host @value{GDBN}
21703 @code{target remote} command.
21705 The @code{stdio} connection is useful when starting @code{gdbserver}
21709 (gdb) target remote | ssh -T hostname gdbserver - hello
21712 The @samp{-T} option to ssh is provided because we don't need a remote pty,
21713 and we don't want escape-character handling. Ssh does this by default when
21714 a command is provided, the flag is provided to make it explicit.
21715 You could elide it if you want to.
21717 Programs started with stdio-connected gdbserver have @file{/dev/null} for
21718 @code{stdin}, and @code{stdout},@code{stderr} are sent back to gdb for
21719 display through a pipe connected to gdbserver.
21720 Both @code{stdout} and @code{stderr} use the same pipe.
21722 @anchor{Attaching to a program}
21723 @subsubsection Attaching to a Running Program
21724 @cindex attach to a program, @code{gdbserver}
21725 @cindex @option{--attach}, @code{gdbserver} option
21727 On some targets, @code{gdbserver} can also attach to running programs.
21728 This is accomplished via the @code{--attach} argument. The syntax is:
21731 target> gdbserver --attach @var{comm} @var{pid}
21734 @var{pid} is the process ID of a currently running process. It isn't
21735 necessary to point @code{gdbserver} at a binary for the running process.
21737 In @code{target extended-remote} mode, you can also attach using the
21738 @value{GDBN} attach command
21739 (@pxref{Attaching in Types of Remote Connections}).
21742 You can debug processes by name instead of process ID if your target has the
21743 @code{pidof} utility:
21746 target> gdbserver --attach @var{comm} `pidof @var{program}`
21749 In case more than one copy of @var{program} is running, or @var{program}
21750 has multiple threads, most versions of @code{pidof} support the
21751 @code{-s} option to only return the first process ID.
21753 @subsubsection TCP port allocation lifecycle of @code{gdbserver}
21755 This section applies only when @code{gdbserver} is run to listen on a TCP
21758 @code{gdbserver} normally terminates after all of its debugged processes have
21759 terminated in @kbd{target remote} mode. On the other hand, for @kbd{target
21760 extended-remote}, @code{gdbserver} stays running even with no processes left.
21761 @value{GDBN} normally terminates the spawned debugged process on its exit,
21762 which normally also terminates @code{gdbserver} in the @kbd{target remote}
21763 mode. Therefore, when the connection drops unexpectedly, and @value{GDBN}
21764 cannot ask @code{gdbserver} to kill its debugged processes, @code{gdbserver}
21765 stays running even in the @kbd{target remote} mode.
21767 When @code{gdbserver} stays running, @value{GDBN} can connect to it again later.
21768 Such reconnecting is useful for features like @ref{disconnected tracing}. For
21769 completeness, at most one @value{GDBN} can be connected at a time.
21771 @cindex @option{--once}, @code{gdbserver} option
21772 By default, @code{gdbserver} keeps the listening TCP port open, so that
21773 subsequent connections are possible. However, if you start @code{gdbserver}
21774 with the @option{--once} option, it will stop listening for any further
21775 connection attempts after connecting to the first @value{GDBN} session. This
21776 means no further connections to @code{gdbserver} will be possible after the
21777 first one. It also means @code{gdbserver} will terminate after the first
21778 connection with remote @value{GDBN} has closed, even for unexpectedly closed
21779 connections and even in the @kbd{target extended-remote} mode. The
21780 @option{--once} option allows reusing the same port number for connecting to
21781 multiple instances of @code{gdbserver} running on the same host, since each
21782 instance closes its port after the first connection.
21784 @anchor{Other Command-Line Arguments for gdbserver}
21785 @subsubsection Other Command-Line Arguments for @code{gdbserver}
21787 You can use the @option{--multi} option to start @code{gdbserver} without
21788 specifying a program to debug or a process to attach to. Then you can
21789 attach in @code{target extended-remote} mode and run or attach to a
21790 program. For more information,
21791 @pxref{--multi Option in Types of Remote Connnections}.
21793 @cindex @option{--debug}, @code{gdbserver} option
21794 The @option{--debug} option tells @code{gdbserver} to display extra
21795 status information about the debugging process.
21796 @cindex @option{--remote-debug}, @code{gdbserver} option
21797 The @option{--remote-debug} option tells @code{gdbserver} to display
21798 remote protocol debug output.
21799 @cindex @option{--debug-file}, @code{gdbserver} option
21800 @cindex @code{gdbserver}, send all debug output to a single file
21801 The @option{--debug-file=@var{filename}} option tells @code{gdbserver} to
21802 write any debug output to the given @var{filename}. These options are intended
21803 for @code{gdbserver} development and for bug reports to the developers.
21805 @cindex @option{--debug-format}, @code{gdbserver} option
21806 The @option{--debug-format=option1[,option2,...]} option tells
21807 @code{gdbserver} to include additional information in each output.
21808 Possible options are:
21812 Turn off all extra information in debugging output.
21814 Turn on all extra information in debugging output.
21816 Include a timestamp in each line of debugging output.
21819 Options are processed in order. Thus, for example, if @option{none}
21820 appears last then no additional information is added to debugging output.
21822 @cindex @option{--wrapper}, @code{gdbserver} option
21823 The @option{--wrapper} option specifies a wrapper to launch programs
21824 for debugging. The option should be followed by the name of the
21825 wrapper, then any command-line arguments to pass to the wrapper, then
21826 @kbd{--} indicating the end of the wrapper arguments.
21828 @code{gdbserver} runs the specified wrapper program with a combined
21829 command line including the wrapper arguments, then the name of the
21830 program to debug, then any arguments to the program. The wrapper
21831 runs until it executes your program, and then @value{GDBN} gains control.
21833 You can use any program that eventually calls @code{execve} with
21834 its arguments as a wrapper. Several standard Unix utilities do
21835 this, e.g.@: @code{env} and @code{nohup}. Any Unix shell script ending
21836 with @code{exec "$@@"} will also work.
21838 For example, you can use @code{env} to pass an environment variable to
21839 the debugged program, without setting the variable in @code{gdbserver}'s
21843 $ gdbserver --wrapper env LD_PRELOAD=libtest.so -- :2222 ./testprog
21846 @cindex @option{--selftest}
21847 The @option{--selftest} option runs the self tests in @code{gdbserver}:
21850 $ gdbserver --selftest
21851 Ran 2 unit tests, 0 failed
21854 These tests are disabled in release.
21855 @subsection Connecting to @code{gdbserver}
21857 The basic procedure for connecting to the remote target is:
21861 Run @value{GDBN} on the host system.
21864 Make sure you have the necessary symbol files
21865 (@pxref{Host and target files}).
21866 Load symbols for your application using the @code{file} command before you
21867 connect. Use @code{set sysroot} to locate target libraries (unless your
21868 @value{GDBN} was compiled with the correct sysroot using
21869 @code{--with-sysroot}).
21872 Connect to your target (@pxref{Connecting,,Connecting to a Remote Target}).
21873 For TCP connections, you must start up @code{gdbserver} prior to using
21874 the @code{target} command. Otherwise you may get an error whose
21875 text depends on the host system, but which usually looks something like
21876 @samp{Connection refused}. Don't use the @code{load}
21877 command in @value{GDBN} when using @code{target remote} mode, since the
21878 program is already on the target.
21882 @anchor{Monitor Commands for gdbserver}
21883 @subsection Monitor Commands for @code{gdbserver}
21884 @cindex monitor commands, for @code{gdbserver}
21886 During a @value{GDBN} session using @code{gdbserver}, you can use the
21887 @code{monitor} command to send special requests to @code{gdbserver}.
21888 Here are the available commands.
21892 List the available monitor commands.
21894 @item monitor set debug 0
21895 @itemx monitor set debug 1
21896 Disable or enable general debugging messages.
21898 @item monitor set remote-debug 0
21899 @itemx monitor set remote-debug 1
21900 Disable or enable specific debugging messages associated with the remote
21901 protocol (@pxref{Remote Protocol}).
21903 @item monitor set debug-file filename
21904 @itemx monitor set debug-file
21905 Send any debug output to the given file, or to stderr.
21907 @item monitor set debug-format option1@r{[},option2,...@r{]}
21908 Specify additional text to add to debugging messages.
21909 Possible options are:
21913 Turn off all extra information in debugging output.
21915 Turn on all extra information in debugging output.
21917 Include a timestamp in each line of debugging output.
21920 Options are processed in order. Thus, for example, if @option{none}
21921 appears last then no additional information is added to debugging output.
21923 @item monitor set libthread-db-search-path [PATH]
21924 @cindex gdbserver, search path for @code{libthread_db}
21925 When this command is issued, @var{path} is a colon-separated list of
21926 directories to search for @code{libthread_db} (@pxref{Threads,,set
21927 libthread-db-search-path}). If you omit @var{path},
21928 @samp{libthread-db-search-path} will be reset to its default value.
21930 The special entry @samp{$pdir} for @samp{libthread-db-search-path} is
21931 not supported in @code{gdbserver}.
21934 Tell gdbserver to exit immediately. This command should be followed by
21935 @code{disconnect} to close the debugging session. @code{gdbserver} will
21936 detach from any attached processes and kill any processes it created.
21937 Use @code{monitor exit} to terminate @code{gdbserver} at the end
21938 of a multi-process mode debug session.
21942 @subsection Tracepoints support in @code{gdbserver}
21943 @cindex tracepoints support in @code{gdbserver}
21945 On some targets, @code{gdbserver} supports tracepoints, fast
21946 tracepoints and static tracepoints.
21948 For fast or static tracepoints to work, a special library called the
21949 @dfn{in-process agent} (IPA), must be loaded in the inferior process.
21950 This library is built and distributed as an integral part of
21951 @code{gdbserver}. In addition, support for static tracepoints
21952 requires building the in-process agent library with static tracepoints
21953 support. At present, the UST (LTTng Userspace Tracer,
21954 @url{http://lttng.org/ust}) tracing engine is supported. This support
21955 is automatically available if UST development headers are found in the
21956 standard include path when @code{gdbserver} is built, or if
21957 @code{gdbserver} was explicitly configured using @option{--with-ust}
21958 to point at such headers. You can explicitly disable the support
21959 using @option{--with-ust=no}.
21961 There are several ways to load the in-process agent in your program:
21964 @item Specifying it as dependency at link time
21966 You can link your program dynamically with the in-process agent
21967 library. On most systems, this is accomplished by adding
21968 @code{-linproctrace} to the link command.
21970 @item Using the system's preloading mechanisms
21972 You can force loading the in-process agent at startup time by using
21973 your system's support for preloading shared libraries. Many Unixes
21974 support the concept of preloading user defined libraries. In most
21975 cases, you do that by specifying @code{LD_PRELOAD=libinproctrace.so}
21976 in the environment. See also the description of @code{gdbserver}'s
21977 @option{--wrapper} command line option.
21979 @item Using @value{GDBN} to force loading the agent at run time
21981 On some systems, you can force the inferior to load a shared library,
21982 by calling a dynamic loader function in the inferior that takes care
21983 of dynamically looking up and loading a shared library. On most Unix
21984 systems, the function is @code{dlopen}. You'll use the @code{call}
21985 command for that. For example:
21988 (@value{GDBP}) call dlopen ("libinproctrace.so", ...)
21991 Note that on most Unix systems, for the @code{dlopen} function to be
21992 available, the program needs to be linked with @code{-ldl}.
21995 On systems that have a userspace dynamic loader, like most Unix
21996 systems, when you connect to @code{gdbserver} using @code{target
21997 remote}, you'll find that the program is stopped at the dynamic
21998 loader's entry point, and no shared library has been loaded in the
21999 program's address space yet, including the in-process agent. In that
22000 case, before being able to use any of the fast or static tracepoints
22001 features, you need to let the loader run and load the shared
22002 libraries. The simplest way to do that is to run the program to the
22003 main procedure. E.g., if debugging a C or C@t{++} program, start
22004 @code{gdbserver} like so:
22007 $ gdbserver :9999 myprogram
22010 Start GDB and connect to @code{gdbserver} like so, and run to main:
22014 (@value{GDBP}) target remote myhost:9999
22015 0x00007f215893ba60 in ?? () from /lib64/ld-linux-x86-64.so.2
22016 (@value{GDBP}) b main
22017 (@value{GDBP}) continue
22020 The in-process tracing agent library should now be loaded into the
22021 process; you can confirm it with the @code{info sharedlibrary}
22022 command, which will list @file{libinproctrace.so} as loaded in the
22023 process. You are now ready to install fast tracepoints, list static
22024 tracepoint markers, probe static tracepoints markers, and start
22027 @node Remote Configuration
22028 @section Remote Configuration
22031 @kindex show remote
22032 This section documents the configuration options available when
22033 debugging remote programs. For the options related to the File I/O
22034 extensions of the remote protocol, see @ref{system,
22035 system-call-allowed}.
22038 @item set remoteaddresssize @var{bits}
22039 @cindex address size for remote targets
22040 @cindex bits in remote address
22041 Set the maximum size of address in a memory packet to the specified
22042 number of bits. @value{GDBN} will mask off the address bits above
22043 that number, when it passes addresses to the remote target. The
22044 default value is the number of bits in the target's address.
22046 @item show remoteaddresssize
22047 Show the current value of remote address size in bits.
22049 @item set serial baud @var{n}
22050 @cindex baud rate for remote targets
22051 Set the baud rate for the remote serial I/O to @var{n} baud. The
22052 value is used to set the speed of the serial port used for debugging
22055 @item show serial baud
22056 Show the current speed of the remote connection.
22058 @item set serial parity @var{parity}
22059 Set the parity for the remote serial I/O. Supported values of @var{parity} are:
22060 @code{even}, @code{none}, and @code{odd}. The default is @code{none}.
22062 @item show serial parity
22063 Show the current parity of the serial port.
22065 @item set remotebreak
22066 @cindex interrupt remote programs
22067 @cindex BREAK signal instead of Ctrl-C
22068 @anchor{set remotebreak}
22069 If set to on, @value{GDBN} sends a @code{BREAK} signal to the remote
22070 when you type @kbd{Ctrl-c} to interrupt the program running
22071 on the remote. If set to off, @value{GDBN} sends the @samp{Ctrl-C}
22072 character instead. The default is off, since most remote systems
22073 expect to see @samp{Ctrl-C} as the interrupt signal.
22075 @item show remotebreak
22076 Show whether @value{GDBN} sends @code{BREAK} or @samp{Ctrl-C} to
22077 interrupt the remote program.
22079 @item set remoteflow on
22080 @itemx set remoteflow off
22081 @kindex set remoteflow
22082 Enable or disable hardware flow control (@code{RTS}/@code{CTS})
22083 on the serial port used to communicate to the remote target.
22085 @item show remoteflow
22086 @kindex show remoteflow
22087 Show the current setting of hardware flow control.
22089 @item set remotelogbase @var{base}
22090 Set the base (a.k.a.@: radix) of logging serial protocol
22091 communications to @var{base}. Supported values of @var{base} are:
22092 @code{ascii}, @code{octal}, and @code{hex}. The default is
22095 @item show remotelogbase
22096 Show the current setting of the radix for logging remote serial
22099 @item set remotelogfile @var{file}
22100 @cindex record serial communications on file
22101 Record remote serial communications on the named @var{file}. The
22102 default is not to record at all.
22104 @item show remotelogfile
22105 Show the current setting of the file name on which to record the
22106 serial communications.
22108 @item set remotetimeout @var{num}
22109 @cindex timeout for serial communications
22110 @cindex remote timeout
22111 Set the timeout limit to wait for the remote target to respond to
22112 @var{num} seconds. The default is 2 seconds.
22114 @item show remotetimeout
22115 Show the current number of seconds to wait for the remote target
22118 @cindex limit hardware breakpoints and watchpoints
22119 @cindex remote target, limit break- and watchpoints
22120 @anchor{set remote hardware-watchpoint-limit}
22121 @anchor{set remote hardware-breakpoint-limit}
22122 @item set remote hardware-watchpoint-limit @var{limit}
22123 @itemx set remote hardware-breakpoint-limit @var{limit}
22124 Restrict @value{GDBN} to using @var{limit} remote hardware watchpoints
22125 or breakpoints. The @var{limit} can be set to 0 to disable hardware
22126 watchpoints or breakpoints, and @code{unlimited} for unlimited
22127 watchpoints or breakpoints.
22129 @item show remote hardware-watchpoint-limit
22130 @itemx show remote hardware-breakpoint-limit
22131 Show the current limit for the number of hardware watchpoints or
22132 breakpoints that @value{GDBN} can use.
22134 @cindex limit hardware watchpoints length
22135 @cindex remote target, limit watchpoints length
22136 @anchor{set remote hardware-watchpoint-length-limit}
22137 @item set remote hardware-watchpoint-length-limit @var{limit}
22138 Restrict @value{GDBN} to using @var{limit} bytes for the maximum
22139 length of a remote hardware watchpoint. A @var{limit} of 0 disables
22140 hardware watchpoints and @code{unlimited} allows watchpoints of any
22143 @item show remote hardware-watchpoint-length-limit
22144 Show the current limit (in bytes) of the maximum length of
22145 a remote hardware watchpoint.
22147 @item set remote exec-file @var{filename}
22148 @itemx show remote exec-file
22149 @anchor{set remote exec-file}
22150 @cindex executable file, for remote target
22151 Select the file used for @code{run} with @code{target
22152 extended-remote}. This should be set to a filename valid on the
22153 target system. If it is not set, the target will use a default
22154 filename (e.g.@: the last program run).
22156 @item set remote interrupt-sequence
22157 @cindex interrupt remote programs
22158 @cindex select Ctrl-C, BREAK or BREAK-g
22159 Allow the user to select one of @samp{Ctrl-C}, a @code{BREAK} or
22160 @samp{BREAK-g} as the
22161 sequence to the remote target in order to interrupt the execution.
22162 @samp{Ctrl-C} is a default. Some system prefers @code{BREAK} which
22163 is high level of serial line for some certain time.
22164 Linux kernel prefers @samp{BREAK-g}, a.k.a Magic SysRq g.
22165 It is @code{BREAK} signal followed by character @code{g}.
22167 @item show interrupt-sequence
22168 Show which of @samp{Ctrl-C}, @code{BREAK} or @code{BREAK-g}
22169 is sent by @value{GDBN} to interrupt the remote program.
22170 @code{BREAK-g} is BREAK signal followed by @code{g} and
22171 also known as Magic SysRq g.
22173 @item set remote interrupt-on-connect
22174 @cindex send interrupt-sequence on start
22175 Specify whether interrupt-sequence is sent to remote target when
22176 @value{GDBN} connects to it. This is mostly needed when you debug
22177 Linux kernel. Linux kernel expects @code{BREAK} followed by @code{g}
22178 which is known as Magic SysRq g in order to connect @value{GDBN}.
22180 @item show interrupt-on-connect
22181 Show whether interrupt-sequence is sent
22182 to remote target when @value{GDBN} connects to it.
22186 @item set tcp auto-retry on
22187 @cindex auto-retry, for remote TCP target
22188 Enable auto-retry for remote TCP connections. This is useful if the remote
22189 debugging agent is launched in parallel with @value{GDBN}; there is a race
22190 condition because the agent may not become ready to accept the connection
22191 before @value{GDBN} attempts to connect. When auto-retry is
22192 enabled, if the initial attempt to connect fails, @value{GDBN} reattempts
22193 to establish the connection using the timeout specified by
22194 @code{set tcp connect-timeout}.
22196 @item set tcp auto-retry off
22197 Do not auto-retry failed TCP connections.
22199 @item show tcp auto-retry
22200 Show the current auto-retry setting.
22202 @item set tcp connect-timeout @var{seconds}
22203 @itemx set tcp connect-timeout unlimited
22204 @cindex connection timeout, for remote TCP target
22205 @cindex timeout, for remote target connection
22206 Set the timeout for establishing a TCP connection to the remote target to
22207 @var{seconds}. The timeout affects both polling to retry failed connections
22208 (enabled by @code{set tcp auto-retry on}) and waiting for connections
22209 that are merely slow to complete, and represents an approximate cumulative
22210 value. If @var{seconds} is @code{unlimited}, there is no timeout and
22211 @value{GDBN} will keep attempting to establish a connection forever,
22212 unless interrupted with @kbd{Ctrl-c}. The default is 15 seconds.
22214 @item show tcp connect-timeout
22215 Show the current connection timeout setting.
22218 @cindex remote packets, enabling and disabling
22219 The @value{GDBN} remote protocol autodetects the packets supported by
22220 your debugging stub. If you need to override the autodetection, you
22221 can use these commands to enable or disable individual packets. Each
22222 packet can be set to @samp{on} (the remote target supports this
22223 packet), @samp{off} (the remote target does not support this packet),
22224 or @samp{auto} (detect remote target support for this packet). They
22225 all default to @samp{auto}. For more information about each packet,
22226 see @ref{Remote Protocol}.
22228 During normal use, you should not have to use any of these commands.
22229 If you do, that may be a bug in your remote debugging stub, or a bug
22230 in @value{GDBN}. You may want to report the problem to the
22231 @value{GDBN} developers.
22233 For each packet @var{name}, the command to enable or disable the
22234 packet is @code{set remote @var{name}-packet}. The available settings
22237 @multitable @columnfractions 0.28 0.32 0.25
22240 @tab Related Features
22242 @item @code{fetch-register}
22244 @tab @code{info registers}
22246 @item @code{set-register}
22250 @item @code{binary-download}
22252 @tab @code{load}, @code{set}
22254 @item @code{read-aux-vector}
22255 @tab @code{qXfer:auxv:read}
22256 @tab @code{info auxv}
22258 @item @code{symbol-lookup}
22259 @tab @code{qSymbol}
22260 @tab Detecting multiple threads
22262 @item @code{attach}
22263 @tab @code{vAttach}
22266 @item @code{verbose-resume}
22268 @tab Stepping or resuming multiple threads
22274 @item @code{software-breakpoint}
22278 @item @code{hardware-breakpoint}
22282 @item @code{write-watchpoint}
22286 @item @code{read-watchpoint}
22290 @item @code{access-watchpoint}
22294 @item @code{pid-to-exec-file}
22295 @tab @code{qXfer:exec-file:read}
22296 @tab @code{attach}, @code{run}
22298 @item @code{target-features}
22299 @tab @code{qXfer:features:read}
22300 @tab @code{set architecture}
22302 @item @code{library-info}
22303 @tab @code{qXfer:libraries:read}
22304 @tab @code{info sharedlibrary}
22306 @item @code{memory-map}
22307 @tab @code{qXfer:memory-map:read}
22308 @tab @code{info mem}
22310 @item @code{read-sdata-object}
22311 @tab @code{qXfer:sdata:read}
22312 @tab @code{print $_sdata}
22314 @item @code{read-spu-object}
22315 @tab @code{qXfer:spu:read}
22316 @tab @code{info spu}
22318 @item @code{write-spu-object}
22319 @tab @code{qXfer:spu:write}
22320 @tab @code{info spu}
22322 @item @code{read-siginfo-object}
22323 @tab @code{qXfer:siginfo:read}
22324 @tab @code{print $_siginfo}
22326 @item @code{write-siginfo-object}
22327 @tab @code{qXfer:siginfo:write}
22328 @tab @code{set $_siginfo}
22330 @item @code{threads}
22331 @tab @code{qXfer:threads:read}
22332 @tab @code{info threads}
22334 @item @code{get-thread-local-@*storage-address}
22335 @tab @code{qGetTLSAddr}
22336 @tab Displaying @code{__thread} variables
22338 @item @code{get-thread-information-block-address}
22339 @tab @code{qGetTIBAddr}
22340 @tab Display MS-Windows Thread Information Block.
22342 @item @code{search-memory}
22343 @tab @code{qSearch:memory}
22346 @item @code{supported-packets}
22347 @tab @code{qSupported}
22348 @tab Remote communications parameters
22350 @item @code{catch-syscalls}
22351 @tab @code{QCatchSyscalls}
22352 @tab @code{catch syscall}
22354 @item @code{pass-signals}
22355 @tab @code{QPassSignals}
22356 @tab @code{handle @var{signal}}
22358 @item @code{program-signals}
22359 @tab @code{QProgramSignals}
22360 @tab @code{handle @var{signal}}
22362 @item @code{hostio-close-packet}
22363 @tab @code{vFile:close}
22364 @tab @code{remote get}, @code{remote put}
22366 @item @code{hostio-open-packet}
22367 @tab @code{vFile:open}
22368 @tab @code{remote get}, @code{remote put}
22370 @item @code{hostio-pread-packet}
22371 @tab @code{vFile:pread}
22372 @tab @code{remote get}, @code{remote put}
22374 @item @code{hostio-pwrite-packet}
22375 @tab @code{vFile:pwrite}
22376 @tab @code{remote get}, @code{remote put}
22378 @item @code{hostio-unlink-packet}
22379 @tab @code{vFile:unlink}
22380 @tab @code{remote delete}
22382 @item @code{hostio-readlink-packet}
22383 @tab @code{vFile:readlink}
22386 @item @code{hostio-fstat-packet}
22387 @tab @code{vFile:fstat}
22390 @item @code{hostio-setfs-packet}
22391 @tab @code{vFile:setfs}
22394 @item @code{noack-packet}
22395 @tab @code{QStartNoAckMode}
22396 @tab Packet acknowledgment
22398 @item @code{osdata}
22399 @tab @code{qXfer:osdata:read}
22400 @tab @code{info os}
22402 @item @code{query-attached}
22403 @tab @code{qAttached}
22404 @tab Querying remote process attach state.
22406 @item @code{trace-buffer-size}
22407 @tab @code{QTBuffer:size}
22408 @tab @code{set trace-buffer-size}
22410 @item @code{trace-status}
22411 @tab @code{qTStatus}
22412 @tab @code{tstatus}
22414 @item @code{traceframe-info}
22415 @tab @code{qXfer:traceframe-info:read}
22416 @tab Traceframe info
22418 @item @code{install-in-trace}
22419 @tab @code{InstallInTrace}
22420 @tab Install tracepoint in tracing
22422 @item @code{disable-randomization}
22423 @tab @code{QDisableRandomization}
22424 @tab @code{set disable-randomization}
22426 @item @code{startup-with-shell}
22427 @tab @code{QStartupWithShell}
22428 @tab @code{set startup-with-shell}
22430 @item @code{environment-hex-encoded}
22431 @tab @code{QEnvironmentHexEncoded}
22432 @tab @code{set environment}
22434 @item @code{environment-unset}
22435 @tab @code{QEnvironmentUnset}
22436 @tab @code{unset environment}
22438 @item @code{environment-reset}
22439 @tab @code{QEnvironmentReset}
22440 @tab @code{Reset the inferior environment (i.e., unset user-set variables)}
22442 @item @code{set-working-dir}
22443 @tab @code{QSetWorkingDir}
22444 @tab @code{set cwd}
22446 @item @code{conditional-breakpoints-packet}
22447 @tab @code{Z0 and Z1}
22448 @tab @code{Support for target-side breakpoint condition evaluation}
22450 @item @code{multiprocess-extensions}
22451 @tab @code{multiprocess extensions}
22452 @tab Debug multiple processes and remote process PID awareness
22454 @item @code{swbreak-feature}
22455 @tab @code{swbreak stop reason}
22458 @item @code{hwbreak-feature}
22459 @tab @code{hwbreak stop reason}
22462 @item @code{fork-event-feature}
22463 @tab @code{fork stop reason}
22466 @item @code{vfork-event-feature}
22467 @tab @code{vfork stop reason}
22470 @item @code{exec-event-feature}
22471 @tab @code{exec stop reason}
22474 @item @code{thread-events}
22475 @tab @code{QThreadEvents}
22476 @tab Tracking thread lifetime.
22478 @item @code{no-resumed-stop-reply}
22479 @tab @code{no resumed thread left stop reply}
22480 @tab Tracking thread lifetime.
22485 @section Implementing a Remote Stub
22487 @cindex debugging stub, example
22488 @cindex remote stub, example
22489 @cindex stub example, remote debugging
22490 The stub files provided with @value{GDBN} implement the target side of the
22491 communication protocol, and the @value{GDBN} side is implemented in the
22492 @value{GDBN} source file @file{remote.c}. Normally, you can simply allow
22493 these subroutines to communicate, and ignore the details. (If you're
22494 implementing your own stub file, you can still ignore the details: start
22495 with one of the existing stub files. @file{sparc-stub.c} is the best
22496 organized, and therefore the easiest to read.)
22498 @cindex remote serial debugging, overview
22499 To debug a program running on another machine (the debugging
22500 @dfn{target} machine), you must first arrange for all the usual
22501 prerequisites for the program to run by itself. For example, for a C
22506 A startup routine to set up the C runtime environment; these usually
22507 have a name like @file{crt0}. The startup routine may be supplied by
22508 your hardware supplier, or you may have to write your own.
22511 A C subroutine library to support your program's
22512 subroutine calls, notably managing input and output.
22515 A way of getting your program to the other machine---for example, a
22516 download program. These are often supplied by the hardware
22517 manufacturer, but you may have to write your own from hardware
22521 The next step is to arrange for your program to use a serial port to
22522 communicate with the machine where @value{GDBN} is running (the @dfn{host}
22523 machine). In general terms, the scheme looks like this:
22527 @value{GDBN} already understands how to use this protocol; when everything
22528 else is set up, you can simply use the @samp{target remote} command
22529 (@pxref{Targets,,Specifying a Debugging Target}).
22531 @item On the target,
22532 you must link with your program a few special-purpose subroutines that
22533 implement the @value{GDBN} remote serial protocol. The file containing these
22534 subroutines is called a @dfn{debugging stub}.
22536 On certain remote targets, you can use an auxiliary program
22537 @code{gdbserver} instead of linking a stub into your program.
22538 @xref{Server,,Using the @code{gdbserver} Program}, for details.
22541 The debugging stub is specific to the architecture of the remote
22542 machine; for example, use @file{sparc-stub.c} to debug programs on
22545 @cindex remote serial stub list
22546 These working remote stubs are distributed with @value{GDBN}:
22551 @cindex @file{i386-stub.c}
22554 For Intel 386 and compatible architectures.
22557 @cindex @file{m68k-stub.c}
22558 @cindex Motorola 680x0
22560 For Motorola 680x0 architectures.
22563 @cindex @file{sh-stub.c}
22566 For Renesas SH architectures.
22569 @cindex @file{sparc-stub.c}
22571 For @sc{sparc} architectures.
22573 @item sparcl-stub.c
22574 @cindex @file{sparcl-stub.c}
22577 For Fujitsu @sc{sparclite} architectures.
22581 The @file{README} file in the @value{GDBN} distribution may list other
22582 recently added stubs.
22585 * Stub Contents:: What the stub can do for you
22586 * Bootstrapping:: What you must do for the stub
22587 * Debug Session:: Putting it all together
22590 @node Stub Contents
22591 @subsection What the Stub Can Do for You
22593 @cindex remote serial stub
22594 The debugging stub for your architecture supplies these three
22598 @item set_debug_traps
22599 @findex set_debug_traps
22600 @cindex remote serial stub, initialization
22601 This routine arranges for @code{handle_exception} to run when your
22602 program stops. You must call this subroutine explicitly in your
22603 program's startup code.
22605 @item handle_exception
22606 @findex handle_exception
22607 @cindex remote serial stub, main routine
22608 This is the central workhorse, but your program never calls it
22609 explicitly---the setup code arranges for @code{handle_exception} to
22610 run when a trap is triggered.
22612 @code{handle_exception} takes control when your program stops during
22613 execution (for example, on a breakpoint), and mediates communications
22614 with @value{GDBN} on the host machine. This is where the communications
22615 protocol is implemented; @code{handle_exception} acts as the @value{GDBN}
22616 representative on the target machine. It begins by sending summary
22617 information on the state of your program, then continues to execute,
22618 retrieving and transmitting any information @value{GDBN} needs, until you
22619 execute a @value{GDBN} command that makes your program resume; at that point,
22620 @code{handle_exception} returns control to your own code on the target
22624 @cindex @code{breakpoint} subroutine, remote
22625 Use this auxiliary subroutine to make your program contain a
22626 breakpoint. Depending on the particular situation, this may be the only
22627 way for @value{GDBN} to get control. For instance, if your target
22628 machine has some sort of interrupt button, you won't need to call this;
22629 pressing the interrupt button transfers control to
22630 @code{handle_exception}---in effect, to @value{GDBN}. On some machines,
22631 simply receiving characters on the serial port may also trigger a trap;
22632 again, in that situation, you don't need to call @code{breakpoint} from
22633 your own program---simply running @samp{target remote} from the host
22634 @value{GDBN} session gets control.
22636 Call @code{breakpoint} if none of these is true, or if you simply want
22637 to make certain your program stops at a predetermined point for the
22638 start of your debugging session.
22641 @node Bootstrapping
22642 @subsection What You Must Do for the Stub
22644 @cindex remote stub, support routines
22645 The debugging stubs that come with @value{GDBN} are set up for a particular
22646 chip architecture, but they have no information about the rest of your
22647 debugging target machine.
22649 First of all you need to tell the stub how to communicate with the
22653 @item int getDebugChar()
22654 @findex getDebugChar
22655 Write this subroutine to read a single character from the serial port.
22656 It may be identical to @code{getchar} for your target system; a
22657 different name is used to allow you to distinguish the two if you wish.
22659 @item void putDebugChar(int)
22660 @findex putDebugChar
22661 Write this subroutine to write a single character to the serial port.
22662 It may be identical to @code{putchar} for your target system; a
22663 different name is used to allow you to distinguish the two if you wish.
22666 @cindex control C, and remote debugging
22667 @cindex interrupting remote targets
22668 If you want @value{GDBN} to be able to stop your program while it is
22669 running, you need to use an interrupt-driven serial driver, and arrange
22670 for it to stop when it receives a @code{^C} (@samp{\003}, the control-C
22671 character). That is the character which @value{GDBN} uses to tell the
22672 remote system to stop.
22674 Getting the debugging target to return the proper status to @value{GDBN}
22675 probably requires changes to the standard stub; one quick and dirty way
22676 is to just execute a breakpoint instruction (the ``dirty'' part is that
22677 @value{GDBN} reports a @code{SIGTRAP} instead of a @code{SIGINT}).
22679 Other routines you need to supply are:
22682 @item void exceptionHandler (int @var{exception_number}, void *@var{exception_address})
22683 @findex exceptionHandler
22684 Write this function to install @var{exception_address} in the exception
22685 handling tables. You need to do this because the stub does not have any
22686 way of knowing what the exception handling tables on your target system
22687 are like (for example, the processor's table might be in @sc{rom},
22688 containing entries which point to a table in @sc{ram}).
22689 The @var{exception_number} specifies the exception which should be changed;
22690 its meaning is architecture-dependent (for example, different numbers
22691 might represent divide by zero, misaligned access, etc). When this
22692 exception occurs, control should be transferred directly to
22693 @var{exception_address}, and the processor state (stack, registers,
22694 and so on) should be just as it is when a processor exception occurs. So if
22695 you want to use a jump instruction to reach @var{exception_address}, it
22696 should be a simple jump, not a jump to subroutine.
22698 For the 386, @var{exception_address} should be installed as an interrupt
22699 gate so that interrupts are masked while the handler runs. The gate
22700 should be at privilege level 0 (the most privileged level). The
22701 @sc{sparc} and 68k stubs are able to mask interrupts themselves without
22702 help from @code{exceptionHandler}.
22704 @item void flush_i_cache()
22705 @findex flush_i_cache
22706 On @sc{sparc} and @sc{sparclite} only, write this subroutine to flush the
22707 instruction cache, if any, on your target machine. If there is no
22708 instruction cache, this subroutine may be a no-op.
22710 On target machines that have instruction caches, @value{GDBN} requires this
22711 function to make certain that the state of your program is stable.
22715 You must also make sure this library routine is available:
22718 @item void *memset(void *, int, int)
22720 This is the standard library function @code{memset} that sets an area of
22721 memory to a known value. If you have one of the free versions of
22722 @code{libc.a}, @code{memset} can be found there; otherwise, you must
22723 either obtain it from your hardware manufacturer, or write your own.
22726 If you do not use the GNU C compiler, you may need other standard
22727 library subroutines as well; this varies from one stub to another,
22728 but in general the stubs are likely to use any of the common library
22729 subroutines which @code{@value{NGCC}} generates as inline code.
22732 @node Debug Session
22733 @subsection Putting it All Together
22735 @cindex remote serial debugging summary
22736 In summary, when your program is ready to debug, you must follow these
22741 Make sure you have defined the supporting low-level routines
22742 (@pxref{Bootstrapping,,What You Must Do for the Stub}):
22744 @code{getDebugChar}, @code{putDebugChar},
22745 @code{flush_i_cache}, @code{memset}, @code{exceptionHandler}.
22749 Insert these lines in your program's startup code, before the main
22750 procedure is called:
22757 On some machines, when a breakpoint trap is raised, the hardware
22758 automatically makes the PC point to the instruction after the
22759 breakpoint. If your machine doesn't do that, you may need to adjust
22760 @code{handle_exception} to arrange for it to return to the instruction
22761 after the breakpoint on this first invocation, so that your program
22762 doesn't keep hitting the initial breakpoint instead of making
22766 For the 680x0 stub only, you need to provide a variable called
22767 @code{exceptionHook}. Normally you just use:
22770 void (*exceptionHook)() = 0;
22774 but if before calling @code{set_debug_traps}, you set it to point to a
22775 function in your program, that function is called when
22776 @code{@value{GDBN}} continues after stopping on a trap (for example, bus
22777 error). The function indicated by @code{exceptionHook} is called with
22778 one parameter: an @code{int} which is the exception number.
22781 Compile and link together: your program, the @value{GDBN} debugging stub for
22782 your target architecture, and the supporting subroutines.
22785 Make sure you have a serial connection between your target machine and
22786 the @value{GDBN} host, and identify the serial port on the host.
22789 @c The "remote" target now provides a `load' command, so we should
22790 @c document that. FIXME.
22791 Download your program to your target machine (or get it there by
22792 whatever means the manufacturer provides), and start it.
22795 Start @value{GDBN} on the host, and connect to the target
22796 (@pxref{Connecting,,Connecting to a Remote Target}).
22800 @node Configurations
22801 @chapter Configuration-Specific Information
22803 While nearly all @value{GDBN} commands are available for all native and
22804 cross versions of the debugger, there are some exceptions. This chapter
22805 describes things that are only available in certain configurations.
22807 There are three major categories of configurations: native
22808 configurations, where the host and target are the same, embedded
22809 operating system configurations, which are usually the same for several
22810 different processor architectures, and bare embedded processors, which
22811 are quite different from each other.
22816 * Embedded Processors::
22823 This section describes details specific to particular native
22827 * BSD libkvm Interface:: Debugging BSD kernel memory images
22828 * Process Information:: Process information
22829 * DJGPP Native:: Features specific to the DJGPP port
22830 * Cygwin Native:: Features specific to the Cygwin port
22831 * Hurd Native:: Features specific to @sc{gnu} Hurd
22832 * Darwin:: Features specific to Darwin
22833 * FreeBSD:: Features specific to FreeBSD
22836 @node BSD libkvm Interface
22837 @subsection BSD libkvm Interface
22840 @cindex kernel memory image
22841 @cindex kernel crash dump
22843 BSD-derived systems (FreeBSD/NetBSD/OpenBSD) have a kernel memory
22844 interface that provides a uniform interface for accessing kernel virtual
22845 memory images, including live systems and crash dumps. @value{GDBN}
22846 uses this interface to allow you to debug live kernels and kernel crash
22847 dumps on many native BSD configurations. This is implemented as a
22848 special @code{kvm} debugging target. For debugging a live system, load
22849 the currently running kernel into @value{GDBN} and connect to the
22853 (@value{GDBP}) @b{target kvm}
22856 For debugging crash dumps, provide the file name of the crash dump as an
22860 (@value{GDBP}) @b{target kvm /var/crash/bsd.0}
22863 Once connected to the @code{kvm} target, the following commands are
22869 Set current context from the @dfn{Process Control Block} (PCB) address.
22872 Set current context from proc address. This command isn't available on
22873 modern FreeBSD systems.
22876 @node Process Information
22877 @subsection Process Information
22879 @cindex examine process image
22880 @cindex process info via @file{/proc}
22882 Some operating systems provide interfaces to fetch additional
22883 information about running processes beyond memory and per-thread
22884 register state. If @value{GDBN} is configured for an operating system
22885 with a supported interface, the command @code{info proc} is available
22886 to report information about the process running your program, or about
22887 any process running on your system.
22889 One supported interface is a facility called @samp{/proc} that can be
22890 used to examine the image of a running process using file-system
22891 subroutines. This facility is supported on @sc{gnu}/Linux and Solaris
22894 On FreeBSD systems, system control nodes are used to query process
22897 In addition, some systems may provide additional process information
22898 in core files. Note that a core file may include a subset of the
22899 information available from a live process. Process information is
22900 currently avaiable from cores created on @sc{gnu}/Linux and FreeBSD
22907 @itemx info proc @var{process-id}
22908 Summarize available information about a process. If a
22909 process ID is specified by @var{process-id}, display information about
22910 that process; otherwise display information about the program being
22911 debugged. The summary includes the debugged process ID, the command
22912 line used to invoke it, its current working directory, and its
22913 executable file's absolute file name.
22915 On some systems, @var{process-id} can be of the form
22916 @samp{[@var{pid}]/@var{tid}} which specifies a certain thread ID
22917 within a process. If the optional @var{pid} part is missing, it means
22918 a thread from the process being debugged (the leading @samp{/} still
22919 needs to be present, or else @value{GDBN} will interpret the number as
22920 a process ID rather than a thread ID).
22922 @item info proc cmdline
22923 @cindex info proc cmdline
22924 Show the original command line of the process. This command is
22925 supported on @sc{gnu}/Linux and FreeBSD.
22927 @item info proc cwd
22928 @cindex info proc cwd
22929 Show the current working directory of the process. This command is
22930 supported on @sc{gnu}/Linux and FreeBSD.
22932 @item info proc exe
22933 @cindex info proc exe
22934 Show the name of executable of the process. This command is supported
22935 on @sc{gnu}/Linux and FreeBSD.
22937 @item info proc files
22938 @cindex info proc files
22939 Show the file descriptors open by the process. For each open file
22940 descriptor, @value{GDBN} shows its number, type (file, directory,
22941 character device, socket), file pointer offset, and the name of the
22942 resource open on the descriptor. The resource name can be a file name
22943 (for files, directories, and devices) or a protocol followed by socket
22944 address (for network connections). This command is supported on
22947 This example shows the open file descriptors for a process using a
22948 tty for standard input and output as well as two network sockets:
22951 (gdb) info proc files 22136
22955 FD Type Offset Flags Name
22956 text file - r-------- /usr/bin/ssh
22957 ctty chr - rw------- /dev/pts/20
22958 cwd dir - r-------- /usr/home/john
22959 root dir - r-------- /
22960 0 chr 0x32933a4 rw------- /dev/pts/20
22961 1 chr 0x32933a4 rw------- /dev/pts/20
22962 2 chr 0x32933a4 rw------- /dev/pts/20
22963 3 socket 0x0 rw----n-- tcp4 10.0.1.2:53014 -> 10.0.1.10:22
22964 4 socket 0x0 rw------- unix stream:/tmp/ssh-FIt89oAzOn5f/agent.2456
22967 @item info proc mappings
22968 @cindex memory address space mappings
22969 Report the memory address space ranges accessible in a process. On
22970 Solaris and FreeBSD systems, each memory range includes information on
22971 whether the process has read, write, or execute access rights to each
22972 range. On @sc{gnu}/Linux and FreeBSD systems, each memory range
22973 includes the object file which is mapped to that range.
22975 @item info proc stat
22976 @itemx info proc status
22977 @cindex process detailed status information
22978 Show additional process-related information, including the user ID and
22979 group ID; virtual memory usage; the signals that are pending, blocked,
22980 and ignored; its TTY; its consumption of system and user time; its
22981 stack size; its @samp{nice} value; etc. These commands are supported
22982 on @sc{gnu}/Linux and FreeBSD.
22984 For @sc{gnu}/Linux systems, see the @samp{proc} man page for more
22985 information (type @kbd{man 5 proc} from your shell prompt).
22987 For FreeBSD systems, @code{info proc stat} is an alias for @code{info
22990 @item info proc all
22991 Show all the information about the process described under all of the
22992 above @code{info proc} subcommands.
22995 @comment These sub-options of 'info proc' were not included when
22996 @comment procfs.c was re-written. Keep their descriptions around
22997 @comment against the day when someone finds the time to put them back in.
22998 @kindex info proc times
22999 @item info proc times
23000 Starting time, user CPU time, and system CPU time for your program and
23003 @kindex info proc id
23005 Report on the process IDs related to your program: its own process ID,
23006 the ID of its parent, the process group ID, and the session ID.
23009 @item set procfs-trace
23010 @kindex set procfs-trace
23011 @cindex @code{procfs} API calls
23012 This command enables and disables tracing of @code{procfs} API calls.
23014 @item show procfs-trace
23015 @kindex show procfs-trace
23016 Show the current state of @code{procfs} API call tracing.
23018 @item set procfs-file @var{file}
23019 @kindex set procfs-file
23020 Tell @value{GDBN} to write @code{procfs} API trace to the named
23021 @var{file}. @value{GDBN} appends the trace info to the previous
23022 contents of the file. The default is to display the trace on the
23025 @item show procfs-file
23026 @kindex show procfs-file
23027 Show the file to which @code{procfs} API trace is written.
23029 @item proc-trace-entry
23030 @itemx proc-trace-exit
23031 @itemx proc-untrace-entry
23032 @itemx proc-untrace-exit
23033 @kindex proc-trace-entry
23034 @kindex proc-trace-exit
23035 @kindex proc-untrace-entry
23036 @kindex proc-untrace-exit
23037 These commands enable and disable tracing of entries into and exits
23038 from the @code{syscall} interface.
23041 @kindex info pidlist
23042 @cindex process list, QNX Neutrino
23043 For QNX Neutrino only, this command displays the list of all the
23044 processes and all the threads within each process.
23047 @kindex info meminfo
23048 @cindex mapinfo list, QNX Neutrino
23049 For QNX Neutrino only, this command displays the list of all mapinfos.
23053 @subsection Features for Debugging @sc{djgpp} Programs
23054 @cindex @sc{djgpp} debugging
23055 @cindex native @sc{djgpp} debugging
23056 @cindex MS-DOS-specific commands
23059 @sc{djgpp} is a port of the @sc{gnu} development tools to MS-DOS and
23060 MS-Windows. @sc{djgpp} programs are 32-bit protected-mode programs
23061 that use the @dfn{DPMI} (DOS Protected-Mode Interface) API to run on
23062 top of real-mode DOS systems and their emulations.
23064 @value{GDBN} supports native debugging of @sc{djgpp} programs, and
23065 defines a few commands specific to the @sc{djgpp} port. This
23066 subsection describes those commands.
23071 This is a prefix of @sc{djgpp}-specific commands which print
23072 information about the target system and important OS structures.
23075 @cindex MS-DOS system info
23076 @cindex free memory information (MS-DOS)
23077 @item info dos sysinfo
23078 This command displays assorted information about the underlying
23079 platform: the CPU type and features, the OS version and flavor, the
23080 DPMI version, and the available conventional and DPMI memory.
23085 @cindex segment descriptor tables
23086 @cindex descriptor tables display
23088 @itemx info dos ldt
23089 @itemx info dos idt
23090 These 3 commands display entries from, respectively, Global, Local,
23091 and Interrupt Descriptor Tables (GDT, LDT, and IDT). The descriptor
23092 tables are data structures which store a descriptor for each segment
23093 that is currently in use. The segment's selector is an index into a
23094 descriptor table; the table entry for that index holds the
23095 descriptor's base address and limit, and its attributes and access
23098 A typical @sc{djgpp} program uses 3 segments: a code segment, a data
23099 segment (used for both data and the stack), and a DOS segment (which
23100 allows access to DOS/BIOS data structures and absolute addresses in
23101 conventional memory). However, the DPMI host will usually define
23102 additional segments in order to support the DPMI environment.
23104 @cindex garbled pointers
23105 These commands allow to display entries from the descriptor tables.
23106 Without an argument, all entries from the specified table are
23107 displayed. An argument, which should be an integer expression, means
23108 display a single entry whose index is given by the argument. For
23109 example, here's a convenient way to display information about the
23110 debugged program's data segment:
23113 @exdent @code{(@value{GDBP}) info dos ldt $ds}
23114 @exdent @code{0x13f: base=0x11970000 limit=0x0009ffff 32-Bit Data (Read/Write, Exp-up)}
23118 This comes in handy when you want to see whether a pointer is outside
23119 the data segment's limit (i.e.@: @dfn{garbled}).
23121 @cindex page tables display (MS-DOS)
23123 @itemx info dos pte
23124 These two commands display entries from, respectively, the Page
23125 Directory and the Page Tables. Page Directories and Page Tables are
23126 data structures which control how virtual memory addresses are mapped
23127 into physical addresses. A Page Table includes an entry for every
23128 page of memory that is mapped into the program's address space; there
23129 may be several Page Tables, each one holding up to 4096 entries. A
23130 Page Directory has up to 4096 entries, one each for every Page Table
23131 that is currently in use.
23133 Without an argument, @kbd{info dos pde} displays the entire Page
23134 Directory, and @kbd{info dos pte} displays all the entries in all of
23135 the Page Tables. An argument, an integer expression, given to the
23136 @kbd{info dos pde} command means display only that entry from the Page
23137 Directory table. An argument given to the @kbd{info dos pte} command
23138 means display entries from a single Page Table, the one pointed to by
23139 the specified entry in the Page Directory.
23141 @cindex direct memory access (DMA) on MS-DOS
23142 These commands are useful when your program uses @dfn{DMA} (Direct
23143 Memory Access), which needs physical addresses to program the DMA
23146 These commands are supported only with some DPMI servers.
23148 @cindex physical address from linear address
23149 @item info dos address-pte @var{addr}
23150 This command displays the Page Table entry for a specified linear
23151 address. The argument @var{addr} is a linear address which should
23152 already have the appropriate segment's base address added to it,
23153 because this command accepts addresses which may belong to @emph{any}
23154 segment. For example, here's how to display the Page Table entry for
23155 the page where a variable @code{i} is stored:
23158 @exdent @code{(@value{GDBP}) info dos address-pte __djgpp_base_address + (char *)&i}
23159 @exdent @code{Page Table entry for address 0x11a00d30:}
23160 @exdent @code{Base=0x02698000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0xd30}
23164 This says that @code{i} is stored at offset @code{0xd30} from the page
23165 whose physical base address is @code{0x02698000}, and shows all the
23166 attributes of that page.
23168 Note that you must cast the addresses of variables to a @code{char *},
23169 since otherwise the value of @code{__djgpp_base_address}, the base
23170 address of all variables and functions in a @sc{djgpp} program, will
23171 be added using the rules of C pointer arithmetics: if @code{i} is
23172 declared an @code{int}, @value{GDBN} will add 4 times the value of
23173 @code{__djgpp_base_address} to the address of @code{i}.
23175 Here's another example, it displays the Page Table entry for the
23179 @exdent @code{(@value{GDBP}) info dos address-pte *((unsigned *)&_go32_info_block + 3)}
23180 @exdent @code{Page Table entry for address 0x29110:}
23181 @exdent @code{Base=0x00029000 Dirty Acc. Not-Cached Write-Back Usr Read-Write +0x110}
23185 (The @code{+ 3} offset is because the transfer buffer's address is the
23186 3rd member of the @code{_go32_info_block} structure.) The output
23187 clearly shows that this DPMI server maps the addresses in conventional
23188 memory 1:1, i.e.@: the physical (@code{0x00029000} + @code{0x110}) and
23189 linear (@code{0x29110}) addresses are identical.
23191 This command is supported only with some DPMI servers.
23194 @cindex DOS serial data link, remote debugging
23195 In addition to native debugging, the DJGPP port supports remote
23196 debugging via a serial data link. The following commands are specific
23197 to remote serial debugging in the DJGPP port of @value{GDBN}.
23200 @kindex set com1base
23201 @kindex set com1irq
23202 @kindex set com2base
23203 @kindex set com2irq
23204 @kindex set com3base
23205 @kindex set com3irq
23206 @kindex set com4base
23207 @kindex set com4irq
23208 @item set com1base @var{addr}
23209 This command sets the base I/O port address of the @file{COM1} serial
23212 @item set com1irq @var{irq}
23213 This command sets the @dfn{Interrupt Request} (@code{IRQ}) line to use
23214 for the @file{COM1} serial port.
23216 There are similar commands @samp{set com2base}, @samp{set com3irq},
23217 etc.@: for setting the port address and the @code{IRQ} lines for the
23220 @kindex show com1base
23221 @kindex show com1irq
23222 @kindex show com2base
23223 @kindex show com2irq
23224 @kindex show com3base
23225 @kindex show com3irq
23226 @kindex show com4base
23227 @kindex show com4irq
23228 The related commands @samp{show com1base}, @samp{show com1irq} etc.@:
23229 display the current settings of the base address and the @code{IRQ}
23230 lines used by the COM ports.
23233 @kindex info serial
23234 @cindex DOS serial port status
23235 This command prints the status of the 4 DOS serial ports. For each
23236 port, it prints whether it's active or not, its I/O base address and
23237 IRQ number, whether it uses a 16550-style FIFO, its baudrate, and the
23238 counts of various errors encountered so far.
23242 @node Cygwin Native
23243 @subsection Features for Debugging MS Windows PE Executables
23244 @cindex MS Windows debugging
23245 @cindex native Cygwin debugging
23246 @cindex Cygwin-specific commands
23248 @value{GDBN} supports native debugging of MS Windows programs, including
23249 DLLs with and without symbolic debugging information.
23251 @cindex Ctrl-BREAK, MS-Windows
23252 @cindex interrupt debuggee on MS-Windows
23253 MS-Windows programs that call @code{SetConsoleMode} to switch off the
23254 special meaning of the @samp{Ctrl-C} keystroke cannot be interrupted
23255 by typing @kbd{C-c}. For this reason, @value{GDBN} on MS-Windows
23256 supports @kbd{C-@key{BREAK}} as an alternative interrupt key
23257 sequence, which can be used to interrupt the debuggee even if it
23260 There are various additional Cygwin-specific commands, described in
23261 this section. Working with DLLs that have no debugging symbols is
23262 described in @ref{Non-debug DLL Symbols}.
23267 This is a prefix of MS Windows-specific commands which print
23268 information about the target system and important OS structures.
23270 @item info w32 selector
23271 This command displays information returned by
23272 the Win32 API @code{GetThreadSelectorEntry} function.
23273 It takes an optional argument that is evaluated to
23274 a long value to give the information about this given selector.
23275 Without argument, this command displays information
23276 about the six segment registers.
23278 @item info w32 thread-information-block
23279 This command displays thread specific information stored in the
23280 Thread Information Block (readable on the X86 CPU family using @code{$fs}
23281 selector for 32-bit programs and @code{$gs} for 64-bit programs).
23283 @kindex signal-event
23284 @item signal-event @var{id}
23285 This command signals an event with user-provided @var{id}. Used to resume
23286 crashing process when attached to it using MS-Windows JIT debugging (AeDebug).
23288 To use it, create or edit the following keys in
23289 @code{HKLM\SOFTWARE\Microsoft\Windows NT\CurrentVersion\AeDebug} and/or
23290 @code{HKLM\SOFTWARE\Wow6432Node\Microsoft\Windows NT\CurrentVersion\AeDebug}
23291 (for x86_64 versions):
23295 @code{Debugger} (REG_SZ) --- a command to launch the debugger.
23296 Suggested command is: @code{@var{fully-qualified-path-to-gdb.exe} -ex
23297 "attach %ld" -ex "signal-event %ld" -ex "continue"}.
23299 The first @code{%ld} will be replaced by the process ID of the
23300 crashing process, the second @code{%ld} will be replaced by the ID of
23301 the event that blocks the crashing process, waiting for @value{GDBN}
23305 @code{Auto} (REG_SZ) --- either @code{1} or @code{0}. @code{1} will
23306 make the system run debugger specified by the Debugger key
23307 automatically, @code{0} will cause a dialog box with ``OK'' and
23308 ``Cancel'' buttons to appear, which allows the user to either
23309 terminate the crashing process (OK) or debug it (Cancel).
23312 @kindex set cygwin-exceptions
23313 @cindex debugging the Cygwin DLL
23314 @cindex Cygwin DLL, debugging
23315 @item set cygwin-exceptions @var{mode}
23316 If @var{mode} is @code{on}, @value{GDBN} will break on exceptions that
23317 happen inside the Cygwin DLL. If @var{mode} is @code{off},
23318 @value{GDBN} will delay recognition of exceptions, and may ignore some
23319 exceptions which seem to be caused by internal Cygwin DLL
23320 ``bookkeeping''. This option is meant primarily for debugging the
23321 Cygwin DLL itself; the default value is @code{off} to avoid annoying
23322 @value{GDBN} users with false @code{SIGSEGV} signals.
23324 @kindex show cygwin-exceptions
23325 @item show cygwin-exceptions
23326 Displays whether @value{GDBN} will break on exceptions that happen
23327 inside the Cygwin DLL itself.
23329 @kindex set new-console
23330 @item set new-console @var{mode}
23331 If @var{mode} is @code{on} the debuggee will
23332 be started in a new console on next start.
23333 If @var{mode} is @code{off}, the debuggee will
23334 be started in the same console as the debugger.
23336 @kindex show new-console
23337 @item show new-console
23338 Displays whether a new console is used
23339 when the debuggee is started.
23341 @kindex set new-group
23342 @item set new-group @var{mode}
23343 This boolean value controls whether the debuggee should
23344 start a new group or stay in the same group as the debugger.
23345 This affects the way the Windows OS handles
23348 @kindex show new-group
23349 @item show new-group
23350 Displays current value of new-group boolean.
23352 @kindex set debugevents
23353 @item set debugevents
23354 This boolean value adds debug output concerning kernel events related
23355 to the debuggee seen by the debugger. This includes events that
23356 signal thread and process creation and exit, DLL loading and
23357 unloading, console interrupts, and debugging messages produced by the
23358 Windows @code{OutputDebugString} API call.
23360 @kindex set debugexec
23361 @item set debugexec
23362 This boolean value adds debug output concerning execute events
23363 (such as resume thread) seen by the debugger.
23365 @kindex set debugexceptions
23366 @item set debugexceptions
23367 This boolean value adds debug output concerning exceptions in the
23368 debuggee seen by the debugger.
23370 @kindex set debugmemory
23371 @item set debugmemory
23372 This boolean value adds debug output concerning debuggee memory reads
23373 and writes by the debugger.
23377 This boolean values specifies whether the debuggee is called
23378 via a shell or directly (default value is on).
23382 Displays if the debuggee will be started with a shell.
23387 * Non-debug DLL Symbols:: Support for DLLs without debugging symbols
23390 @node Non-debug DLL Symbols
23391 @subsubsection Support for DLLs without Debugging Symbols
23392 @cindex DLLs with no debugging symbols
23393 @cindex Minimal symbols and DLLs
23395 Very often on windows, some of the DLLs that your program relies on do
23396 not include symbolic debugging information (for example,
23397 @file{kernel32.dll}). When @value{GDBN} doesn't recognize any debugging
23398 symbols in a DLL, it relies on the minimal amount of symbolic
23399 information contained in the DLL's export table. This section
23400 describes working with such symbols, known internally to @value{GDBN} as
23401 ``minimal symbols''.
23403 Note that before the debugged program has started execution, no DLLs
23404 will have been loaded. The easiest way around this problem is simply to
23405 start the program --- either by setting a breakpoint or letting the
23406 program run once to completion.
23408 @subsubsection DLL Name Prefixes
23410 In keeping with the naming conventions used by the Microsoft debugging
23411 tools, DLL export symbols are made available with a prefix based on the
23412 DLL name, for instance @code{KERNEL32!CreateFileA}. The plain name is
23413 also entered into the symbol table, so @code{CreateFileA} is often
23414 sufficient. In some cases there will be name clashes within a program
23415 (particularly if the executable itself includes full debugging symbols)
23416 necessitating the use of the fully qualified name when referring to the
23417 contents of the DLL. Use single-quotes around the name to avoid the
23418 exclamation mark (``!'') being interpreted as a language operator.
23420 Note that the internal name of the DLL may be all upper-case, even
23421 though the file name of the DLL is lower-case, or vice-versa. Since
23422 symbols within @value{GDBN} are @emph{case-sensitive} this may cause
23423 some confusion. If in doubt, try the @code{info functions} and
23424 @code{info variables} commands or even @code{maint print msymbols}
23425 (@pxref{Symbols}). Here's an example:
23428 (@value{GDBP}) info function CreateFileA
23429 All functions matching regular expression "CreateFileA":
23431 Non-debugging symbols:
23432 0x77e885f4 CreateFileA
23433 0x77e885f4 KERNEL32!CreateFileA
23437 (@value{GDBP}) info function !
23438 All functions matching regular expression "!":
23440 Non-debugging symbols:
23441 0x6100114c cygwin1!__assert
23442 0x61004034 cygwin1!_dll_crt0@@0
23443 0x61004240 cygwin1!dll_crt0(per_process *)
23447 @subsubsection Working with Minimal Symbols
23449 Symbols extracted from a DLL's export table do not contain very much
23450 type information. All that @value{GDBN} can do is guess whether a symbol
23451 refers to a function or variable depending on the linker section that
23452 contains the symbol. Also note that the actual contents of the memory
23453 contained in a DLL are not available unless the program is running. This
23454 means that you cannot examine the contents of a variable or disassemble
23455 a function within a DLL without a running program.
23457 Variables are generally treated as pointers and dereferenced
23458 automatically. For this reason, it is often necessary to prefix a
23459 variable name with the address-of operator (``&'') and provide explicit
23460 type information in the command. Here's an example of the type of
23464 (@value{GDBP}) print 'cygwin1!__argv'
23465 'cygwin1!__argv' has unknown type; cast it to its declared type
23469 (@value{GDBP}) x 'cygwin1!__argv'
23470 'cygwin1!__argv' has unknown type; cast it to its declared type
23473 And two possible solutions:
23476 (@value{GDBP}) print ((char **)'cygwin1!__argv')[0]
23477 $2 = 0x22fd98 "/cygdrive/c/mydirectory/myprogram"
23481 (@value{GDBP}) x/2x &'cygwin1!__argv'
23482 0x610c0aa8 <cygwin1!__argv>: 0x10021608 0x00000000
23483 (@value{GDBP}) x/x 0x10021608
23484 0x10021608: 0x0022fd98
23485 (@value{GDBP}) x/s 0x0022fd98
23486 0x22fd98: "/cygdrive/c/mydirectory/myprogram"
23489 Setting a break point within a DLL is possible even before the program
23490 starts execution. However, under these circumstances, @value{GDBN} can't
23491 examine the initial instructions of the function in order to skip the
23492 function's frame set-up code. You can work around this by using ``*&''
23493 to set the breakpoint at a raw memory address:
23496 (@value{GDBP}) break *&'python22!PyOS_Readline'
23497 Breakpoint 1 at 0x1e04eff0
23500 The author of these extensions is not entirely convinced that setting a
23501 break point within a shared DLL like @file{kernel32.dll} is completely
23505 @subsection Commands Specific to @sc{gnu} Hurd Systems
23506 @cindex @sc{gnu} Hurd debugging
23508 This subsection describes @value{GDBN} commands specific to the
23509 @sc{gnu} Hurd native debugging.
23514 @kindex set signals@r{, Hurd command}
23515 @kindex set sigs@r{, Hurd command}
23516 This command toggles the state of inferior signal interception by
23517 @value{GDBN}. Mach exceptions, such as breakpoint traps, are not
23518 affected by this command. @code{sigs} is a shorthand alias for
23523 @kindex show signals@r{, Hurd command}
23524 @kindex show sigs@r{, Hurd command}
23525 Show the current state of intercepting inferior's signals.
23527 @item set signal-thread
23528 @itemx set sigthread
23529 @kindex set signal-thread
23530 @kindex set sigthread
23531 This command tells @value{GDBN} which thread is the @code{libc} signal
23532 thread. That thread is run when a signal is delivered to a running
23533 process. @code{set sigthread} is the shorthand alias of @code{set
23536 @item show signal-thread
23537 @itemx show sigthread
23538 @kindex show signal-thread
23539 @kindex show sigthread
23540 These two commands show which thread will run when the inferior is
23541 delivered a signal.
23544 @kindex set stopped@r{, Hurd command}
23545 This commands tells @value{GDBN} that the inferior process is stopped,
23546 as with the @code{SIGSTOP} signal. The stopped process can be
23547 continued by delivering a signal to it.
23550 @kindex show stopped@r{, Hurd command}
23551 This command shows whether @value{GDBN} thinks the debuggee is
23554 @item set exceptions
23555 @kindex set exceptions@r{, Hurd command}
23556 Use this command to turn off trapping of exceptions in the inferior.
23557 When exception trapping is off, neither breakpoints nor
23558 single-stepping will work. To restore the default, set exception
23561 @item show exceptions
23562 @kindex show exceptions@r{, Hurd command}
23563 Show the current state of trapping exceptions in the inferior.
23565 @item set task pause
23566 @kindex set task@r{, Hurd commands}
23567 @cindex task attributes (@sc{gnu} Hurd)
23568 @cindex pause current task (@sc{gnu} Hurd)
23569 This command toggles task suspension when @value{GDBN} has control.
23570 Setting it to on takes effect immediately, and the task is suspended
23571 whenever @value{GDBN} gets control. Setting it to off will take
23572 effect the next time the inferior is continued. If this option is set
23573 to off, you can use @code{set thread default pause on} or @code{set
23574 thread pause on} (see below) to pause individual threads.
23576 @item show task pause
23577 @kindex show task@r{, Hurd commands}
23578 Show the current state of task suspension.
23580 @item set task detach-suspend-count
23581 @cindex task suspend count
23582 @cindex detach from task, @sc{gnu} Hurd
23583 This command sets the suspend count the task will be left with when
23584 @value{GDBN} detaches from it.
23586 @item show task detach-suspend-count
23587 Show the suspend count the task will be left with when detaching.
23589 @item set task exception-port
23590 @itemx set task excp
23591 @cindex task exception port, @sc{gnu} Hurd
23592 This command sets the task exception port to which @value{GDBN} will
23593 forward exceptions. The argument should be the value of the @dfn{send
23594 rights} of the task. @code{set task excp} is a shorthand alias.
23596 @item set noninvasive
23597 @cindex noninvasive task options
23598 This command switches @value{GDBN} to a mode that is the least
23599 invasive as far as interfering with the inferior is concerned. This
23600 is the same as using @code{set task pause}, @code{set exceptions}, and
23601 @code{set signals} to values opposite to the defaults.
23603 @item info send-rights
23604 @itemx info receive-rights
23605 @itemx info port-rights
23606 @itemx info port-sets
23607 @itemx info dead-names
23610 @cindex send rights, @sc{gnu} Hurd
23611 @cindex receive rights, @sc{gnu} Hurd
23612 @cindex port rights, @sc{gnu} Hurd
23613 @cindex port sets, @sc{gnu} Hurd
23614 @cindex dead names, @sc{gnu} Hurd
23615 These commands display information about, respectively, send rights,
23616 receive rights, port rights, port sets, and dead names of a task.
23617 There are also shorthand aliases: @code{info ports} for @code{info
23618 port-rights} and @code{info psets} for @code{info port-sets}.
23620 @item set thread pause
23621 @kindex set thread@r{, Hurd command}
23622 @cindex thread properties, @sc{gnu} Hurd
23623 @cindex pause current thread (@sc{gnu} Hurd)
23624 This command toggles current thread suspension when @value{GDBN} has
23625 control. Setting it to on takes effect immediately, and the current
23626 thread is suspended whenever @value{GDBN} gets control. Setting it to
23627 off will take effect the next time the inferior is continued.
23628 Normally, this command has no effect, since when @value{GDBN} has
23629 control, the whole task is suspended. However, if you used @code{set
23630 task pause off} (see above), this command comes in handy to suspend
23631 only the current thread.
23633 @item show thread pause
23634 @kindex show thread@r{, Hurd command}
23635 This command shows the state of current thread suspension.
23637 @item set thread run
23638 This command sets whether the current thread is allowed to run.
23640 @item show thread run
23641 Show whether the current thread is allowed to run.
23643 @item set thread detach-suspend-count
23644 @cindex thread suspend count, @sc{gnu} Hurd
23645 @cindex detach from thread, @sc{gnu} Hurd
23646 This command sets the suspend count @value{GDBN} will leave on a
23647 thread when detaching. This number is relative to the suspend count
23648 found by @value{GDBN} when it notices the thread; use @code{set thread
23649 takeover-suspend-count} to force it to an absolute value.
23651 @item show thread detach-suspend-count
23652 Show the suspend count @value{GDBN} will leave on the thread when
23655 @item set thread exception-port
23656 @itemx set thread excp
23657 Set the thread exception port to which to forward exceptions. This
23658 overrides the port set by @code{set task exception-port} (see above).
23659 @code{set thread excp} is the shorthand alias.
23661 @item set thread takeover-suspend-count
23662 Normally, @value{GDBN}'s thread suspend counts are relative to the
23663 value @value{GDBN} finds when it notices each thread. This command
23664 changes the suspend counts to be absolute instead.
23666 @item set thread default
23667 @itemx show thread default
23668 @cindex thread default settings, @sc{gnu} Hurd
23669 Each of the above @code{set thread} commands has a @code{set thread
23670 default} counterpart (e.g., @code{set thread default pause}, @code{set
23671 thread default exception-port}, etc.). The @code{thread default}
23672 variety of commands sets the default thread properties for all
23673 threads; you can then change the properties of individual threads with
23674 the non-default commands.
23681 @value{GDBN} provides the following commands specific to the Darwin target:
23684 @item set debug darwin @var{num}
23685 @kindex set debug darwin
23686 When set to a non zero value, enables debugging messages specific to
23687 the Darwin support. Higher values produce more verbose output.
23689 @item show debug darwin
23690 @kindex show debug darwin
23691 Show the current state of Darwin messages.
23693 @item set debug mach-o @var{num}
23694 @kindex set debug mach-o
23695 When set to a non zero value, enables debugging messages while
23696 @value{GDBN} is reading Darwin object files. (@dfn{Mach-O} is the
23697 file format used on Darwin for object and executable files.) Higher
23698 values produce more verbose output. This is a command to diagnose
23699 problems internal to @value{GDBN} and should not be needed in normal
23702 @item show debug mach-o
23703 @kindex show debug mach-o
23704 Show the current state of Mach-O file messages.
23706 @item set mach-exceptions on
23707 @itemx set mach-exceptions off
23708 @kindex set mach-exceptions
23709 On Darwin, faults are first reported as a Mach exception and are then
23710 mapped to a Posix signal. Use this command to turn on trapping of
23711 Mach exceptions in the inferior. This might be sometimes useful to
23712 better understand the cause of a fault. The default is off.
23714 @item show mach-exceptions
23715 @kindex show mach-exceptions
23716 Show the current state of exceptions trapping.
23720 @subsection FreeBSD
23723 When the ABI of a system call is changed in the FreeBSD kernel, this
23724 is implemented by leaving a compatibility system call using the old
23725 ABI at the existing number and allocating a new system call number for
23726 the version using the new ABI. As a convenience, when a system call
23727 is caught by name (@pxref{catch syscall}), compatibility system calls
23730 For example, FreeBSD 12 introduced a new variant of the @code{kevent}
23731 system call and catching the @code{kevent} system call by name catches
23735 (@value{GDBP}) catch syscall kevent
23736 Catchpoint 1 (syscalls 'freebsd11_kevent' [363] 'kevent' [560])
23742 @section Embedded Operating Systems
23744 This section describes configurations involving the debugging of
23745 embedded operating systems that are available for several different
23748 @value{GDBN} includes the ability to debug programs running on
23749 various real-time operating systems.
23751 @node Embedded Processors
23752 @section Embedded Processors
23754 This section goes into details specific to particular embedded
23757 @cindex send command to simulator
23758 Whenever a specific embedded processor has a simulator, @value{GDBN}
23759 allows to send an arbitrary command to the simulator.
23762 @item sim @var{command}
23763 @kindex sim@r{, a command}
23764 Send an arbitrary @var{command} string to the simulator. Consult the
23765 documentation for the specific simulator in use for information about
23766 acceptable commands.
23771 * ARC:: Synopsys ARC
23773 * M68K:: Motorola M68K
23774 * MicroBlaze:: Xilinx MicroBlaze
23775 * MIPS Embedded:: MIPS Embedded
23776 * OpenRISC 1000:: OpenRISC 1000 (or1k)
23777 * PowerPC Embedded:: PowerPC Embedded
23780 * Super-H:: Renesas Super-H
23784 @subsection Synopsys ARC
23785 @cindex Synopsys ARC
23786 @cindex ARC specific commands
23792 @value{GDBN} provides the following ARC-specific commands:
23795 @item set debug arc
23796 @kindex set debug arc
23797 Control the level of ARC specific debug messages. Use 0 for no messages (the
23798 default), 1 for debug messages, and 2 for even more debug messages.
23800 @item show debug arc
23801 @kindex show debug arc
23802 Show the level of ARC specific debugging in operation.
23804 @item maint print arc arc-instruction @var{address}
23805 @kindex maint print arc arc-instruction
23806 Print internal disassembler information about instruction at a given address.
23813 @value{GDBN} provides the following ARM-specific commands:
23816 @item set arm disassembler
23818 This commands selects from a list of disassembly styles. The
23819 @code{"std"} style is the standard style.
23821 @item show arm disassembler
23823 Show the current disassembly style.
23825 @item set arm apcs32
23826 @cindex ARM 32-bit mode
23827 This command toggles ARM operation mode between 32-bit and 26-bit.
23829 @item show arm apcs32
23830 Display the current usage of the ARM 32-bit mode.
23832 @item set arm fpu @var{fputype}
23833 This command sets the ARM floating-point unit (FPU) type. The
23834 argument @var{fputype} can be one of these:
23838 Determine the FPU type by querying the OS ABI.
23840 Software FPU, with mixed-endian doubles on little-endian ARM
23843 GCC-compiled FPA co-processor.
23845 Software FPU with pure-endian doubles.
23851 Show the current type of the FPU.
23854 This command forces @value{GDBN} to use the specified ABI.
23857 Show the currently used ABI.
23859 @item set arm fallback-mode (arm|thumb|auto)
23860 @value{GDBN} uses the symbol table, when available, to determine
23861 whether instructions are ARM or Thumb. This command controls
23862 @value{GDBN}'s default behavior when the symbol table is not
23863 available. The default is @samp{auto}, which causes @value{GDBN} to
23864 use the current execution mode (from the @code{T} bit in the @code{CPSR}
23867 @item show arm fallback-mode
23868 Show the current fallback instruction mode.
23870 @item set arm force-mode (arm|thumb|auto)
23871 This command overrides use of the symbol table to determine whether
23872 instructions are ARM or Thumb. The default is @samp{auto}, which
23873 causes @value{GDBN} to use the symbol table and then the setting
23874 of @samp{set arm fallback-mode}.
23876 @item show arm force-mode
23877 Show the current forced instruction mode.
23879 @item set debug arm
23880 Toggle whether to display ARM-specific debugging messages from the ARM
23881 target support subsystem.
23883 @item show debug arm
23884 Show whether ARM-specific debugging messages are enabled.
23888 @item target sim @r{[}@var{simargs}@r{]} @dots{}
23889 The @value{GDBN} ARM simulator accepts the following optional arguments.
23892 @item --swi-support=@var{type}
23893 Tell the simulator which SWI interfaces to support. The argument
23894 @var{type} may be a comma separated list of the following values.
23895 The default value is @code{all}.
23910 The Motorola m68k configuration includes ColdFire support.
23913 @subsection MicroBlaze
23914 @cindex Xilinx MicroBlaze
23915 @cindex XMD, Xilinx Microprocessor Debugger
23917 The MicroBlaze is a soft-core processor supported on various Xilinx
23918 FPGAs, such as Spartan or Virtex series. Boards with these processors
23919 usually have JTAG ports which connect to a host system running the Xilinx
23920 Embedded Development Kit (EDK) or Software Development Kit (SDK).
23921 This host system is used to download the configuration bitstream to
23922 the target FPGA. The Xilinx Microprocessor Debugger (XMD) program
23923 communicates with the target board using the JTAG interface and
23924 presents a @code{gdbserver} interface to the board. By default
23925 @code{xmd} uses port @code{1234}. (While it is possible to change
23926 this default port, it requires the use of undocumented @code{xmd}
23927 commands. Contact Xilinx support if you need to do this.)
23929 Use these GDB commands to connect to the MicroBlaze target processor.
23932 @item target remote :1234
23933 Use this command to connect to the target if you are running @value{GDBN}
23934 on the same system as @code{xmd}.
23936 @item target remote @var{xmd-host}:1234
23937 Use this command to connect to the target if it is connected to @code{xmd}
23938 running on a different system named @var{xmd-host}.
23941 Use this command to download a program to the MicroBlaze target.
23943 @item set debug microblaze @var{n}
23944 Enable MicroBlaze-specific debugging messages if non-zero.
23946 @item show debug microblaze @var{n}
23947 Show MicroBlaze-specific debugging level.
23950 @node MIPS Embedded
23951 @subsection @acronym{MIPS} Embedded
23954 @value{GDBN} supports these special commands for @acronym{MIPS} targets:
23957 @item set mipsfpu double
23958 @itemx set mipsfpu single
23959 @itemx set mipsfpu none
23960 @itemx set mipsfpu auto
23961 @itemx show mipsfpu
23962 @kindex set mipsfpu
23963 @kindex show mipsfpu
23964 @cindex @acronym{MIPS} remote floating point
23965 @cindex floating point, @acronym{MIPS} remote
23966 If your target board does not support the @acronym{MIPS} floating point
23967 coprocessor, you should use the command @samp{set mipsfpu none} (if you
23968 need this, you may wish to put the command in your @value{GDBN} init
23969 file). This tells @value{GDBN} how to find the return value of
23970 functions which return floating point values. It also allows
23971 @value{GDBN} to avoid saving the floating point registers when calling
23972 functions on the board. If you are using a floating point coprocessor
23973 with only single precision floating point support, as on the @sc{r4650}
23974 processor, use the command @samp{set mipsfpu single}. The default
23975 double precision floating point coprocessor may be selected using
23976 @samp{set mipsfpu double}.
23978 In previous versions the only choices were double precision or no
23979 floating point, so @samp{set mipsfpu on} will select double precision
23980 and @samp{set mipsfpu off} will select no floating point.
23982 As usual, you can inquire about the @code{mipsfpu} variable with
23983 @samp{show mipsfpu}.
23986 @node OpenRISC 1000
23987 @subsection OpenRISC 1000
23988 @cindex OpenRISC 1000
23991 The OpenRISC 1000 provides a free RISC instruction set architecture. It is
23992 mainly provided as a soft-core which can run on Xilinx, Altera and other
23995 @value{GDBN} for OpenRISC supports the below commands when connecting to
24003 Runs the builtin CPU simulator which can run very basic
24004 programs but does not support most hardware functions like MMU.
24005 For more complex use cases the user is advised to run an external
24006 target, and connect using @samp{target remote}.
24008 Example: @code{target sim}
24010 @item set debug or1k
24011 Toggle whether to display OpenRISC-specific debugging messages from the
24012 OpenRISC target support subsystem.
24014 @item show debug or1k
24015 Show whether OpenRISC-specific debugging messages are enabled.
24018 @node PowerPC Embedded
24019 @subsection PowerPC Embedded
24021 @cindex DVC register
24022 @value{GDBN} supports using the DVC (Data Value Compare) register to
24023 implement in hardware simple hardware watchpoint conditions of the form:
24026 (@value{GDBP}) watch @var{ADDRESS|VARIABLE} \
24027 if @var{ADDRESS|VARIABLE} == @var{CONSTANT EXPRESSION}
24030 The DVC register will be automatically used when @value{GDBN} detects
24031 such pattern in a condition expression, and the created watchpoint uses one
24032 debug register (either the @code{exact-watchpoints} option is on and the
24033 variable is scalar, or the variable has a length of one byte). This feature
24034 is available in native @value{GDBN} running on a Linux kernel version 2.6.34
24037 When running on PowerPC embedded processors, @value{GDBN} automatically uses
24038 ranged hardware watchpoints, unless the @code{exact-watchpoints} option is on,
24039 in which case watchpoints using only one debug register are created when
24040 watching variables of scalar types.
24042 You can create an artificial array to watch an arbitrary memory
24043 region using one of the following commands (@pxref{Expressions}):
24046 (@value{GDBP}) watch *((char *) @var{address})@@@var{length}
24047 (@value{GDBP}) watch @{char[@var{length}]@} @var{address}
24050 PowerPC embedded processors support masked watchpoints. See the discussion
24051 about the @code{mask} argument in @ref{Set Watchpoints}.
24053 @cindex ranged breakpoint
24054 PowerPC embedded processors support hardware accelerated
24055 @dfn{ranged breakpoints}. A ranged breakpoint stops execution of
24056 the inferior whenever it executes an instruction at any address within
24057 the range it specifies. To set a ranged breakpoint in @value{GDBN},
24058 use the @code{break-range} command.
24060 @value{GDBN} provides the following PowerPC-specific commands:
24063 @kindex break-range
24064 @item break-range @var{start-location}, @var{end-location}
24065 Set a breakpoint for an address range given by
24066 @var{start-location} and @var{end-location}, which can specify a function name,
24067 a line number, an offset of lines from the current line or from the start
24068 location, or an address of an instruction (see @ref{Specify Location},
24069 for a list of all the possible ways to specify a @var{location}.)
24070 The breakpoint will stop execution of the inferior whenever it
24071 executes an instruction at any address within the specified range,
24072 (including @var{start-location} and @var{end-location}.)
24074 @kindex set powerpc
24075 @item set powerpc soft-float
24076 @itemx show powerpc soft-float
24077 Force @value{GDBN} to use (or not use) a software floating point calling
24078 convention. By default, @value{GDBN} selects the calling convention based
24079 on the selected architecture and the provided executable file.
24081 @item set powerpc vector-abi
24082 @itemx show powerpc vector-abi
24083 Force @value{GDBN} to use the specified calling convention for vector
24084 arguments and return values. The valid options are @samp{auto};
24085 @samp{generic}, to avoid vector registers even if they are present;
24086 @samp{altivec}, to use AltiVec registers; and @samp{spe} to use SPE
24087 registers. By default, @value{GDBN} selects the calling convention
24088 based on the selected architecture and the provided executable file.
24090 @item set powerpc exact-watchpoints
24091 @itemx show powerpc exact-watchpoints
24092 Allow @value{GDBN} to use only one debug register when watching a variable
24093 of scalar type, thus assuming that the variable is accessed through the
24094 address of its first byte.
24099 @subsection Atmel AVR
24102 When configured for debugging the Atmel AVR, @value{GDBN} supports the
24103 following AVR-specific commands:
24106 @item info io_registers
24107 @kindex info io_registers@r{, AVR}
24108 @cindex I/O registers (Atmel AVR)
24109 This command displays information about the AVR I/O registers. For
24110 each register, @value{GDBN} prints its number and value.
24117 When configured for debugging CRIS, @value{GDBN} provides the
24118 following CRIS-specific commands:
24121 @item set cris-version @var{ver}
24122 @cindex CRIS version
24123 Set the current CRIS version to @var{ver}, either @samp{10} or @samp{32}.
24124 The CRIS version affects register names and sizes. This command is useful in
24125 case autodetection of the CRIS version fails.
24127 @item show cris-version
24128 Show the current CRIS version.
24130 @item set cris-dwarf2-cfi
24131 @cindex DWARF-2 CFI and CRIS
24132 Set the usage of DWARF-2 CFI for CRIS debugging. The default is @samp{on}.
24133 Change to @samp{off} when using @code{gcc-cris} whose version is below
24136 @item show cris-dwarf2-cfi
24137 Show the current state of using DWARF-2 CFI.
24139 @item set cris-mode @var{mode}
24141 Set the current CRIS mode to @var{mode}. It should only be changed when
24142 debugging in guru mode, in which case it should be set to
24143 @samp{guru} (the default is @samp{normal}).
24145 @item show cris-mode
24146 Show the current CRIS mode.
24150 @subsection Renesas Super-H
24153 For the Renesas Super-H processor, @value{GDBN} provides these
24157 @item set sh calling-convention @var{convention}
24158 @kindex set sh calling-convention
24159 Set the calling-convention used when calling functions from @value{GDBN}.
24160 Allowed values are @samp{gcc}, which is the default setting, and @samp{renesas}.
24161 With the @samp{gcc} setting, functions are called using the @value{NGCC} calling
24162 convention. If the DWARF-2 information of the called function specifies
24163 that the function follows the Renesas calling convention, the function
24164 is called using the Renesas calling convention. If the calling convention
24165 is set to @samp{renesas}, the Renesas calling convention is always used,
24166 regardless of the DWARF-2 information. This can be used to override the
24167 default of @samp{gcc} if debug information is missing, or the compiler
24168 does not emit the DWARF-2 calling convention entry for a function.
24170 @item show sh calling-convention
24171 @kindex show sh calling-convention
24172 Show the current calling convention setting.
24177 @node Architectures
24178 @section Architectures
24180 This section describes characteristics of architectures that affect
24181 all uses of @value{GDBN} with the architecture, both native and cross.
24188 * HPPA:: HP PA architecture
24189 * SPU:: Cell Broadband Engine SPU architecture
24197 @subsection AArch64
24198 @cindex AArch64 support
24200 When @value{GDBN} is debugging the AArch64 architecture, it provides the
24201 following special commands:
24204 @item set debug aarch64
24205 @kindex set debug aarch64
24206 This command determines whether AArch64 architecture-specific debugging
24207 messages are to be displayed.
24209 @item show debug aarch64
24210 Show whether AArch64 debugging messages are displayed.
24214 @subsubsection AArch64 SVE.
24215 @cindex AArch64 SVE.
24217 When @value{GDBN} is debugging the AArch64 architecture, if the Scalable Vector
24218 Extension (SVE) is present, then @value{GDBN} will provide the vector registers
24219 @code{$z0} through @code{$z31}, vector predicate registers @code{$p0} through
24220 @code{$p15}, and the @code{$ffr} register. In addition, the pseudo register
24221 @code{$vg} will be provided. This is the vector granule for the current thread
24222 and represents the number of 64-bit chunks in an SVE @code{z} register.
24224 If the vector length changes, then the @code{$vg} register will be updated,
24225 but the lengths of the @code{z} and @code{p} registers will not change. This
24226 is a known limitation of @value{GDBN} and does not affect the execution of the
24231 @subsection x86 Architecture-specific Issues
24234 @item set struct-convention @var{mode}
24235 @kindex set struct-convention
24236 @cindex struct return convention
24237 @cindex struct/union returned in registers
24238 Set the convention used by the inferior to return @code{struct}s and
24239 @code{union}s from functions to @var{mode}. Possible values of
24240 @var{mode} are @code{"pcc"}, @code{"reg"}, and @code{"default"} (the
24241 default). @code{"default"} or @code{"pcc"} means that @code{struct}s
24242 are returned on the stack, while @code{"reg"} means that a
24243 @code{struct} or a @code{union} whose size is 1, 2, 4, or 8 bytes will
24244 be returned in a register.
24246 @item show struct-convention
24247 @kindex show struct-convention
24248 Show the current setting of the convention to return @code{struct}s
24253 @subsubsection Intel @dfn{Memory Protection Extensions} (MPX).
24254 @cindex Intel Memory Protection Extensions (MPX).
24256 Memory Protection Extension (MPX) adds the bound registers @samp{BND0}
24257 @footnote{The register named with capital letters represent the architecture
24258 registers.} through @samp{BND3}. Bound registers store a pair of 64-bit values
24259 which are the lower bound and upper bound. Bounds are effective addresses or
24260 memory locations. The upper bounds are architecturally represented in 1's
24261 complement form. A bound having lower bound = 0, and upper bound = 0
24262 (1's complement of all bits set) will allow access to the entire address space.
24264 @samp{BND0} through @samp{BND3} are represented in @value{GDBN} as @samp{bnd0raw}
24265 through @samp{bnd3raw}. Pseudo registers @samp{bnd0} through @samp{bnd3}
24266 display the upper bound performing the complement of one operation on the
24267 upper bound value, i.e.@ when upper bound in @samp{bnd0raw} is 0 in the
24268 @value{GDBN} @samp{bnd0} it will be @code{0xfff@dots{}}. In this sense it
24269 can also be noted that the upper bounds are inclusive.
24271 As an example, assume that the register BND0 holds bounds for a pointer having
24272 access allowed for the range between 0x32 and 0x71. The values present on
24273 bnd0raw and bnd registers are presented as follows:
24276 bnd0raw = @{0x32, 0xffffffff8e@}
24277 bnd0 = @{lbound = 0x32, ubound = 0x71@} : size 64
24280 This way the raw value can be accessed via bnd0raw@dots{}bnd3raw. Any
24281 change on bnd0@dots{}bnd3 or bnd0raw@dots{}bnd3raw is reflect on its
24282 counterpart. When the bnd0@dots{}bnd3 registers are displayed via
24283 Python, the display includes the memory size, in bits, accessible to
24286 Bounds can also be stored in bounds tables, which are stored in
24287 application memory. These tables store bounds for pointers by specifying
24288 the bounds pointer's value along with its bounds. Evaluating and changing
24289 bounds located in bound tables is therefore interesting while investigating
24290 bugs on MPX context. @value{GDBN} provides commands for this purpose:
24293 @item show mpx bound @var{pointer}
24294 @kindex show mpx bound
24295 Display bounds of the given @var{pointer}.
24297 @item set mpx bound @var{pointer}, @var{lbound}, @var{ubound}
24298 @kindex set mpx bound
24299 Set the bounds of a pointer in the bound table.
24300 This command takes three parameters: @var{pointer} is the pointers
24301 whose bounds are to be changed, @var{lbound} and @var{ubound} are new values
24302 for lower and upper bounds respectively.
24305 When you call an inferior function on an Intel MPX enabled program,
24306 GDB sets the inferior's bound registers to the init (disabled) state
24307 before calling the function. As a consequence, bounds checks for the
24308 pointer arguments passed to the function will always pass.
24310 This is necessary because when you call an inferior function, the
24311 program is usually in the middle of the execution of other function.
24312 Since at that point bound registers are in an arbitrary state, not
24313 clearing them would lead to random bound violations in the called
24316 You can still examine the influence of the bound registers on the
24317 execution of the called function by stopping the execution of the
24318 called function at its prologue, setting bound registers, and
24319 continuing the execution. For example:
24323 Breakpoint 2 at 0x4009de: file i386-mpx-call.c, line 47.
24324 $ print upper (a, b, c, d, 1)
24325 Breakpoint 2, upper (a=0x0, b=0x6e0000005b, c=0x0, d=0x0, len=48)....
24327 @{lbound = 0x0, ubound = ffffffff@} : size -1
24330 At this last step the value of bnd0 can be changed for investigation of bound
24331 violations caused along the execution of the call. In order to know how to
24332 set the bound registers or bound table for the call consult the ABI.
24337 See the following section.
24340 @subsection @acronym{MIPS}
24342 @cindex stack on Alpha
24343 @cindex stack on @acronym{MIPS}
24344 @cindex Alpha stack
24345 @cindex @acronym{MIPS} stack
24346 Alpha- and @acronym{MIPS}-based computers use an unusual stack frame, which
24347 sometimes requires @value{GDBN} to search backward in the object code to
24348 find the beginning of a function.
24350 @cindex response time, @acronym{MIPS} debugging
24351 To improve response time (especially for embedded applications, where
24352 @value{GDBN} may be restricted to a slow serial line for this search)
24353 you may want to limit the size of this search, using one of these
24357 @cindex @code{heuristic-fence-post} (Alpha, @acronym{MIPS})
24358 @item set heuristic-fence-post @var{limit}
24359 Restrict @value{GDBN} to examining at most @var{limit} bytes in its
24360 search for the beginning of a function. A value of @var{0} (the
24361 default) means there is no limit. However, except for @var{0}, the
24362 larger the limit the more bytes @code{heuristic-fence-post} must search
24363 and therefore the longer it takes to run. You should only need to use
24364 this command when debugging a stripped executable.
24366 @item show heuristic-fence-post
24367 Display the current limit.
24371 These commands are available @emph{only} when @value{GDBN} is configured
24372 for debugging programs on Alpha or @acronym{MIPS} processors.
24374 Several @acronym{MIPS}-specific commands are available when debugging @acronym{MIPS}
24378 @item set mips abi @var{arg}
24379 @kindex set mips abi
24380 @cindex set ABI for @acronym{MIPS}
24381 Tell @value{GDBN} which @acronym{MIPS} ABI is used by the inferior. Possible
24382 values of @var{arg} are:
24386 The default ABI associated with the current binary (this is the
24396 @item show mips abi
24397 @kindex show mips abi
24398 Show the @acronym{MIPS} ABI used by @value{GDBN} to debug the inferior.
24400 @item set mips compression @var{arg}
24401 @kindex set mips compression
24402 @cindex code compression, @acronym{MIPS}
24403 Tell @value{GDBN} which @acronym{MIPS} compressed
24404 @acronym{ISA, Instruction Set Architecture} encoding is used by the
24405 inferior. @value{GDBN} uses this for code disassembly and other
24406 internal interpretation purposes. This setting is only referred to
24407 when no executable has been associated with the debugging session or
24408 the executable does not provide information about the encoding it uses.
24409 Otherwise this setting is automatically updated from information
24410 provided by the executable.
24412 Possible values of @var{arg} are @samp{mips16} and @samp{micromips}.
24413 The default compressed @acronym{ISA} encoding is @samp{mips16}, as
24414 executables containing @acronym{MIPS16} code frequently are not
24415 identified as such.
24417 This setting is ``sticky''; that is, it retains its value across
24418 debugging sessions until reset either explicitly with this command or
24419 implicitly from an executable.
24421 The compiler and/or assembler typically add symbol table annotations to
24422 identify functions compiled for the @acronym{MIPS16} or
24423 @acronym{microMIPS} @acronym{ISA}s. If these function-scope annotations
24424 are present, @value{GDBN} uses them in preference to the global
24425 compressed @acronym{ISA} encoding setting.
24427 @item show mips compression
24428 @kindex show mips compression
24429 Show the @acronym{MIPS} compressed @acronym{ISA} encoding used by
24430 @value{GDBN} to debug the inferior.
24433 @itemx show mipsfpu
24434 @xref{MIPS Embedded, set mipsfpu}.
24436 @item set mips mask-address @var{arg}
24437 @kindex set mips mask-address
24438 @cindex @acronym{MIPS} addresses, masking
24439 This command determines whether the most-significant 32 bits of 64-bit
24440 @acronym{MIPS} addresses are masked off. The argument @var{arg} can be
24441 @samp{on}, @samp{off}, or @samp{auto}. The latter is the default
24442 setting, which lets @value{GDBN} determine the correct value.
24444 @item show mips mask-address
24445 @kindex show mips mask-address
24446 Show whether the upper 32 bits of @acronym{MIPS} addresses are masked off or
24449 @item set remote-mips64-transfers-32bit-regs
24450 @kindex set remote-mips64-transfers-32bit-regs
24451 This command controls compatibility with 64-bit @acronym{MIPS} targets that
24452 transfer data in 32-bit quantities. If you have an old @acronym{MIPS} 64 target
24453 that transfers 32 bits for some registers, like @sc{sr} and @sc{fsr},
24454 and 64 bits for other registers, set this option to @samp{on}.
24456 @item show remote-mips64-transfers-32bit-regs
24457 @kindex show remote-mips64-transfers-32bit-regs
24458 Show the current setting of compatibility with older @acronym{MIPS} 64 targets.
24460 @item set debug mips
24461 @kindex set debug mips
24462 This command turns on and off debugging messages for the @acronym{MIPS}-specific
24463 target code in @value{GDBN}.
24465 @item show debug mips
24466 @kindex show debug mips
24467 Show the current setting of @acronym{MIPS} debugging messages.
24473 @cindex HPPA support
24475 When @value{GDBN} is debugging the HP PA architecture, it provides the
24476 following special commands:
24479 @item set debug hppa
24480 @kindex set debug hppa
24481 This command determines whether HPPA architecture-specific debugging
24482 messages are to be displayed.
24484 @item show debug hppa
24485 Show whether HPPA debugging messages are displayed.
24487 @item maint print unwind @var{address}
24488 @kindex maint print unwind@r{, HPPA}
24489 This command displays the contents of the unwind table entry at the
24490 given @var{address}.
24496 @subsection Cell Broadband Engine SPU architecture
24497 @cindex Cell Broadband Engine
24500 When @value{GDBN} is debugging the Cell Broadband Engine SPU architecture,
24501 it provides the following special commands:
24504 @item info spu event
24506 Display SPU event facility status. Shows current event mask
24507 and pending event status.
24509 @item info spu signal
24510 Display SPU signal notification facility status. Shows pending
24511 signal-control word and signal notification mode of both signal
24512 notification channels.
24514 @item info spu mailbox
24515 Display SPU mailbox facility status. Shows all pending entries,
24516 in order of processing, in each of the SPU Write Outbound,
24517 SPU Write Outbound Interrupt, and SPU Read Inbound mailboxes.
24520 Display MFC DMA status. Shows all pending commands in the MFC
24521 DMA queue. For each entry, opcode, tag, class IDs, effective
24522 and local store addresses and transfer size are shown.
24524 @item info spu proxydma
24525 Display MFC Proxy-DMA status. Shows all pending commands in the MFC
24526 Proxy-DMA queue. For each entry, opcode, tag, class IDs, effective
24527 and local store addresses and transfer size are shown.
24531 When @value{GDBN} is debugging a combined PowerPC/SPU application
24532 on the Cell Broadband Engine, it provides in addition the following
24536 @item set spu stop-on-load @var{arg}
24538 Set whether to stop for new SPE threads. When set to @code{on}, @value{GDBN}
24539 will give control to the user when a new SPE thread enters its @code{main}
24540 function. The default is @code{off}.
24542 @item show spu stop-on-load
24544 Show whether to stop for new SPE threads.
24546 @item set spu auto-flush-cache @var{arg}
24547 Set whether to automatically flush the software-managed cache. When set to
24548 @code{on}, @value{GDBN} will automatically cause the SPE software-managed
24549 cache to be flushed whenever SPE execution stops. This provides a consistent
24550 view of PowerPC memory that is accessed via the cache. If an application
24551 does not use the software-managed cache, this option has no effect.
24553 @item show spu auto-flush-cache
24554 Show whether to automatically flush the software-managed cache.
24559 @subsection PowerPC
24560 @cindex PowerPC architecture
24562 When @value{GDBN} is debugging the PowerPC architecture, it provides a set of
24563 pseudo-registers to enable inspection of 128-bit wide Decimal Floating Point
24564 numbers stored in the floating point registers. These values must be stored
24565 in two consecutive registers, always starting at an even register like
24566 @code{f0} or @code{f2}.
24568 The pseudo-registers go from @code{$dl0} through @code{$dl15}, and are formed
24569 by joining the even/odd register pairs @code{f0} and @code{f1} for @code{$dl0},
24570 @code{f2} and @code{f3} for @code{$dl1} and so on.
24572 For POWER7 processors, @value{GDBN} provides a set of pseudo-registers, the 64-bit
24573 wide Extended Floating Point Registers (@samp{f32} through @samp{f63}).
24576 @subsection Nios II
24577 @cindex Nios II architecture
24579 When @value{GDBN} is debugging the Nios II architecture,
24580 it provides the following special commands:
24584 @item set debug nios2
24585 @kindex set debug nios2
24586 This command turns on and off debugging messages for the Nios II
24587 target code in @value{GDBN}.
24589 @item show debug nios2
24590 @kindex show debug nios2
24591 Show the current setting of Nios II debugging messages.
24595 @subsection Sparc64
24596 @cindex Sparc64 support
24597 @cindex Application Data Integrity
24598 @subsubsection ADI Support
24600 The M7 processor supports an Application Data Integrity (ADI) feature that
24601 detects invalid data accesses. When software allocates memory and enables
24602 ADI on the allocated memory, it chooses a 4-bit version number, sets the
24603 version in the upper 4 bits of the 64-bit pointer to that data, and stores
24604 the 4-bit version in every cacheline of that data. Hardware saves the latter
24605 in spare bits in the cache and memory hierarchy. On each load and store,
24606 the processor compares the upper 4 VA (virtual address) bits to the
24607 cacheline's version. If there is a mismatch, the processor generates a
24608 version mismatch trap which can be either precise or disrupting. The trap
24609 is an error condition which the kernel delivers to the process as a SIGSEGV
24612 Note that only 64-bit applications can use ADI and need to be built with
24615 Values of the ADI version tags, which are in granularity of a
24616 cacheline (64 bytes), can be viewed or modified.
24620 @kindex adi examine
24621 @item adi (examine | x) [ / @var{n} ] @var{addr}
24623 The @code{adi examine} command displays the value of one ADI version tag per
24626 @var{n} is a decimal integer specifying the number in bytes; the default
24627 is 1. It specifies how much ADI version information, at the ratio of 1:ADI
24628 block size, to display.
24630 @var{addr} is the address in user address space where you want @value{GDBN}
24631 to begin displaying the ADI version tags.
24633 Below is an example of displaying ADI versions of variable "shmaddr".
24636 (@value{GDBP}) adi x/100 shmaddr
24637 0xfff800010002c000: 0 0
24641 @item adi (assign | a) [ / @var{n} ] @var{addr} = @var{tag}
24643 The @code{adi assign} command is used to assign new ADI version tag
24646 @var{n} is a decimal integer specifying the number in bytes;
24647 the default is 1. It specifies how much ADI version information, at the
24648 ratio of 1:ADI block size, to modify.
24650 @var{addr} is the address in user address space where you want @value{GDBN}
24651 to begin modifying the ADI version tags.
24653 @var{tag} is the new ADI version tag.
24655 For example, do the following to modify then verify ADI versions of
24656 variable "shmaddr":
24659 (@value{GDBP}) adi a/100 shmaddr = 7
24660 (@value{GDBP}) adi x/100 shmaddr
24661 0xfff800010002c000: 7 7
24668 @cindex S12Z support
24670 When @value{GDBN} is debugging the S12Z architecture,
24671 it provides the following special command:
24674 @item maint info bdccsr
24675 @kindex maint info bdccsr@r{, S12Z}
24676 This command displays the current value of the microprocessor's
24681 @node Controlling GDB
24682 @chapter Controlling @value{GDBN}
24684 You can alter the way @value{GDBN} interacts with you by using the
24685 @code{set} command. For commands controlling how @value{GDBN} displays
24686 data, see @ref{Print Settings, ,Print Settings}. Other settings are
24691 * Editing:: Command editing
24692 * Command History:: Command history
24693 * Screen Size:: Screen size
24694 * Output Styling:: Output styling
24695 * Numbers:: Numbers
24696 * ABI:: Configuring the current ABI
24697 * Auto-loading:: Automatically loading associated files
24698 * Messages/Warnings:: Optional warnings and messages
24699 * Debugging Output:: Optional messages about internal happenings
24700 * Other Misc Settings:: Other Miscellaneous Settings
24708 @value{GDBN} indicates its readiness to read a command by printing a string
24709 called the @dfn{prompt}. This string is normally @samp{(@value{GDBP})}. You
24710 can change the prompt string with the @code{set prompt} command. For
24711 instance, when debugging @value{GDBN} with @value{GDBN}, it is useful to change
24712 the prompt in one of the @value{GDBN} sessions so that you can always tell
24713 which one you are talking to.
24715 @emph{Note:} @code{set prompt} does not add a space for you after the
24716 prompt you set. This allows you to set a prompt which ends in a space
24717 or a prompt that does not.
24721 @item set prompt @var{newprompt}
24722 Directs @value{GDBN} to use @var{newprompt} as its prompt string henceforth.
24724 @kindex show prompt
24726 Prints a line of the form: @samp{Gdb's prompt is: @var{your-prompt}}
24729 Versions of @value{GDBN} that ship with Python scripting enabled have
24730 prompt extensions. The commands for interacting with these extensions
24734 @kindex set extended-prompt
24735 @item set extended-prompt @var{prompt}
24736 Set an extended prompt that allows for substitutions.
24737 @xref{gdb.prompt}, for a list of escape sequences that can be used for
24738 substitution. Any escape sequences specified as part of the prompt
24739 string are replaced with the corresponding strings each time the prompt
24745 set extended-prompt Current working directory: \w (gdb)
24748 Note that when an extended-prompt is set, it takes control of the
24749 @var{prompt_hook} hook. @xref{prompt_hook}, for further information.
24751 @kindex show extended-prompt
24752 @item show extended-prompt
24753 Prints the extended prompt. Any escape sequences specified as part of
24754 the prompt string with @code{set extended-prompt}, are replaced with the
24755 corresponding strings each time the prompt is displayed.
24759 @section Command Editing
24761 @cindex command line editing
24763 @value{GDBN} reads its input commands via the @dfn{Readline} interface. This
24764 @sc{gnu} library provides consistent behavior for programs which provide a
24765 command line interface to the user. Advantages are @sc{gnu} Emacs-style
24766 or @dfn{vi}-style inline editing of commands, @code{csh}-like history
24767 substitution, and a storage and recall of command history across
24768 debugging sessions.
24770 You may control the behavior of command line editing in @value{GDBN} with the
24771 command @code{set}.
24774 @kindex set editing
24777 @itemx set editing on
24778 Enable command line editing (enabled by default).
24780 @item set editing off
24781 Disable command line editing.
24783 @kindex show editing
24785 Show whether command line editing is enabled.
24788 @ifset SYSTEM_READLINE
24789 @xref{Command Line Editing, , , rluserman, GNU Readline Library},
24791 @ifclear SYSTEM_READLINE
24792 @xref{Command Line Editing},
24794 for more details about the Readline
24795 interface. Users unfamiliar with @sc{gnu} Emacs or @code{vi} are
24796 encouraged to read that chapter.
24798 @node Command History
24799 @section Command History
24800 @cindex command history
24802 @value{GDBN} can keep track of the commands you type during your
24803 debugging sessions, so that you can be certain of precisely what
24804 happened. Use these commands to manage the @value{GDBN} command
24807 @value{GDBN} uses the @sc{gnu} History library, a part of the Readline
24808 package, to provide the history facility.
24809 @ifset SYSTEM_READLINE
24810 @xref{Using History Interactively, , , history, GNU History Library},
24812 @ifclear SYSTEM_READLINE
24813 @xref{Using History Interactively},
24815 for the detailed description of the History library.
24817 To issue a command to @value{GDBN} without affecting certain aspects of
24818 the state which is seen by users, prefix it with @samp{server }
24819 (@pxref{Server Prefix}). This
24820 means that this command will not affect the command history, nor will it
24821 affect @value{GDBN}'s notion of which command to repeat if @key{RET} is
24822 pressed on a line by itself.
24824 @cindex @code{server}, command prefix
24825 The server prefix does not affect the recording of values into the value
24826 history; to print a value without recording it into the value history,
24827 use the @code{output} command instead of the @code{print} command.
24829 Here is the description of @value{GDBN} commands related to command
24833 @cindex history substitution
24834 @cindex history file
24835 @kindex set history filename
24836 @cindex @env{GDBHISTFILE}, environment variable
24837 @item set history filename @var{fname}
24838 Set the name of the @value{GDBN} command history file to @var{fname}.
24839 This is the file where @value{GDBN} reads an initial command history
24840 list, and where it writes the command history from this session when it
24841 exits. You can access this list through history expansion or through
24842 the history command editing characters listed below. This file defaults
24843 to the value of the environment variable @code{GDBHISTFILE}, or to
24844 @file{./.gdb_history} (@file{./_gdb_history} on MS-DOS) if this variable
24847 @cindex save command history
24848 @kindex set history save
24849 @item set history save
24850 @itemx set history save on
24851 Record command history in a file, whose name may be specified with the
24852 @code{set history filename} command. By default, this option is disabled.
24854 @item set history save off
24855 Stop recording command history in a file.
24857 @cindex history size
24858 @kindex set history size
24859 @cindex @env{GDBHISTSIZE}, environment variable
24860 @item set history size @var{size}
24861 @itemx set history size unlimited
24862 Set the number of commands which @value{GDBN} keeps in its history list.
24863 This defaults to the value of the environment variable @env{GDBHISTSIZE}, or
24864 to 256 if this variable is not set. Non-numeric values of @env{GDBHISTSIZE}
24865 are ignored. If @var{size} is @code{unlimited} or if @env{GDBHISTSIZE} is
24866 either a negative number or the empty string, then the number of commands
24867 @value{GDBN} keeps in the history list is unlimited.
24869 @cindex remove duplicate history
24870 @kindex set history remove-duplicates
24871 @item set history remove-duplicates @var{count}
24872 @itemx set history remove-duplicates unlimited
24873 Control the removal of duplicate history entries in the command history list.
24874 If @var{count} is non-zero, @value{GDBN} will look back at the last @var{count}
24875 history entries and remove the first entry that is a duplicate of the current
24876 entry being added to the command history list. If @var{count} is
24877 @code{unlimited} then this lookbehind is unbounded. If @var{count} is 0, then
24878 removal of duplicate history entries is disabled.
24880 Only history entries added during the current session are considered for
24881 removal. This option is set to 0 by default.
24885 History expansion assigns special meaning to the character @kbd{!}.
24886 @ifset SYSTEM_READLINE
24887 @xref{Event Designators, , , history, GNU History Library},
24889 @ifclear SYSTEM_READLINE
24890 @xref{Event Designators},
24894 @cindex history expansion, turn on/off
24895 Since @kbd{!} is also the logical not operator in C, history expansion
24896 is off by default. If you decide to enable history expansion with the
24897 @code{set history expansion on} command, you may sometimes need to
24898 follow @kbd{!} (when it is used as logical not, in an expression) with
24899 a space or a tab to prevent it from being expanded. The readline
24900 history facilities do not attempt substitution on the strings
24901 @kbd{!=} and @kbd{!(}, even when history expansion is enabled.
24903 The commands to control history expansion are:
24906 @item set history expansion on
24907 @itemx set history expansion
24908 @kindex set history expansion
24909 Enable history expansion. History expansion is off by default.
24911 @item set history expansion off
24912 Disable history expansion.
24915 @kindex show history
24917 @itemx show history filename
24918 @itemx show history save
24919 @itemx show history size
24920 @itemx show history expansion
24921 These commands display the state of the @value{GDBN} history parameters.
24922 @code{show history} by itself displays all four states.
24927 @kindex show commands
24928 @cindex show last commands
24929 @cindex display command history
24930 @item show commands
24931 Display the last ten commands in the command history.
24933 @item show commands @var{n}
24934 Print ten commands centered on command number @var{n}.
24936 @item show commands +
24937 Print ten commands just after the commands last printed.
24941 @section Screen Size
24942 @cindex size of screen
24943 @cindex screen size
24946 @cindex pauses in output
24948 Certain commands to @value{GDBN} may produce large amounts of
24949 information output to the screen. To help you read all of it,
24950 @value{GDBN} pauses and asks you for input at the end of each page of
24951 output. Type @key{RET} when you want to see one more page of output,
24952 @kbd{q} to discard the remaining output, or @kbd{c} to continue
24953 without paging for the rest of the current command. Also, the screen
24954 width setting determines when to wrap lines of output. Depending on
24955 what is being printed, @value{GDBN} tries to break the line at a
24956 readable place, rather than simply letting it overflow onto the
24959 Normally @value{GDBN} knows the size of the screen from the terminal
24960 driver software. For example, on Unix @value{GDBN} uses the termcap data base
24961 together with the value of the @code{TERM} environment variable and the
24962 @code{stty rows} and @code{stty cols} settings. If this is not correct,
24963 you can override it with the @code{set height} and @code{set
24970 @kindex show height
24971 @item set height @var{lpp}
24972 @itemx set height unlimited
24974 @itemx set width @var{cpl}
24975 @itemx set width unlimited
24977 These @code{set} commands specify a screen height of @var{lpp} lines and
24978 a screen width of @var{cpl} characters. The associated @code{show}
24979 commands display the current settings.
24981 If you specify a height of either @code{unlimited} or zero lines,
24982 @value{GDBN} does not pause during output no matter how long the
24983 output is. This is useful if output is to a file or to an editor
24986 Likewise, you can specify @samp{set width unlimited} or @samp{set
24987 width 0} to prevent @value{GDBN} from wrapping its output.
24989 @item set pagination on
24990 @itemx set pagination off
24991 @kindex set pagination
24992 Turn the output pagination on or off; the default is on. Turning
24993 pagination off is the alternative to @code{set height unlimited}. Note that
24994 running @value{GDBN} with the @option{--batch} option (@pxref{Mode
24995 Options, -batch}) also automatically disables pagination.
24997 @item show pagination
24998 @kindex show pagination
24999 Show the current pagination mode.
25002 @node Output Styling
25003 @section Output Styling
25009 @value{GDBN} can style its output on a capable terminal. This is
25010 enabled by default on most systems, but disabled by default when in
25011 batch mode (@pxref{Mode Options}). Various style settings are available;
25012 and styles can also be disabled entirely.
25015 @item set style enabled @samp{on|off}
25016 Enable or disable all styling. The default is host-dependent, with
25017 most hosts defaulting to @samp{on}.
25019 @item show style enabled
25020 Show the current state of styling.
25022 @item set style sources @samp{on|off}
25023 Enable or disable source code styling. This affects whether source
25024 code, such as the output of the @code{list} command, is styled. Note
25025 that source styling only works if styling in general is enabled, and
25026 if @value{GDBN} was linked with the GNU Source Highlight library. The
25027 default is @samp{on}.
25029 @item show style sources
25030 Show the current state of source code styling.
25033 Subcommands of @code{set style} control specific forms of styling.
25034 These subcommands all follow the same pattern: each style-able object
25035 can be styled with a foreground color, a background color, and an
25038 For example, the style of file names can be controlled using the
25039 @code{set style filename} group of commands:
25042 @item set style filename background @var{color}
25043 Set the background to @var{color}. Valid colors are @samp{none}
25044 (meaning the terminal's default color), @samp{black}, @samp{red},
25045 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
25048 @item set style filename foreground @var{color}
25049 Set the foreground to @var{color}. Valid colors are @samp{none}
25050 (meaning the terminal's default color), @samp{black}, @samp{red},
25051 @samp{green}, @samp{yellow}, @samp{blue}, @samp{magenta}, @samp{cyan},
25054 @item set style filename intensity @var{value}
25055 Set the intensity to @var{value}. Valid intensities are @samp{normal}
25056 (the default), @samp{bold}, and @samp{dim}.
25059 The @code{show style} command and its subcommands are styling
25060 a style name in their output using its own style.
25061 So, use @command{show style} to see the complete list of styles,
25062 their characteristics and the visual aspect of each style.
25064 The style-able objects are:
25067 Control the styling of file names. By default, this style's
25068 foreground color is green.
25071 Control the styling of function names. These are managed with the
25072 @code{set style function} family of commands. By default, this
25073 style's foreground color is yellow.
25076 Control the styling of variable names. These are managed with the
25077 @code{set style variable} family of commands. By default, this style's
25078 foreground color is cyan.
25081 Control the styling of addresses. These are managed with the
25082 @code{set style address} family of commands. By default, this style's
25083 foreground color is blue.
25086 Control the styling of titles. These are managed with the
25087 @code{set style title} family of commands. By default, this style's
25088 intensity is bold. Commands are using the title style to improve
25089 the readibility of large output. For example, the commands
25090 @command{apropos} and @command{help} are using the title style
25091 for the command names.
25094 Control the styling of highlightings. These are managed with the
25095 @code{set style highlight} family of commands. By default, this style's
25096 foreground color is red. Commands are using the highlight style to draw
25097 the user attention to some specific parts of their output. For example,
25098 the command @command{apropos -v REGEXP} uses the highlight style to
25099 mark the documentation parts matching @var{regexp}.
25105 @cindex number representation
25106 @cindex entering numbers
25108 You can always enter numbers in octal, decimal, or hexadecimal in
25109 @value{GDBN} by the usual conventions: octal numbers begin with
25110 @samp{0}, decimal numbers end with @samp{.}, and hexadecimal numbers
25111 begin with @samp{0x}. Numbers that neither begin with @samp{0} or
25112 @samp{0x}, nor end with a @samp{.} are, by default, entered in base
25113 10; likewise, the default display for numbers---when no particular
25114 format is specified---is base 10. You can change the default base for
25115 both input and output with the commands described below.
25118 @kindex set input-radix
25119 @item set input-radix @var{base}
25120 Set the default base for numeric input. Supported choices
25121 for @var{base} are decimal 8, 10, or 16. The base must itself be
25122 specified either unambiguously or using the current input radix; for
25126 set input-radix 012
25127 set input-radix 10.
25128 set input-radix 0xa
25132 sets the input base to decimal. On the other hand, @samp{set input-radix 10}
25133 leaves the input radix unchanged, no matter what it was, since
25134 @samp{10}, being without any leading or trailing signs of its base, is
25135 interpreted in the current radix. Thus, if the current radix is 16,
25136 @samp{10} is interpreted in hex, i.e.@: as 16 decimal, which doesn't
25139 @kindex set output-radix
25140 @item set output-radix @var{base}
25141 Set the default base for numeric display. Supported choices
25142 for @var{base} are decimal 8, 10, or 16. The base must itself be
25143 specified either unambiguously or using the current input radix.
25145 @kindex show input-radix
25146 @item show input-radix
25147 Display the current default base for numeric input.
25149 @kindex show output-radix
25150 @item show output-radix
25151 Display the current default base for numeric display.
25153 @item set radix @r{[}@var{base}@r{]}
25157 These commands set and show the default base for both input and output
25158 of numbers. @code{set radix} sets the radix of input and output to
25159 the same base; without an argument, it resets the radix back to its
25160 default value of 10.
25165 @section Configuring the Current ABI
25167 @value{GDBN} can determine the @dfn{ABI} (Application Binary Interface) of your
25168 application automatically. However, sometimes you need to override its
25169 conclusions. Use these commands to manage @value{GDBN}'s view of the
25175 @cindex Newlib OS ABI and its influence on the longjmp handling
25177 One @value{GDBN} configuration can debug binaries for multiple operating
25178 system targets, either via remote debugging or native emulation.
25179 @value{GDBN} will autodetect the @dfn{OS ABI} (Operating System ABI) in use,
25180 but you can override its conclusion using the @code{set osabi} command.
25181 One example where this is useful is in debugging of binaries which use
25182 an alternate C library (e.g.@: @sc{uClibc} for @sc{gnu}/Linux) which does
25183 not have the same identifying marks that the standard C library for your
25186 When @value{GDBN} is debugging the AArch64 architecture, it provides a
25187 ``Newlib'' OS ABI. This is useful for handling @code{setjmp} and
25188 @code{longjmp} when debugging binaries that use the @sc{newlib} C library.
25189 The ``Newlib'' OS ABI can be selected by @code{set osabi Newlib}.
25193 Show the OS ABI currently in use.
25196 With no argument, show the list of registered available OS ABI's.
25198 @item set osabi @var{abi}
25199 Set the current OS ABI to @var{abi}.
25202 @cindex float promotion
25204 Generally, the way that an argument of type @code{float} is passed to a
25205 function depends on whether the function is prototyped. For a prototyped
25206 (i.e.@: ANSI/ISO style) function, @code{float} arguments are passed unchanged,
25207 according to the architecture's convention for @code{float}. For unprototyped
25208 (i.e.@: K&R style) functions, @code{float} arguments are first promoted to type
25209 @code{double} and then passed.
25211 Unfortunately, some forms of debug information do not reliably indicate whether
25212 a function is prototyped. If @value{GDBN} calls a function that is not marked
25213 as prototyped, it consults @kbd{set coerce-float-to-double}.
25216 @kindex set coerce-float-to-double
25217 @item set coerce-float-to-double
25218 @itemx set coerce-float-to-double on
25219 Arguments of type @code{float} will be promoted to @code{double} when passed
25220 to an unprototyped function. This is the default setting.
25222 @item set coerce-float-to-double off
25223 Arguments of type @code{float} will be passed directly to unprototyped
25226 @kindex show coerce-float-to-double
25227 @item show coerce-float-to-double
25228 Show the current setting of promoting @code{float} to @code{double}.
25232 @kindex show cp-abi
25233 @value{GDBN} needs to know the ABI used for your program's C@t{++}
25234 objects. The correct C@t{++} ABI depends on which C@t{++} compiler was
25235 used to build your application. @value{GDBN} only fully supports
25236 programs with a single C@t{++} ABI; if your program contains code using
25237 multiple C@t{++} ABI's or if @value{GDBN} can not identify your
25238 program's ABI correctly, you can tell @value{GDBN} which ABI to use.
25239 Currently supported ABI's include ``gnu-v2'', for @code{g++} versions
25240 before 3.0, ``gnu-v3'', for @code{g++} versions 3.0 and later, and
25241 ``hpaCC'' for the HP ANSI C@t{++} compiler. Other C@t{++} compilers may
25242 use the ``gnu-v2'' or ``gnu-v3'' ABI's as well. The default setting is
25247 Show the C@t{++} ABI currently in use.
25250 With no argument, show the list of supported C@t{++} ABI's.
25252 @item set cp-abi @var{abi}
25253 @itemx set cp-abi auto
25254 Set the current C@t{++} ABI to @var{abi}, or return to automatic detection.
25258 @section Automatically loading associated files
25259 @cindex auto-loading
25261 @value{GDBN} sometimes reads files with commands and settings automatically,
25262 without being explicitly told so by the user. We call this feature
25263 @dfn{auto-loading}. While auto-loading is useful for automatically adapting
25264 @value{GDBN} to the needs of your project, it can sometimes produce unexpected
25265 results or introduce security risks (e.g., if the file comes from untrusted
25269 * Init File in the Current Directory:: @samp{set/show/info auto-load local-gdbinit}
25270 * libthread_db.so.1 file:: @samp{set/show/info auto-load libthread-db}
25272 * Auto-loading safe path:: @samp{set/show/info auto-load safe-path}
25273 * Auto-loading verbose mode:: @samp{set/show debug auto-load}
25276 There are various kinds of files @value{GDBN} can automatically load.
25277 In addition to these files, @value{GDBN} supports auto-loading code written
25278 in various extension languages. @xref{Auto-loading extensions}.
25280 Note that loading of these associated files (including the local @file{.gdbinit}
25281 file) requires accordingly configured @code{auto-load safe-path}
25282 (@pxref{Auto-loading safe path}).
25284 For these reasons, @value{GDBN} includes commands and options to let you
25285 control when to auto-load files and which files should be auto-loaded.
25288 @anchor{set auto-load off}
25289 @kindex set auto-load off
25290 @item set auto-load off
25291 Globally disable loading of all auto-loaded files.
25292 You may want to use this command with the @samp{-iex} option
25293 (@pxref{Option -init-eval-command}) such as:
25295 $ @kbd{gdb -iex "set auto-load off" untrusted-executable corefile}
25298 Be aware that system init file (@pxref{System-wide configuration})
25299 and init files from your home directory (@pxref{Home Directory Init File})
25300 still get read (as they come from generally trusted directories).
25301 To prevent @value{GDBN} from auto-loading even those init files, use the
25302 @option{-nx} option (@pxref{Mode Options}), in addition to
25303 @code{set auto-load no}.
25305 @anchor{show auto-load}
25306 @kindex show auto-load
25307 @item show auto-load
25308 Show whether auto-loading of each specific @samp{auto-load} file(s) is enabled
25312 (gdb) show auto-load
25313 gdb-scripts: Auto-loading of canned sequences of commands scripts is on.
25314 libthread-db: Auto-loading of inferior specific libthread_db is on.
25315 local-gdbinit: Auto-loading of .gdbinit script from current directory
25317 python-scripts: Auto-loading of Python scripts is on.
25318 safe-path: List of directories from which it is safe to auto-load files
25319 is $debugdir:$datadir/auto-load.
25320 scripts-directory: List of directories from which to load auto-loaded scripts
25321 is $debugdir:$datadir/auto-load.
25324 @anchor{info auto-load}
25325 @kindex info auto-load
25326 @item info auto-load
25327 Print whether each specific @samp{auto-load} file(s) have been auto-loaded or
25331 (gdb) info auto-load
25334 Yes /home/user/gdb/gdb-gdb.gdb
25335 libthread-db: No auto-loaded libthread-db.
25336 local-gdbinit: Local .gdbinit file "/home/user/gdb/.gdbinit" has been
25340 Yes /home/user/gdb/gdb-gdb.py
25344 These are @value{GDBN} control commands for the auto-loading:
25346 @multitable @columnfractions .5 .5
25347 @item @xref{set auto-load off}.
25348 @tab Disable auto-loading globally.
25349 @item @xref{show auto-load}.
25350 @tab Show setting of all kinds of files.
25351 @item @xref{info auto-load}.
25352 @tab Show state of all kinds of files.
25353 @item @xref{set auto-load gdb-scripts}.
25354 @tab Control for @value{GDBN} command scripts.
25355 @item @xref{show auto-load gdb-scripts}.
25356 @tab Show setting of @value{GDBN} command scripts.
25357 @item @xref{info auto-load gdb-scripts}.
25358 @tab Show state of @value{GDBN} command scripts.
25359 @item @xref{set auto-load python-scripts}.
25360 @tab Control for @value{GDBN} Python scripts.
25361 @item @xref{show auto-load python-scripts}.
25362 @tab Show setting of @value{GDBN} Python scripts.
25363 @item @xref{info auto-load python-scripts}.
25364 @tab Show state of @value{GDBN} Python scripts.
25365 @item @xref{set auto-load guile-scripts}.
25366 @tab Control for @value{GDBN} Guile scripts.
25367 @item @xref{show auto-load guile-scripts}.
25368 @tab Show setting of @value{GDBN} Guile scripts.
25369 @item @xref{info auto-load guile-scripts}.
25370 @tab Show state of @value{GDBN} Guile scripts.
25371 @item @xref{set auto-load scripts-directory}.
25372 @tab Control for @value{GDBN} auto-loaded scripts location.
25373 @item @xref{show auto-load scripts-directory}.
25374 @tab Show @value{GDBN} auto-loaded scripts location.
25375 @item @xref{add-auto-load-scripts-directory}.
25376 @tab Add directory for auto-loaded scripts location list.
25377 @item @xref{set auto-load local-gdbinit}.
25378 @tab Control for init file in the current directory.
25379 @item @xref{show auto-load local-gdbinit}.
25380 @tab Show setting of init file in the current directory.
25381 @item @xref{info auto-load local-gdbinit}.
25382 @tab Show state of init file in the current directory.
25383 @item @xref{set auto-load libthread-db}.
25384 @tab Control for thread debugging library.
25385 @item @xref{show auto-load libthread-db}.
25386 @tab Show setting of thread debugging library.
25387 @item @xref{info auto-load libthread-db}.
25388 @tab Show state of thread debugging library.
25389 @item @xref{set auto-load safe-path}.
25390 @tab Control directories trusted for automatic loading.
25391 @item @xref{show auto-load safe-path}.
25392 @tab Show directories trusted for automatic loading.
25393 @item @xref{add-auto-load-safe-path}.
25394 @tab Add directory trusted for automatic loading.
25397 @node Init File in the Current Directory
25398 @subsection Automatically loading init file in the current directory
25399 @cindex auto-loading init file in the current directory
25401 By default, @value{GDBN} reads and executes the canned sequences of commands
25402 from init file (if any) in the current working directory,
25403 see @ref{Init File in the Current Directory during Startup}.
25405 Note that loading of this local @file{.gdbinit} file also requires accordingly
25406 configured @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25409 @anchor{set auto-load local-gdbinit}
25410 @kindex set auto-load local-gdbinit
25411 @item set auto-load local-gdbinit [on|off]
25412 Enable or disable the auto-loading of canned sequences of commands
25413 (@pxref{Sequences}) found in init file in the current directory.
25415 @anchor{show auto-load local-gdbinit}
25416 @kindex show auto-load local-gdbinit
25417 @item show auto-load local-gdbinit
25418 Show whether auto-loading of canned sequences of commands from init file in the
25419 current directory is enabled or disabled.
25421 @anchor{info auto-load local-gdbinit}
25422 @kindex info auto-load local-gdbinit
25423 @item info auto-load local-gdbinit
25424 Print whether canned sequences of commands from init file in the
25425 current directory have been auto-loaded.
25428 @node libthread_db.so.1 file
25429 @subsection Automatically loading thread debugging library
25430 @cindex auto-loading libthread_db.so.1
25432 This feature is currently present only on @sc{gnu}/Linux native hosts.
25434 @value{GDBN} reads in some cases thread debugging library from places specific
25435 to the inferior (@pxref{set libthread-db-search-path}).
25437 The special @samp{libthread-db-search-path} entry @samp{$sdir} is processed
25438 without checking this @samp{set auto-load libthread-db} switch as system
25439 libraries have to be trusted in general. In all other cases of
25440 @samp{libthread-db-search-path} entries @value{GDBN} checks first if @samp{set
25441 auto-load libthread-db} is enabled before trying to open such thread debugging
25444 Note that loading of this debugging library also requires accordingly configured
25445 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
25448 @anchor{set auto-load libthread-db}
25449 @kindex set auto-load libthread-db
25450 @item set auto-load libthread-db [on|off]
25451 Enable or disable the auto-loading of inferior specific thread debugging library.
25453 @anchor{show auto-load libthread-db}
25454 @kindex show auto-load libthread-db
25455 @item show auto-load libthread-db
25456 Show whether auto-loading of inferior specific thread debugging library is
25457 enabled or disabled.
25459 @anchor{info auto-load libthread-db}
25460 @kindex info auto-load libthread-db
25461 @item info auto-load libthread-db
25462 Print the list of all loaded inferior specific thread debugging libraries and
25463 for each such library print list of inferior @var{pid}s using it.
25466 @node Auto-loading safe path
25467 @subsection Security restriction for auto-loading
25468 @cindex auto-loading safe-path
25470 As the files of inferior can come from untrusted source (such as submitted by
25471 an application user) @value{GDBN} does not always load any files automatically.
25472 @value{GDBN} provides the @samp{set auto-load safe-path} setting to list
25473 directories trusted for loading files not explicitly requested by user.
25474 Each directory can also be a shell wildcard pattern.
25476 If the path is not set properly you will see a warning and the file will not
25481 Reading symbols from /home/user/gdb/gdb...done.
25482 warning: File "/home/user/gdb/gdb-gdb.gdb" auto-loading has been
25483 declined by your `auto-load safe-path' set
25484 to "$debugdir:$datadir/auto-load".
25485 warning: File "/home/user/gdb/gdb-gdb.py" auto-loading has been
25486 declined by your `auto-load safe-path' set
25487 to "$debugdir:$datadir/auto-load".
25491 To instruct @value{GDBN} to go ahead and use the init files anyway,
25492 invoke @value{GDBN} like this:
25495 $ gdb -q -iex "set auto-load safe-path /home/user/gdb" ./gdb
25498 The list of trusted directories is controlled by the following commands:
25501 @anchor{set auto-load safe-path}
25502 @kindex set auto-load safe-path
25503 @item set auto-load safe-path @r{[}@var{directories}@r{]}
25504 Set the list of directories (and their subdirectories) trusted for automatic
25505 loading and execution of scripts. You can also enter a specific trusted file.
25506 Each directory can also be a shell wildcard pattern; wildcards do not match
25507 directory separator - see @code{FNM_PATHNAME} for system function @code{fnmatch}
25508 (@pxref{Wildcard Matching, fnmatch, , libc, GNU C Library Reference Manual}).
25509 If you omit @var{directories}, @samp{auto-load safe-path} will be reset to
25510 its default value as specified during @value{GDBN} compilation.
25512 The list of directories uses path separator (@samp{:} on GNU and Unix
25513 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
25514 to the @env{PATH} environment variable.
25516 @anchor{show auto-load safe-path}
25517 @kindex show auto-load safe-path
25518 @item show auto-load safe-path
25519 Show the list of directories trusted for automatic loading and execution of
25522 @anchor{add-auto-load-safe-path}
25523 @kindex add-auto-load-safe-path
25524 @item add-auto-load-safe-path
25525 Add an entry (or list of entries) to the list of directories trusted for
25526 automatic loading and execution of scripts. Multiple entries may be delimited
25527 by the host platform path separator in use.
25530 This variable defaults to what @code{--with-auto-load-dir} has been configured
25531 to (@pxref{with-auto-load-dir}). @file{$debugdir} and @file{$datadir}
25532 substitution applies the same as for @ref{set auto-load scripts-directory}.
25533 The default @code{set auto-load safe-path} value can be also overriden by
25534 @value{GDBN} configuration option @option{--with-auto-load-safe-path}.
25536 Setting this variable to @file{/} disables this security protection,
25537 corresponding @value{GDBN} configuration option is
25538 @option{--without-auto-load-safe-path}.
25539 This variable is supposed to be set to the system directories writable by the
25540 system superuser only. Users can add their source directories in init files in
25541 their home directories (@pxref{Home Directory Init File}). See also deprecated
25542 init file in the current directory
25543 (@pxref{Init File in the Current Directory during Startup}).
25545 To force @value{GDBN} to load the files it declined to load in the previous
25546 example, you could use one of the following ways:
25549 @item @file{~/.gdbinit}: @samp{add-auto-load-safe-path ~/src/gdb}
25550 Specify this trusted directory (or a file) as additional component of the list.
25551 You have to specify also any existing directories displayed by
25552 by @samp{show auto-load safe-path} (such as @samp{/usr:/bin} in this example).
25554 @item @kbd{gdb -iex "set auto-load safe-path /usr:/bin:~/src/gdb" @dots{}}
25555 Specify this directory as in the previous case but just for a single
25556 @value{GDBN} session.
25558 @item @kbd{gdb -iex "set auto-load safe-path /" @dots{}}
25559 Disable auto-loading safety for a single @value{GDBN} session.
25560 This assumes all the files you debug during this @value{GDBN} session will come
25561 from trusted sources.
25563 @item @kbd{./configure --without-auto-load-safe-path}
25564 During compilation of @value{GDBN} you may disable any auto-loading safety.
25565 This assumes all the files you will ever debug with this @value{GDBN} come from
25569 On the other hand you can also explicitly forbid automatic files loading which
25570 also suppresses any such warning messages:
25573 @item @kbd{gdb -iex "set auto-load no" @dots{}}
25574 You can use @value{GDBN} command-line option for a single @value{GDBN} session.
25576 @item @file{~/.gdbinit}: @samp{set auto-load no}
25577 Disable auto-loading globally for the user
25578 (@pxref{Home Directory Init File}). While it is improbable, you could also
25579 use system init file instead (@pxref{System-wide configuration}).
25582 This setting applies to the file names as entered by user. If no entry matches
25583 @value{GDBN} tries as a last resort to also resolve all the file names into
25584 their canonical form (typically resolving symbolic links) and compare the
25585 entries again. @value{GDBN} already canonicalizes most of the filenames on its
25586 own before starting the comparison so a canonical form of directories is
25587 recommended to be entered.
25589 @node Auto-loading verbose mode
25590 @subsection Displaying files tried for auto-load
25591 @cindex auto-loading verbose mode
25593 For better visibility of all the file locations where you can place scripts to
25594 be auto-loaded with inferior --- or to protect yourself against accidental
25595 execution of untrusted scripts --- @value{GDBN} provides a feature for printing
25596 all the files attempted to be loaded. Both existing and non-existing files may
25599 For example the list of directories from which it is safe to auto-load files
25600 (@pxref{Auto-loading safe path}) applies also to canonicalized filenames which
25601 may not be too obvious while setting it up.
25604 (gdb) set debug auto-load on
25605 (gdb) file ~/src/t/true
25606 auto-load: Loading canned sequences of commands script "/tmp/true-gdb.gdb"
25607 for objfile "/tmp/true".
25608 auto-load: Updating directories of "/usr:/opt".
25609 auto-load: Using directory "/usr".
25610 auto-load: Using directory "/opt".
25611 warning: File "/tmp/true-gdb.gdb" auto-loading has been declined
25612 by your `auto-load safe-path' set to "/usr:/opt".
25616 @anchor{set debug auto-load}
25617 @kindex set debug auto-load
25618 @item set debug auto-load [on|off]
25619 Set whether to print the filenames attempted to be auto-loaded.
25621 @anchor{show debug auto-load}
25622 @kindex show debug auto-load
25623 @item show debug auto-load
25624 Show whether printing of the filenames attempted to be auto-loaded is turned
25628 @node Messages/Warnings
25629 @section Optional Warnings and Messages
25631 @cindex verbose operation
25632 @cindex optional warnings
25633 By default, @value{GDBN} is silent about its inner workings. If you are
25634 running on a slow machine, you may want to use the @code{set verbose}
25635 command. This makes @value{GDBN} tell you when it does a lengthy
25636 internal operation, so you will not think it has crashed.
25638 Currently, the messages controlled by @code{set verbose} are those
25639 which announce that the symbol table for a source file is being read;
25640 see @code{symbol-file} in @ref{Files, ,Commands to Specify Files}.
25643 @kindex set verbose
25644 @item set verbose on
25645 Enables @value{GDBN} output of certain informational messages.
25647 @item set verbose off
25648 Disables @value{GDBN} output of certain informational messages.
25650 @kindex show verbose
25652 Displays whether @code{set verbose} is on or off.
25655 By default, if @value{GDBN} encounters bugs in the symbol table of an
25656 object file, it is silent; but if you are debugging a compiler, you may
25657 find this information useful (@pxref{Symbol Errors, ,Errors Reading
25662 @kindex set complaints
25663 @item set complaints @var{limit}
25664 Permits @value{GDBN} to output @var{limit} complaints about each type of
25665 unusual symbols before becoming silent about the problem. Set
25666 @var{limit} to zero to suppress all complaints; set it to a large number
25667 to prevent complaints from being suppressed.
25669 @kindex show complaints
25670 @item show complaints
25671 Displays how many symbol complaints @value{GDBN} is permitted to produce.
25675 @anchor{confirmation requests}
25676 By default, @value{GDBN} is cautious, and asks what sometimes seems to be a
25677 lot of stupid questions to confirm certain commands. For example, if
25678 you try to run a program which is already running:
25682 The program being debugged has been started already.
25683 Start it from the beginning? (y or n)
25686 If you are willing to unflinchingly face the consequences of your own
25687 commands, you can disable this ``feature'':
25691 @kindex set confirm
25693 @cindex confirmation
25694 @cindex stupid questions
25695 @item set confirm off
25696 Disables confirmation requests. Note that running @value{GDBN} with
25697 the @option{--batch} option (@pxref{Mode Options, -batch}) also
25698 automatically disables confirmation requests.
25700 @item set confirm on
25701 Enables confirmation requests (the default).
25703 @kindex show confirm
25705 Displays state of confirmation requests.
25709 @cindex command tracing
25710 If you need to debug user-defined commands or sourced files you may find it
25711 useful to enable @dfn{command tracing}. In this mode each command will be
25712 printed as it is executed, prefixed with one or more @samp{+} symbols, the
25713 quantity denoting the call depth of each command.
25716 @kindex set trace-commands
25717 @cindex command scripts, debugging
25718 @item set trace-commands on
25719 Enable command tracing.
25720 @item set trace-commands off
25721 Disable command tracing.
25722 @item show trace-commands
25723 Display the current state of command tracing.
25726 @node Debugging Output
25727 @section Optional Messages about Internal Happenings
25728 @cindex optional debugging messages
25730 @value{GDBN} has commands that enable optional debugging messages from
25731 various @value{GDBN} subsystems; normally these commands are of
25732 interest to @value{GDBN} maintainers, or when reporting a bug. This
25733 section documents those commands.
25736 @kindex set exec-done-display
25737 @item set exec-done-display
25738 Turns on or off the notification of asynchronous commands'
25739 completion. When on, @value{GDBN} will print a message when an
25740 asynchronous command finishes its execution. The default is off.
25741 @kindex show exec-done-display
25742 @item show exec-done-display
25743 Displays the current setting of asynchronous command completion
25746 @cindex ARM AArch64
25747 @item set debug aarch64
25748 Turns on or off display of debugging messages related to ARM AArch64.
25749 The default is off.
25751 @item show debug aarch64
25752 Displays the current state of displaying debugging messages related to
25754 @cindex gdbarch debugging info
25755 @cindex architecture debugging info
25756 @item set debug arch
25757 Turns on or off display of gdbarch debugging info. The default is off
25758 @item show debug arch
25759 Displays the current state of displaying gdbarch debugging info.
25760 @item set debug aix-solib
25761 @cindex AIX shared library debugging
25762 Control display of debugging messages from the AIX shared library
25763 support module. The default is off.
25764 @item show debug aix-thread
25765 Show the current state of displaying AIX shared library debugging messages.
25766 @item set debug aix-thread
25767 @cindex AIX threads
25768 Display debugging messages about inner workings of the AIX thread
25770 @item show debug aix-thread
25771 Show the current state of AIX thread debugging info display.
25772 @item set debug check-physname
25774 Check the results of the ``physname'' computation. When reading DWARF
25775 debugging information for C@t{++}, @value{GDBN} attempts to compute
25776 each entity's name. @value{GDBN} can do this computation in two
25777 different ways, depending on exactly what information is present.
25778 When enabled, this setting causes @value{GDBN} to compute the names
25779 both ways and display any discrepancies.
25780 @item show debug check-physname
25781 Show the current state of ``physname'' checking.
25782 @item set debug coff-pe-read
25783 @cindex COFF/PE exported symbols
25784 Control display of debugging messages related to reading of COFF/PE
25785 exported symbols. The default is off.
25786 @item show debug coff-pe-read
25787 Displays the current state of displaying debugging messages related to
25788 reading of COFF/PE exported symbols.
25789 @item set debug dwarf-die
25791 Dump DWARF DIEs after they are read in.
25792 The value is the number of nesting levels to print.
25793 A value of zero turns off the display.
25794 @item show debug dwarf-die
25795 Show the current state of DWARF DIE debugging.
25796 @item set debug dwarf-line
25797 @cindex DWARF Line Tables
25798 Turns on or off display of debugging messages related to reading
25799 DWARF line tables. The default is 0 (off).
25800 A value of 1 provides basic information.
25801 A value greater than 1 provides more verbose information.
25802 @item show debug dwarf-line
25803 Show the current state of DWARF line table debugging.
25804 @item set debug dwarf-read
25805 @cindex DWARF Reading
25806 Turns on or off display of debugging messages related to reading
25807 DWARF debug info. The default is 0 (off).
25808 A value of 1 provides basic information.
25809 A value greater than 1 provides more verbose information.
25810 @item show debug dwarf-read
25811 Show the current state of DWARF reader debugging.
25812 @item set debug displaced
25813 @cindex displaced stepping debugging info
25814 Turns on or off display of @value{GDBN} debugging info for the
25815 displaced stepping support. The default is off.
25816 @item show debug displaced
25817 Displays the current state of displaying @value{GDBN} debugging info
25818 related to displaced stepping.
25819 @item set debug event
25820 @cindex event debugging info
25821 Turns on or off display of @value{GDBN} event debugging info. The
25823 @item show debug event
25824 Displays the current state of displaying @value{GDBN} event debugging
25826 @item set debug expression
25827 @cindex expression debugging info
25828 Turns on or off display of debugging info about @value{GDBN}
25829 expression parsing. The default is off.
25830 @item show debug expression
25831 Displays the current state of displaying debugging info about
25832 @value{GDBN} expression parsing.
25833 @item set debug fbsd-lwp
25834 @cindex FreeBSD LWP debug messages
25835 Turns on or off debugging messages from the FreeBSD LWP debug support.
25836 @item show debug fbsd-lwp
25837 Show the current state of FreeBSD LWP debugging messages.
25838 @item set debug fbsd-nat
25839 @cindex FreeBSD native target debug messages
25840 Turns on or off debugging messages from the FreeBSD native target.
25841 @item show debug fbsd-nat
25842 Show the current state of FreeBSD native target debugging messages.
25843 @item set debug frame
25844 @cindex frame debugging info
25845 Turns on or off display of @value{GDBN} frame debugging info. The
25847 @item show debug frame
25848 Displays the current state of displaying @value{GDBN} frame debugging
25850 @item set debug gnu-nat
25851 @cindex @sc{gnu}/Hurd debug messages
25852 Turn on or off debugging messages from the @sc{gnu}/Hurd debug support.
25853 @item show debug gnu-nat
25854 Show the current state of @sc{gnu}/Hurd debugging messages.
25855 @item set debug infrun
25856 @cindex inferior debugging info
25857 Turns on or off display of @value{GDBN} debugging info for running the inferior.
25858 The default is off. @file{infrun.c} contains GDB's runtime state machine used
25859 for implementing operations such as single-stepping the inferior.
25860 @item show debug infrun
25861 Displays the current state of @value{GDBN} inferior debugging.
25862 @item set debug jit
25863 @cindex just-in-time compilation, debugging messages
25864 Turn on or off debugging messages from JIT debug support.
25865 @item show debug jit
25866 Displays the current state of @value{GDBN} JIT debugging.
25867 @item set debug lin-lwp
25868 @cindex @sc{gnu}/Linux LWP debug messages
25869 @cindex Linux lightweight processes
25870 Turn on or off debugging messages from the Linux LWP debug support.
25871 @item show debug lin-lwp
25872 Show the current state of Linux LWP debugging messages.
25873 @item set debug linux-namespaces
25874 @cindex @sc{gnu}/Linux namespaces debug messages
25875 Turn on or off debugging messages from the Linux namespaces debug support.
25876 @item show debug linux-namespaces
25877 Show the current state of Linux namespaces debugging messages.
25878 @item set debug mach-o
25879 @cindex Mach-O symbols processing
25880 Control display of debugging messages related to Mach-O symbols
25881 processing. The default is off.
25882 @item show debug mach-o
25883 Displays the current state of displaying debugging messages related to
25884 reading of COFF/PE exported symbols.
25885 @item set debug notification
25886 @cindex remote async notification debugging info
25887 Turn on or off debugging messages about remote async notification.
25888 The default is off.
25889 @item show debug notification
25890 Displays the current state of remote async notification debugging messages.
25891 @item set debug observer
25892 @cindex observer debugging info
25893 Turns on or off display of @value{GDBN} observer debugging. This
25894 includes info such as the notification of observable events.
25895 @item show debug observer
25896 Displays the current state of observer debugging.
25897 @item set debug overload
25898 @cindex C@t{++} overload debugging info
25899 Turns on or off display of @value{GDBN} C@t{++} overload debugging
25900 info. This includes info such as ranking of functions, etc. The default
25902 @item show debug overload
25903 Displays the current state of displaying @value{GDBN} C@t{++} overload
25905 @cindex expression parser, debugging info
25906 @cindex debug expression parser
25907 @item set debug parser
25908 Turns on or off the display of expression parser debugging output.
25909 Internally, this sets the @code{yydebug} variable in the expression
25910 parser. @xref{Tracing, , Tracing Your Parser, bison, Bison}, for
25911 details. The default is off.
25912 @item show debug parser
25913 Show the current state of expression parser debugging.
25914 @cindex packets, reporting on stdout
25915 @cindex serial connections, debugging
25916 @cindex debug remote protocol
25917 @cindex remote protocol debugging
25918 @cindex display remote packets
25919 @item set debug remote
25920 Turns on or off display of reports on all packets sent back and forth across
25921 the serial line to the remote machine. The info is printed on the
25922 @value{GDBN} standard output stream. The default is off.
25923 @item show debug remote
25924 Displays the state of display of remote packets.
25926 @item set debug separate-debug-file
25927 Turns on or off display of debug output about separate debug file search.
25928 @item show debug separate-debug-file
25929 Displays the state of separate debug file search debug output.
25931 @item set debug serial
25932 Turns on or off display of @value{GDBN} serial debugging info. The
25934 @item show debug serial
25935 Displays the current state of displaying @value{GDBN} serial debugging
25937 @item set debug solib-frv
25938 @cindex FR-V shared-library debugging
25939 Turn on or off debugging messages for FR-V shared-library code.
25940 @item show debug solib-frv
25941 Display the current state of FR-V shared-library code debugging
25943 @item set debug symbol-lookup
25944 @cindex symbol lookup
25945 Turns on or off display of debugging messages related to symbol lookup.
25946 The default is 0 (off).
25947 A value of 1 provides basic information.
25948 A value greater than 1 provides more verbose information.
25949 @item show debug symbol-lookup
25950 Show the current state of symbol lookup debugging messages.
25951 @item set debug symfile
25952 @cindex symbol file functions
25953 Turns on or off display of debugging messages related to symbol file functions.
25954 The default is off. @xref{Files}.
25955 @item show debug symfile
25956 Show the current state of symbol file debugging messages.
25957 @item set debug symtab-create
25958 @cindex symbol table creation
25959 Turns on or off display of debugging messages related to symbol table creation.
25960 The default is 0 (off).
25961 A value of 1 provides basic information.
25962 A value greater than 1 provides more verbose information.
25963 @item show debug symtab-create
25964 Show the current state of symbol table creation debugging.
25965 @item set debug target
25966 @cindex target debugging info
25967 Turns on or off display of @value{GDBN} target debugging info. This info
25968 includes what is going on at the target level of GDB, as it happens. The
25969 default is 0. Set it to 1 to track events, and to 2 to also track the
25970 value of large memory transfers.
25971 @item show debug target
25972 Displays the current state of displaying @value{GDBN} target debugging
25974 @item set debug timestamp
25975 @cindex timestampping debugging info
25976 Turns on or off display of timestamps with @value{GDBN} debugging info.
25977 When enabled, seconds and microseconds are displayed before each debugging
25979 @item show debug timestamp
25980 Displays the current state of displaying timestamps with @value{GDBN}
25982 @item set debug varobj
25983 @cindex variable object debugging info
25984 Turns on or off display of @value{GDBN} variable object debugging
25985 info. The default is off.
25986 @item show debug varobj
25987 Displays the current state of displaying @value{GDBN} variable object
25989 @item set debug xml
25990 @cindex XML parser debugging
25991 Turn on or off debugging messages for built-in XML parsers.
25992 @item show debug xml
25993 Displays the current state of XML debugging messages.
25996 @node Other Misc Settings
25997 @section Other Miscellaneous Settings
25998 @cindex miscellaneous settings
26001 @kindex set interactive-mode
26002 @item set interactive-mode
26003 If @code{on}, forces @value{GDBN} to assume that GDB was started
26004 in a terminal. In practice, this means that @value{GDBN} should wait
26005 for the user to answer queries generated by commands entered at
26006 the command prompt. If @code{off}, forces @value{GDBN} to operate
26007 in the opposite mode, and it uses the default answers to all queries.
26008 If @code{auto} (the default), @value{GDBN} tries to determine whether
26009 its standard input is a terminal, and works in interactive-mode if it
26010 is, non-interactively otherwise.
26012 In the vast majority of cases, the debugger should be able to guess
26013 correctly which mode should be used. But this setting can be useful
26014 in certain specific cases, such as running a MinGW @value{GDBN}
26015 inside a cygwin window.
26017 @kindex show interactive-mode
26018 @item show interactive-mode
26019 Displays whether the debugger is operating in interactive mode or not.
26022 @node Extending GDB
26023 @chapter Extending @value{GDBN}
26024 @cindex extending GDB
26026 @value{GDBN} provides several mechanisms for extension.
26027 @value{GDBN} also provides the ability to automatically load
26028 extensions when it reads a file for debugging. This allows the
26029 user to automatically customize @value{GDBN} for the program
26033 * Sequences:: Canned Sequences of @value{GDBN} Commands
26034 * Python:: Extending @value{GDBN} using Python
26035 * Guile:: Extending @value{GDBN} using Guile
26036 * Auto-loading extensions:: Automatically loading extensions
26037 * Multiple Extension Languages:: Working with multiple extension languages
26038 * Aliases:: Creating new spellings of existing commands
26041 To facilitate the use of extension languages, @value{GDBN} is capable
26042 of evaluating the contents of a file. When doing so, @value{GDBN}
26043 can recognize which extension language is being used by looking at
26044 the filename extension. Files with an unrecognized filename extension
26045 are always treated as a @value{GDBN} Command Files.
26046 @xref{Command Files,, Command files}.
26048 You can control how @value{GDBN} evaluates these files with the following
26052 @kindex set script-extension
26053 @kindex show script-extension
26054 @item set script-extension off
26055 All scripts are always evaluated as @value{GDBN} Command Files.
26057 @item set script-extension soft
26058 The debugger determines the scripting language based on filename
26059 extension. If this scripting language is supported, @value{GDBN}
26060 evaluates the script using that language. Otherwise, it evaluates
26061 the file as a @value{GDBN} Command File.
26063 @item set script-extension strict
26064 The debugger determines the scripting language based on filename
26065 extension, and evaluates the script using that language. If the
26066 language is not supported, then the evaluation fails.
26068 @item show script-extension
26069 Display the current value of the @code{script-extension} option.
26074 @section Canned Sequences of Commands
26076 Aside from breakpoint commands (@pxref{Break Commands, ,Breakpoint
26077 Command Lists}), @value{GDBN} provides two ways to store sequences of
26078 commands for execution as a unit: user-defined commands and command
26082 * Define:: How to define your own commands
26083 * Hooks:: Hooks for user-defined commands
26084 * Command Files:: How to write scripts of commands to be stored in a file
26085 * Output:: Commands for controlled output
26086 * Auto-loading sequences:: Controlling auto-loaded command files
26090 @subsection User-defined Commands
26092 @cindex user-defined command
26093 @cindex arguments, to user-defined commands
26094 A @dfn{user-defined command} is a sequence of @value{GDBN} commands to
26095 which you assign a new name as a command. This is done with the
26096 @code{define} command. User commands may accept an unlimited number of arguments
26097 separated by whitespace. Arguments are accessed within the user command
26098 via @code{$arg0@dots{}$argN}. A trivial example:
26102 print $arg0 + $arg1 + $arg2
26107 To execute the command use:
26114 This defines the command @code{adder}, which prints the sum of
26115 its three arguments. Note the arguments are text substitutions, so they may
26116 reference variables, use complex expressions, or even perform inferior
26119 @cindex argument count in user-defined commands
26120 @cindex how many arguments (user-defined commands)
26121 In addition, @code{$argc} may be used to find out how many arguments have
26127 print $arg0 + $arg1
26130 print $arg0 + $arg1 + $arg2
26135 Combining with the @code{eval} command (@pxref{eval}) makes it easier
26136 to process a variable number of arguments:
26143 eval "set $sum = $sum + $arg%d", $i
26153 @item define @var{commandname}
26154 Define a command named @var{commandname}. If there is already a command
26155 by that name, you are asked to confirm that you want to redefine it.
26156 The argument @var{commandname} may be a bare command name consisting of letters,
26157 numbers, dashes, and underscores. It may also start with any predefined
26158 prefix command. For example, @samp{define target my-target} creates
26159 a user-defined @samp{target my-target} command.
26161 The definition of the command is made up of other @value{GDBN} command lines,
26162 which are given following the @code{define} command. The end of these
26163 commands is marked by a line containing @code{end}.
26166 @kindex end@r{ (user-defined commands)}
26167 @item document @var{commandname}
26168 Document the user-defined command @var{commandname}, so that it can be
26169 accessed by @code{help}. The command @var{commandname} must already be
26170 defined. This command reads lines of documentation just as @code{define}
26171 reads the lines of the command definition, ending with @code{end}.
26172 After the @code{document} command is finished, @code{help} on command
26173 @var{commandname} displays the documentation you have written.
26175 You may use the @code{document} command again to change the
26176 documentation of a command. Redefining the command with @code{define}
26177 does not change the documentation.
26179 @kindex dont-repeat
26180 @cindex don't repeat command
26182 Used inside a user-defined command, this tells @value{GDBN} that this
26183 command should not be repeated when the user hits @key{RET}
26184 (@pxref{Command Syntax, repeat last command}).
26186 @kindex help user-defined
26187 @item help user-defined
26188 List all user-defined commands and all python commands defined in class
26189 COMAND_USER. The first line of the documentation or docstring is
26194 @itemx show user @var{commandname}
26195 Display the @value{GDBN} commands used to define @var{commandname} (but
26196 not its documentation). If no @var{commandname} is given, display the
26197 definitions for all user-defined commands.
26198 This does not work for user-defined python commands.
26200 @cindex infinite recursion in user-defined commands
26201 @kindex show max-user-call-depth
26202 @kindex set max-user-call-depth
26203 @item show max-user-call-depth
26204 @itemx set max-user-call-depth
26205 The value of @code{max-user-call-depth} controls how many recursion
26206 levels are allowed in user-defined commands before @value{GDBN} suspects an
26207 infinite recursion and aborts the command.
26208 This does not apply to user-defined python commands.
26211 In addition to the above commands, user-defined commands frequently
26212 use control flow commands, described in @ref{Command Files}.
26214 When user-defined commands are executed, the
26215 commands of the definition are not printed. An error in any command
26216 stops execution of the user-defined command.
26218 If used interactively, commands that would ask for confirmation proceed
26219 without asking when used inside a user-defined command. Many @value{GDBN}
26220 commands that normally print messages to say what they are doing omit the
26221 messages when used in a user-defined command.
26224 @subsection User-defined Command Hooks
26225 @cindex command hooks
26226 @cindex hooks, for commands
26227 @cindex hooks, pre-command
26230 You may define @dfn{hooks}, which are a special kind of user-defined
26231 command. Whenever you run the command @samp{foo}, if the user-defined
26232 command @samp{hook-foo} exists, it is executed (with no arguments)
26233 before that command.
26235 @cindex hooks, post-command
26237 A hook may also be defined which is run after the command you executed.
26238 Whenever you run the command @samp{foo}, if the user-defined command
26239 @samp{hookpost-foo} exists, it is executed (with no arguments) after
26240 that command. Post-execution hooks may exist simultaneously with
26241 pre-execution hooks, for the same command.
26243 It is valid for a hook to call the command which it hooks. If this
26244 occurs, the hook is not re-executed, thereby avoiding infinite recursion.
26246 @c It would be nice if hookpost could be passed a parameter indicating
26247 @c if the command it hooks executed properly or not. FIXME!
26249 @kindex stop@r{, a pseudo-command}
26250 In addition, a pseudo-command, @samp{stop} exists. Defining
26251 (@samp{hook-stop}) makes the associated commands execute every time
26252 execution stops in your program: before breakpoint commands are run,
26253 displays are printed, or the stack frame is printed.
26255 For example, to ignore @code{SIGALRM} signals while
26256 single-stepping, but treat them normally during normal execution,
26261 handle SIGALRM nopass
26265 handle SIGALRM pass
26268 define hook-continue
26269 handle SIGALRM pass
26273 As a further example, to hook at the beginning and end of the @code{echo}
26274 command, and to add extra text to the beginning and end of the message,
26282 define hookpost-echo
26286 (@value{GDBP}) echo Hello World
26287 <<<---Hello World--->>>
26292 You can define a hook for any single-word command in @value{GDBN}, but
26293 not for command aliases; you should define a hook for the basic command
26294 name, e.g.@: @code{backtrace} rather than @code{bt}.
26295 @c FIXME! So how does Joe User discover whether a command is an alias
26297 You can hook a multi-word command by adding @code{hook-} or
26298 @code{hookpost-} to the last word of the command, e.g.@:
26299 @samp{define target hook-remote} to add a hook to @samp{target remote}.
26301 If an error occurs during the execution of your hook, execution of
26302 @value{GDBN} commands stops and @value{GDBN} issues a prompt
26303 (before the command that you actually typed had a chance to run).
26305 If you try to define a hook which does not match any known command, you
26306 get a warning from the @code{define} command.
26308 @node Command Files
26309 @subsection Command Files
26311 @cindex command files
26312 @cindex scripting commands
26313 A command file for @value{GDBN} is a text file made of lines that are
26314 @value{GDBN} commands. Comments (lines starting with @kbd{#}) may
26315 also be included. An empty line in a command file does nothing; it
26316 does not mean to repeat the last command, as it would from the
26319 You can request the execution of a command file with the @code{source}
26320 command. Note that the @code{source} command is also used to evaluate
26321 scripts that are not Command Files. The exact behavior can be configured
26322 using the @code{script-extension} setting.
26323 @xref{Extending GDB,, Extending GDB}.
26327 @cindex execute commands from a file
26328 @item source [-s] [-v] @var{filename}
26329 Execute the command file @var{filename}.
26332 The lines in a command file are generally executed sequentially,
26333 unless the order of execution is changed by one of the
26334 @emph{flow-control commands} described below. The commands are not
26335 printed as they are executed. An error in any command terminates
26336 execution of the command file and control is returned to the console.
26338 @value{GDBN} first searches for @var{filename} in the current directory.
26339 If the file is not found there, and @var{filename} does not specify a
26340 directory, then @value{GDBN} also looks for the file on the source search path
26341 (specified with the @samp{directory} command);
26342 except that @file{$cdir} is not searched because the compilation directory
26343 is not relevant to scripts.
26345 If @code{-s} is specified, then @value{GDBN} searches for @var{filename}
26346 on the search path even if @var{filename} specifies a directory.
26347 The search is done by appending @var{filename} to each element of the
26348 search path. So, for example, if @var{filename} is @file{mylib/myscript}
26349 and the search path contains @file{/home/user} then @value{GDBN} will
26350 look for the script @file{/home/user/mylib/myscript}.
26351 The search is also done if @var{filename} is an absolute path.
26352 For example, if @var{filename} is @file{/tmp/myscript} and
26353 the search path contains @file{/home/user} then @value{GDBN} will
26354 look for the script @file{/home/user/tmp/myscript}.
26355 For DOS-like systems, if @var{filename} contains a drive specification,
26356 it is stripped before concatenation. For example, if @var{filename} is
26357 @file{d:myscript} and the search path contains @file{c:/tmp} then @value{GDBN}
26358 will look for the script @file{c:/tmp/myscript}.
26360 If @code{-v}, for verbose mode, is given then @value{GDBN} displays
26361 each command as it is executed. The option must be given before
26362 @var{filename}, and is interpreted as part of the filename anywhere else.
26364 Commands that would ask for confirmation if used interactively proceed
26365 without asking when used in a command file. Many @value{GDBN} commands that
26366 normally print messages to say what they are doing omit the messages
26367 when called from command files.
26369 @value{GDBN} also accepts command input from standard input. In this
26370 mode, normal output goes to standard output and error output goes to
26371 standard error. Errors in a command file supplied on standard input do
26372 not terminate execution of the command file---execution continues with
26376 gdb < cmds > log 2>&1
26379 (The syntax above will vary depending on the shell used.) This example
26380 will execute commands from the file @file{cmds}. All output and errors
26381 would be directed to @file{log}.
26383 Since commands stored on command files tend to be more general than
26384 commands typed interactively, they frequently need to deal with
26385 complicated situations, such as different or unexpected values of
26386 variables and symbols, changes in how the program being debugged is
26387 built, etc. @value{GDBN} provides a set of flow-control commands to
26388 deal with these complexities. Using these commands, you can write
26389 complex scripts that loop over data structures, execute commands
26390 conditionally, etc.
26397 This command allows to include in your script conditionally executed
26398 commands. The @code{if} command takes a single argument, which is an
26399 expression to evaluate. It is followed by a series of commands that
26400 are executed only if the expression is true (its value is nonzero).
26401 There can then optionally be an @code{else} line, followed by a series
26402 of commands that are only executed if the expression was false. The
26403 end of the list is marked by a line containing @code{end}.
26407 This command allows to write loops. Its syntax is similar to
26408 @code{if}: the command takes a single argument, which is an expression
26409 to evaluate, and must be followed by the commands to execute, one per
26410 line, terminated by an @code{end}. These commands are called the
26411 @dfn{body} of the loop. The commands in the body of @code{while} are
26412 executed repeatedly as long as the expression evaluates to true.
26416 This command exits the @code{while} loop in whose body it is included.
26417 Execution of the script continues after that @code{while}s @code{end}
26420 @kindex loop_continue
26421 @item loop_continue
26422 This command skips the execution of the rest of the body of commands
26423 in the @code{while} loop in whose body it is included. Execution
26424 branches to the beginning of the @code{while} loop, where it evaluates
26425 the controlling expression.
26427 @kindex end@r{ (if/else/while commands)}
26429 Terminate the block of commands that are the body of @code{if},
26430 @code{else}, or @code{while} flow-control commands.
26435 @subsection Commands for Controlled Output
26437 During the execution of a command file or a user-defined command, normal
26438 @value{GDBN} output is suppressed; the only output that appears is what is
26439 explicitly printed by the commands in the definition. This section
26440 describes three commands useful for generating exactly the output you
26445 @item echo @var{text}
26446 @c I do not consider backslash-space a standard C escape sequence
26447 @c because it is not in ANSI.
26448 Print @var{text}. Nonprinting characters can be included in
26449 @var{text} using C escape sequences, such as @samp{\n} to print a
26450 newline. @strong{No newline is printed unless you specify one.}
26451 In addition to the standard C escape sequences, a backslash followed
26452 by a space stands for a space. This is useful for displaying a
26453 string with spaces at the beginning or the end, since leading and
26454 trailing spaces are otherwise trimmed from all arguments.
26455 To print @samp{@w{ }and foo =@w{ }}, use the command
26456 @samp{echo \@w{ }and foo = \@w{ }}.
26458 A backslash at the end of @var{text} can be used, as in C, to continue
26459 the command onto subsequent lines. For example,
26462 echo This is some text\n\
26463 which is continued\n\
26464 onto several lines.\n
26467 produces the same output as
26470 echo This is some text\n
26471 echo which is continued\n
26472 echo onto several lines.\n
26476 @item output @var{expression}
26477 Print the value of @var{expression} and nothing but that value: no
26478 newlines, no @samp{$@var{nn} = }. The value is not entered in the
26479 value history either. @xref{Expressions, ,Expressions}, for more information
26482 @item output/@var{fmt} @var{expression}
26483 Print the value of @var{expression} in format @var{fmt}. You can use
26484 the same formats as for @code{print}. @xref{Output Formats,,Output
26485 Formats}, for more information.
26488 @item printf @var{template}, @var{expressions}@dots{}
26489 Print the values of one or more @var{expressions} under the control of
26490 the string @var{template}. To print several values, make
26491 @var{expressions} be a comma-separated list of individual expressions,
26492 which may be either numbers or pointers. Their values are printed as
26493 specified by @var{template}, exactly as a C program would do by
26494 executing the code below:
26497 printf (@var{template}, @var{expressions}@dots{});
26500 As in @code{C} @code{printf}, ordinary characters in @var{template}
26501 are printed verbatim, while @dfn{conversion specification} introduced
26502 by the @samp{%} character cause subsequent @var{expressions} to be
26503 evaluated, their values converted and formatted according to type and
26504 style information encoded in the conversion specifications, and then
26507 For example, you can print two values in hex like this:
26510 printf "foo, bar-foo = 0x%x, 0x%x\n", foo, bar-foo
26513 @code{printf} supports all the standard @code{C} conversion
26514 specifications, including the flags and modifiers between the @samp{%}
26515 character and the conversion letter, with the following exceptions:
26519 The argument-ordering modifiers, such as @samp{2$}, are not supported.
26522 The modifier @samp{*} is not supported for specifying precision or
26526 The @samp{'} flag (for separation of digits into groups according to
26527 @code{LC_NUMERIC'}) is not supported.
26530 The type modifiers @samp{hh}, @samp{j}, @samp{t}, and @samp{z} are not
26534 The conversion letter @samp{n} (as in @samp{%n}) is not supported.
26537 The conversion letters @samp{a} and @samp{A} are not supported.
26541 Note that the @samp{ll} type modifier is supported only if the
26542 underlying @code{C} implementation used to build @value{GDBN} supports
26543 the @code{long long int} type, and the @samp{L} type modifier is
26544 supported only if @code{long double} type is available.
26546 As in @code{C}, @code{printf} supports simple backslash-escape
26547 sequences, such as @code{\n}, @samp{\t}, @samp{\\}, @samp{\"},
26548 @samp{\a}, and @samp{\f}, that consist of backslash followed by a
26549 single character. Octal and hexadecimal escape sequences are not
26552 Additionally, @code{printf} supports conversion specifications for DFP
26553 (@dfn{Decimal Floating Point}) types using the following length modifiers
26554 together with a floating point specifier.
26559 @samp{H} for printing @code{Decimal32} types.
26562 @samp{D} for printing @code{Decimal64} types.
26565 @samp{DD} for printing @code{Decimal128} types.
26568 If the underlying @code{C} implementation used to build @value{GDBN} has
26569 support for the three length modifiers for DFP types, other modifiers
26570 such as width and precision will also be available for @value{GDBN} to use.
26572 In case there is no such @code{C} support, no additional modifiers will be
26573 available and the value will be printed in the standard way.
26575 Here's an example of printing DFP types using the above conversion letters:
26577 printf "D32: %Hf - D64: %Df - D128: %DDf\n",1.2345df,1.2E10dd,1.2E1dl
26582 @item eval @var{template}, @var{expressions}@dots{}
26583 Convert the values of one or more @var{expressions} under the control of
26584 the string @var{template} to a command line, and call it.
26588 @node Auto-loading sequences
26589 @subsection Controlling auto-loading native @value{GDBN} scripts
26590 @cindex native script auto-loading
26592 When a new object file is read (for example, due to the @code{file}
26593 command, or because the inferior has loaded a shared library),
26594 @value{GDBN} will look for the command file @file{@var{objfile}-gdb.gdb}.
26595 @xref{Auto-loading extensions}.
26597 Auto-loading can be enabled or disabled,
26598 and the list of auto-loaded scripts can be printed.
26601 @anchor{set auto-load gdb-scripts}
26602 @kindex set auto-load gdb-scripts
26603 @item set auto-load gdb-scripts [on|off]
26604 Enable or disable the auto-loading of canned sequences of commands scripts.
26606 @anchor{show auto-load gdb-scripts}
26607 @kindex show auto-load gdb-scripts
26608 @item show auto-load gdb-scripts
26609 Show whether auto-loading of canned sequences of commands scripts is enabled or
26612 @anchor{info auto-load gdb-scripts}
26613 @kindex info auto-load gdb-scripts
26614 @cindex print list of auto-loaded canned sequences of commands scripts
26615 @item info auto-load gdb-scripts [@var{regexp}]
26616 Print the list of all canned sequences of commands scripts that @value{GDBN}
26620 If @var{regexp} is supplied only canned sequences of commands scripts with
26621 matching names are printed.
26623 @c Python docs live in a separate file.
26624 @include python.texi
26626 @c Guile docs live in a separate file.
26627 @include guile.texi
26629 @node Auto-loading extensions
26630 @section Auto-loading extensions
26631 @cindex auto-loading extensions
26633 @value{GDBN} provides two mechanisms for automatically loading extensions
26634 when a new object file is read (for example, due to the @code{file}
26635 command, or because the inferior has loaded a shared library):
26636 @file{@var{objfile}-gdb.@var{ext}} and the @code{.debug_gdb_scripts}
26637 section of modern file formats like ELF.
26640 * objfile-gdb.ext file: objfile-gdbdotext file. The @file{@var{objfile}-gdb.@var{ext}} file
26641 * .debug_gdb_scripts section: dotdebug_gdb_scripts section. The @code{.debug_gdb_scripts} section
26642 * Which flavor to choose?::
26645 The auto-loading feature is useful for supplying application-specific
26646 debugging commands and features.
26648 Auto-loading can be enabled or disabled,
26649 and the list of auto-loaded scripts can be printed.
26650 See the @samp{auto-loading} section of each extension language
26651 for more information.
26652 For @value{GDBN} command files see @ref{Auto-loading sequences}.
26653 For Python files see @ref{Python Auto-loading}.
26655 Note that loading of this script file also requires accordingly configured
26656 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26658 @node objfile-gdbdotext file
26659 @subsection The @file{@var{objfile}-gdb.@var{ext}} file
26660 @cindex @file{@var{objfile}-gdb.gdb}
26661 @cindex @file{@var{objfile}-gdb.py}
26662 @cindex @file{@var{objfile}-gdb.scm}
26664 When a new object file is read, @value{GDBN} looks for a file named
26665 @file{@var{objfile}-gdb.@var{ext}} (we call it @var{script-name} below),
26666 where @var{objfile} is the object file's name and
26667 where @var{ext} is the file extension for the extension language:
26670 @item @file{@var{objfile}-gdb.gdb}
26671 GDB's own command language
26672 @item @file{@var{objfile}-gdb.py}
26674 @item @file{@var{objfile}-gdb.scm}
26678 @var{script-name} is formed by ensuring that the file name of @var{objfile}
26679 is absolute, following all symlinks, and resolving @code{.} and @code{..}
26680 components, and appending the @file{-gdb.@var{ext}} suffix.
26681 If this file exists and is readable, @value{GDBN} will evaluate it as a
26682 script in the specified extension language.
26684 If this file does not exist, then @value{GDBN} will look for
26685 @var{script-name} file in all of the directories as specified below.
26687 Note that loading of these files requires an accordingly configured
26688 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26690 For object files using @file{.exe} suffix @value{GDBN} tries to load first the
26691 scripts normally according to its @file{.exe} filename. But if no scripts are
26692 found @value{GDBN} also tries script filenames matching the object file without
26693 its @file{.exe} suffix. This @file{.exe} stripping is case insensitive and it
26694 is attempted on any platform. This makes the script filenames compatible
26695 between Unix and MS-Windows hosts.
26698 @anchor{set auto-load scripts-directory}
26699 @kindex set auto-load scripts-directory
26700 @item set auto-load scripts-directory @r{[}@var{directories}@r{]}
26701 Control @value{GDBN} auto-loaded scripts location. Multiple directory entries
26702 may be delimited by the host platform path separator in use
26703 (@samp{:} on Unix, @samp{;} on MS-Windows and MS-DOS).
26705 Each entry here needs to be covered also by the security setting
26706 @code{set auto-load safe-path} (@pxref{set auto-load safe-path}).
26708 @anchor{with-auto-load-dir}
26709 This variable defaults to @file{$debugdir:$datadir/auto-load}. The default
26710 @code{set auto-load safe-path} value can be also overriden by @value{GDBN}
26711 configuration option @option{--with-auto-load-dir}.
26713 Any reference to @file{$debugdir} will get replaced by
26714 @var{debug-file-directory} value (@pxref{Separate Debug Files}) and any
26715 reference to @file{$datadir} will get replaced by @var{data-directory} which is
26716 determined at @value{GDBN} startup (@pxref{Data Files}). @file{$debugdir} and
26717 @file{$datadir} must be placed as a directory component --- either alone or
26718 delimited by @file{/} or @file{\} directory separators, depending on the host
26721 The list of directories uses path separator (@samp{:} on GNU and Unix
26722 systems, @samp{;} on MS-Windows and MS-DOS) to separate directories, similarly
26723 to the @env{PATH} environment variable.
26725 @anchor{show auto-load scripts-directory}
26726 @kindex show auto-load scripts-directory
26727 @item show auto-load scripts-directory
26728 Show @value{GDBN} auto-loaded scripts location.
26730 @anchor{add-auto-load-scripts-directory}
26731 @kindex add-auto-load-scripts-directory
26732 @item add-auto-load-scripts-directory @r{[}@var{directories}@dots{}@r{]}
26733 Add an entry (or list of entries) to the list of auto-loaded scripts locations.
26734 Multiple entries may be delimited by the host platform path separator in use.
26737 @value{GDBN} does not track which files it has already auto-loaded this way.
26738 @value{GDBN} will load the associated script every time the corresponding
26739 @var{objfile} is opened.
26740 So your @file{-gdb.@var{ext}} file should be careful to avoid errors if it
26741 is evaluated more than once.
26743 @node dotdebug_gdb_scripts section
26744 @subsection The @code{.debug_gdb_scripts} section
26745 @cindex @code{.debug_gdb_scripts} section
26747 For systems using file formats like ELF and COFF,
26748 when @value{GDBN} loads a new object file
26749 it will look for a special section named @code{.debug_gdb_scripts}.
26750 If this section exists, its contents is a list of null-terminated entries
26751 specifying scripts to load. Each entry begins with a non-null prefix byte that
26752 specifies the kind of entry, typically the extension language and whether the
26753 script is in a file or inlined in @code{.debug_gdb_scripts}.
26755 The following entries are supported:
26758 @item SECTION_SCRIPT_ID_PYTHON_FILE = 1
26759 @item SECTION_SCRIPT_ID_SCHEME_FILE = 3
26760 @item SECTION_SCRIPT_ID_PYTHON_TEXT = 4
26761 @item SECTION_SCRIPT_ID_SCHEME_TEXT = 6
26764 @subsubsection Script File Entries
26766 If the entry specifies a file, @value{GDBN} will look for the file first
26767 in the current directory and then along the source search path
26768 (@pxref{Source Path, ,Specifying Source Directories}),
26769 except that @file{$cdir} is not searched, since the compilation
26770 directory is not relevant to scripts.
26772 File entries can be placed in section @code{.debug_gdb_scripts} with,
26773 for example, this GCC macro for Python scripts.
26776 /* Note: The "MS" section flags are to remove duplicates. */
26777 #define DEFINE_GDB_PY_SCRIPT(script_name) \
26779 .pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n\
26780 .byte 1 /* Python */\n\
26781 .asciz \"" script_name "\"\n\
26787 For Guile scripts, replace @code{.byte 1} with @code{.byte 3}.
26788 Then one can reference the macro in a header or source file like this:
26791 DEFINE_GDB_PY_SCRIPT ("my-app-scripts.py")
26794 The script name may include directories if desired.
26796 Note that loading of this script file also requires accordingly configured
26797 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26799 If the macro invocation is put in a header, any application or library
26800 using this header will get a reference to the specified script,
26801 and with the use of @code{"MS"} attributes on the section, the linker
26802 will remove duplicates.
26804 @subsubsection Script Text Entries
26806 Script text entries allow to put the executable script in the entry
26807 itself instead of loading it from a file.
26808 The first line of the entry, everything after the prefix byte and up to
26809 the first newline (@code{0xa}) character, is the script name, and must not
26810 contain any kind of space character, e.g., spaces or tabs.
26811 The rest of the entry, up to the trailing null byte, is the script to
26812 execute in the specified language. The name needs to be unique among
26813 all script names, as @value{GDBN} executes each script only once based
26816 Here is an example from file @file{py-section-script.c} in the @value{GDBN}
26820 #include "symcat.h"
26821 #include "gdb/section-scripts.h"
26823 ".pushsection \".debug_gdb_scripts\", \"MS\",@@progbits,1\n"
26824 ".byte " XSTRING (SECTION_SCRIPT_ID_PYTHON_TEXT) "\n"
26825 ".ascii \"gdb.inlined-script\\n\"\n"
26826 ".ascii \"class test_cmd (gdb.Command):\\n\"\n"
26827 ".ascii \" def __init__ (self):\\n\"\n"
26828 ".ascii \" super (test_cmd, self).__init__ ("
26829 "\\\"test-cmd\\\", gdb.COMMAND_OBSCURE)\\n\"\n"
26830 ".ascii \" def invoke (self, arg, from_tty):\\n\"\n"
26831 ".ascii \" print (\\\"test-cmd output, arg = %s\\\" % arg)\\n\"\n"
26832 ".ascii \"test_cmd ()\\n\"\n"
26838 Loading of inlined scripts requires a properly configured
26839 @code{auto-load safe-path} (@pxref{Auto-loading safe path}).
26840 The path to specify in @code{auto-load safe-path} is the path of the file
26841 containing the @code{.debug_gdb_scripts} section.
26843 @node Which flavor to choose?
26844 @subsection Which flavor to choose?
26846 Given the multiple ways of auto-loading extensions, it might not always
26847 be clear which one to choose. This section provides some guidance.
26850 Benefits of the @file{-gdb.@var{ext}} way:
26854 Can be used with file formats that don't support multiple sections.
26857 Ease of finding scripts for public libraries.
26859 Scripts specified in the @code{.debug_gdb_scripts} section are searched for
26860 in the source search path.
26861 For publicly installed libraries, e.g., @file{libstdc++}, there typically
26862 isn't a source directory in which to find the script.
26865 Doesn't require source code additions.
26869 Benefits of the @code{.debug_gdb_scripts} way:
26873 Works with static linking.
26875 Scripts for libraries done the @file{-gdb.@var{ext}} way require an objfile to
26876 trigger their loading. When an application is statically linked the only
26877 objfile available is the executable, and it is cumbersome to attach all the
26878 scripts from all the input libraries to the executable's
26879 @file{-gdb.@var{ext}} script.
26882 Works with classes that are entirely inlined.
26884 Some classes can be entirely inlined, and thus there may not be an associated
26885 shared library to attach a @file{-gdb.@var{ext}} script to.
26888 Scripts needn't be copied out of the source tree.
26890 In some circumstances, apps can be built out of large collections of internal
26891 libraries, and the build infrastructure necessary to install the
26892 @file{-gdb.@var{ext}} scripts in a place where @value{GDBN} can find them is
26893 cumbersome. It may be easier to specify the scripts in the
26894 @code{.debug_gdb_scripts} section as relative paths, and add a path to the
26895 top of the source tree to the source search path.
26898 @node Multiple Extension Languages
26899 @section Multiple Extension Languages
26901 The Guile and Python extension languages do not share any state,
26902 and generally do not interfere with each other.
26903 There are some things to be aware of, however.
26905 @subsection Python comes first
26907 Python was @value{GDBN}'s first extension language, and to avoid breaking
26908 existing behaviour Python comes first. This is generally solved by the
26909 ``first one wins'' principle. @value{GDBN} maintains a list of enabled
26910 extension languages, and when it makes a call to an extension language,
26911 (say to pretty-print a value), it tries each in turn until an extension
26912 language indicates it has performed the request (e.g., has returned the
26913 pretty-printed form of a value).
26914 This extends to errors while performing such requests: If an error happens
26915 while, for example, trying to pretty-print an object then the error is
26916 reported and any following extension languages are not tried.
26919 @section Creating new spellings of existing commands
26920 @cindex aliases for commands
26922 It is often useful to define alternate spellings of existing commands.
26923 For example, if a new @value{GDBN} command defined in Python has
26924 a long name to type, it is handy to have an abbreviated version of it
26925 that involves less typing.
26927 @value{GDBN} itself uses aliases. For example @samp{s} is an alias
26928 of the @samp{step} command even though it is otherwise an ambiguous
26929 abbreviation of other commands like @samp{set} and @samp{show}.
26931 Aliases are also used to provide shortened or more common versions
26932 of multi-word commands. For example, @value{GDBN} provides the
26933 @samp{tty} alias of the @samp{set inferior-tty} command.
26935 You can define a new alias with the @samp{alias} command.
26940 @item alias [-a] [--] @var{ALIAS} = @var{COMMAND}
26944 @var{ALIAS} specifies the name of the new alias.
26945 Each word of @var{ALIAS} must consist of letters, numbers, dashes and
26948 @var{COMMAND} specifies the name of an existing command
26949 that is being aliased.
26951 The @samp{-a} option specifies that the new alias is an abbreviation
26952 of the command. Abbreviations are not shown in command
26953 lists displayed by the @samp{help} command.
26955 The @samp{--} option specifies the end of options,
26956 and is useful when @var{ALIAS} begins with a dash.
26958 Here is a simple example showing how to make an abbreviation
26959 of a command so that there is less to type.
26960 Suppose you were tired of typing @samp{disas}, the current
26961 shortest unambiguous abbreviation of the @samp{disassemble} command
26962 and you wanted an even shorter version named @samp{di}.
26963 The following will accomplish this.
26966 (gdb) alias -a di = disas
26969 Note that aliases are different from user-defined commands.
26970 With a user-defined command, you also need to write documentation
26971 for it with the @samp{document} command.
26972 An alias automatically picks up the documentation of the existing command.
26974 Here is an example where we make @samp{elms} an abbreviation of
26975 @samp{elements} in the @samp{set print elements} command.
26976 This is to show that you can make an abbreviation of any part
26980 (gdb) alias -a set print elms = set print elements
26981 (gdb) alias -a show print elms = show print elements
26982 (gdb) set p elms 20
26984 Limit on string chars or array elements to print is 200.
26987 Note that if you are defining an alias of a @samp{set} command,
26988 and you want to have an alias for the corresponding @samp{show}
26989 command, then you need to define the latter separately.
26991 Unambiguously abbreviated commands are allowed in @var{COMMAND} and
26992 @var{ALIAS}, just as they are normally.
26995 (gdb) alias -a set pr elms = set p ele
26998 Finally, here is an example showing the creation of a one word
26999 alias for a more complex command.
27000 This creates alias @samp{spe} of the command @samp{set print elements}.
27003 (gdb) alias spe = set print elements
27008 @chapter Command Interpreters
27009 @cindex command interpreters
27011 @value{GDBN} supports multiple command interpreters, and some command
27012 infrastructure to allow users or user interface writers to switch
27013 between interpreters or run commands in other interpreters.
27015 @value{GDBN} currently supports two command interpreters, the console
27016 interpreter (sometimes called the command-line interpreter or @sc{cli})
27017 and the machine interface interpreter (or @sc{gdb/mi}). This manual
27018 describes both of these interfaces in great detail.
27020 By default, @value{GDBN} will start with the console interpreter.
27021 However, the user may choose to start @value{GDBN} with another
27022 interpreter by specifying the @option{-i} or @option{--interpreter}
27023 startup options. Defined interpreters include:
27027 @cindex console interpreter
27028 The traditional console or command-line interpreter. This is the most often
27029 used interpreter with @value{GDBN}. With no interpreter specified at runtime,
27030 @value{GDBN} will use this interpreter.
27033 @cindex mi interpreter
27034 The newest @sc{gdb/mi} interface (currently @code{mi3}). Used primarily
27035 by programs wishing to use @value{GDBN} as a backend for a debugger GUI
27036 or an IDE. For more information, see @ref{GDB/MI, ,The @sc{gdb/mi}
27040 @cindex mi3 interpreter
27041 The @sc{gdb/mi} interface introduced in @value{GDBN} 9.1.
27044 @cindex mi2 interpreter
27045 The @sc{gdb/mi} interface introduced in @value{GDBN} 6.0.
27048 @cindex mi1 interpreter
27049 The @sc{gdb/mi} interface introduced in @value{GDBN} 5.1.
27053 @cindex invoke another interpreter
27055 @kindex interpreter-exec
27056 You may execute commands in any interpreter from the current
27057 interpreter using the appropriate command. If you are running the
27058 console interpreter, simply use the @code{interpreter-exec} command:
27061 interpreter-exec mi "-data-list-register-names"
27064 @sc{gdb/mi} has a similar command, although it is only available in versions of
27065 @value{GDBN} which support @sc{gdb/mi} version 2 (or greater).
27067 Note that @code{interpreter-exec} only changes the interpreter for the
27068 duration of the specified command. It does not change the interpreter
27071 @cindex start a new independent interpreter
27073 Although you may only choose a single interpreter at startup, it is
27074 possible to run an independent interpreter on a specified input/output
27075 device (usually a tty).
27077 For example, consider a debugger GUI or IDE that wants to provide a
27078 @value{GDBN} console view. It may do so by embedding a terminal
27079 emulator widget in its GUI, starting @value{GDBN} in the traditional
27080 command-line mode with stdin/stdout/stderr redirected to that
27081 terminal, and then creating an MI interpreter running on a specified
27082 input/output device. The console interpreter created by @value{GDBN}
27083 at startup handles commands the user types in the terminal widget,
27084 while the GUI controls and synchronizes state with @value{GDBN} using
27085 the separate MI interpreter.
27087 To start a new secondary @dfn{user interface} running MI, use the
27088 @code{new-ui} command:
27091 @cindex new user interface
27093 new-ui @var{interpreter} @var{tty}
27096 The @var{interpreter} parameter specifies the interpreter to run.
27097 This accepts the same values as the @code{interpreter-exec} command.
27098 For example, @samp{console}, @samp{mi}, @samp{mi2}, etc. The
27099 @var{tty} parameter specifies the name of the bidirectional file the
27100 interpreter uses for input/output, usually the name of a
27101 pseudoterminal slave on Unix systems. For example:
27104 (@value{GDBP}) new-ui mi /dev/pts/9
27108 runs an MI interpreter on @file{/dev/pts/9}.
27111 @chapter @value{GDBN} Text User Interface
27113 @cindex Text User Interface
27116 * TUI Overview:: TUI overview
27117 * TUI Keys:: TUI key bindings
27118 * TUI Single Key Mode:: TUI single key mode
27119 * TUI Commands:: TUI-specific commands
27120 * TUI Configuration:: TUI configuration variables
27123 The @value{GDBN} Text User Interface (TUI) is a terminal
27124 interface which uses the @code{curses} library to show the source
27125 file, the assembly output, the program registers and @value{GDBN}
27126 commands in separate text windows. The TUI mode is supported only
27127 on platforms where a suitable version of the @code{curses} library
27130 The TUI mode is enabled by default when you invoke @value{GDBN} as
27131 @samp{@value{GDBP} -tui}.
27132 You can also switch in and out of TUI mode while @value{GDBN} runs by
27133 using various TUI commands and key bindings, such as @command{tui
27134 enable} or @kbd{C-x C-a}. @xref{TUI Commands, ,TUI Commands}, and
27135 @ref{TUI Keys, ,TUI Key Bindings}.
27138 @section TUI Overview
27140 In TUI mode, @value{GDBN} can display several text windows:
27144 This window is the @value{GDBN} command window with the @value{GDBN}
27145 prompt and the @value{GDBN} output. The @value{GDBN} input is still
27146 managed using readline.
27149 The source window shows the source file of the program. The current
27150 line and active breakpoints are displayed in this window.
27153 The assembly window shows the disassembly output of the program.
27156 This window shows the processor registers. Registers are highlighted
27157 when their values change.
27160 The source and assembly windows show the current program position
27161 by highlighting the current line and marking it with a @samp{>} marker.
27162 Breakpoints are indicated with two markers. The first marker
27163 indicates the breakpoint type:
27167 Breakpoint which was hit at least once.
27170 Breakpoint which was never hit.
27173 Hardware breakpoint which was hit at least once.
27176 Hardware breakpoint which was never hit.
27179 The second marker indicates whether the breakpoint is enabled or not:
27183 Breakpoint is enabled.
27186 Breakpoint is disabled.
27189 The source, assembly and register windows are updated when the current
27190 thread changes, when the frame changes, or when the program counter
27193 These windows are not all visible at the same time. The command
27194 window is always visible. The others can be arranged in several
27205 source and assembly,
27208 source and registers, or
27211 assembly and registers.
27214 A status line above the command window shows the following information:
27218 Indicates the current @value{GDBN} target.
27219 (@pxref{Targets, ,Specifying a Debugging Target}).
27222 Gives the current process or thread number.
27223 When no process is being debugged, this field is set to @code{No process}.
27226 Gives the current function name for the selected frame.
27227 The name is demangled if demangling is turned on (@pxref{Print Settings}).
27228 When there is no symbol corresponding to the current program counter,
27229 the string @code{??} is displayed.
27232 Indicates the current line number for the selected frame.
27233 When the current line number is not known, the string @code{??} is displayed.
27236 Indicates the current program counter address.
27240 @section TUI Key Bindings
27241 @cindex TUI key bindings
27243 The TUI installs several key bindings in the readline keymaps
27244 @ifset SYSTEM_READLINE
27245 (@pxref{Command Line Editing, , , rluserman, GNU Readline Library}).
27247 @ifclear SYSTEM_READLINE
27248 (@pxref{Command Line Editing}).
27250 The following key bindings are installed for both TUI mode and the
27251 @value{GDBN} standard mode.
27260 Enter or leave the TUI mode. When leaving the TUI mode,
27261 the curses window management stops and @value{GDBN} operates using
27262 its standard mode, writing on the terminal directly. When reentering
27263 the TUI mode, control is given back to the curses windows.
27264 The screen is then refreshed.
27268 Use a TUI layout with only one window. The layout will
27269 either be @samp{source} or @samp{assembly}. When the TUI mode
27270 is not active, it will switch to the TUI mode.
27272 Think of this key binding as the Emacs @kbd{C-x 1} binding.
27276 Use a TUI layout with at least two windows. When the current
27277 layout already has two windows, the next layout with two windows is used.
27278 When a new layout is chosen, one window will always be common to the
27279 previous layout and the new one.
27281 Think of it as the Emacs @kbd{C-x 2} binding.
27285 Change the active window. The TUI associates several key bindings
27286 (like scrolling and arrow keys) with the active window. This command
27287 gives the focus to the next TUI window.
27289 Think of it as the Emacs @kbd{C-x o} binding.
27293 Switch in and out of the TUI SingleKey mode that binds single
27294 keys to @value{GDBN} commands (@pxref{TUI Single Key Mode}).
27297 The following key bindings only work in the TUI mode:
27302 Scroll the active window one page up.
27306 Scroll the active window one page down.
27310 Scroll the active window one line up.
27314 Scroll the active window one line down.
27318 Scroll the active window one column left.
27322 Scroll the active window one column right.
27326 Refresh the screen.
27329 Because the arrow keys scroll the active window in the TUI mode, they
27330 are not available for their normal use by readline unless the command
27331 window has the focus. When another window is active, you must use
27332 other readline key bindings such as @kbd{C-p}, @kbd{C-n}, @kbd{C-b}
27333 and @kbd{C-f} to control the command window.
27335 @node TUI Single Key Mode
27336 @section TUI Single Key Mode
27337 @cindex TUI single key mode
27339 The TUI also provides a @dfn{SingleKey} mode, which binds several
27340 frequently used @value{GDBN} commands to single keys. Type @kbd{C-x s} to
27341 switch into this mode, where the following key bindings are used:
27344 @kindex c @r{(SingleKey TUI key)}
27348 @kindex d @r{(SingleKey TUI key)}
27352 @kindex f @r{(SingleKey TUI key)}
27356 @kindex n @r{(SingleKey TUI key)}
27360 @kindex o @r{(SingleKey TUI key)}
27362 nexti. The shortcut letter @samp{o} stands for ``step Over''.
27364 @kindex q @r{(SingleKey TUI key)}
27366 exit the SingleKey mode.
27368 @kindex r @r{(SingleKey TUI key)}
27372 @kindex s @r{(SingleKey TUI key)}
27376 @kindex i @r{(SingleKey TUI key)}
27378 stepi. The shortcut letter @samp{i} stands for ``step Into''.
27380 @kindex u @r{(SingleKey TUI key)}
27384 @kindex v @r{(SingleKey TUI key)}
27388 @kindex w @r{(SingleKey TUI key)}
27393 Other keys temporarily switch to the @value{GDBN} command prompt.
27394 The key that was pressed is inserted in the editing buffer so that
27395 it is possible to type most @value{GDBN} commands without interaction
27396 with the TUI SingleKey mode. Once the command is entered the TUI
27397 SingleKey mode is restored. The only way to permanently leave
27398 this mode is by typing @kbd{q} or @kbd{C-x s}.
27402 @section TUI-specific Commands
27403 @cindex TUI commands
27405 The TUI has specific commands to control the text windows.
27406 These commands are always available, even when @value{GDBN} is not in
27407 the TUI mode. When @value{GDBN} is in the standard mode, most
27408 of these commands will automatically switch to the TUI mode.
27410 Note that if @value{GDBN}'s @code{stdout} is not connected to a
27411 terminal, or @value{GDBN} has been started with the machine interface
27412 interpreter (@pxref{GDB/MI, ,The @sc{gdb/mi} Interface}), most of
27413 these commands will fail with an error, because it would not be
27414 possible or desirable to enable curses window management.
27419 Activate TUI mode. The last active TUI window layout will be used if
27420 TUI mode has prevsiouly been used in the current debugging session,
27421 otherwise a default layout is used.
27424 @kindex tui disable
27425 Disable TUI mode, returning to the console interpreter.
27429 List and give the size of all displayed windows.
27431 @item layout @var{name}
27433 Changes which TUI windows are displayed. In each layout the command
27434 window is always displayed, the @var{name} parameter controls which
27435 additional windows are displayed, and can be any of the following:
27439 Display the next layout.
27442 Display the previous layout.
27445 Display the source and command windows.
27448 Display the assembly and command windows.
27451 Display the source, assembly, and command windows.
27454 When in @code{src} layout display the register, source, and command
27455 windows. When in @code{asm} or @code{split} layout display the
27456 register, assembler, and command windows.
27459 @item focus @var{name}
27461 Changes which TUI window is currently active for scrolling. The
27462 @var{name} parameter can be any of the following:
27466 Make the next window active for scrolling.
27469 Make the previous window active for scrolling.
27472 Make the source window active for scrolling.
27475 Make the assembly window active for scrolling.
27478 Make the register window active for scrolling.
27481 Make the command window active for scrolling.
27486 Refresh the screen. This is similar to typing @kbd{C-L}.
27488 @item tui reg @var{group}
27490 Changes the register group displayed in the tui register window to
27491 @var{group}. If the register window is not currently displayed this
27492 command will cause the register window to be displayed. The list of
27493 register groups, as well as their order is target specific. The
27494 following groups are available on most targets:
27497 Repeatedly selecting this group will cause the display to cycle
27498 through all of the available register groups.
27501 Repeatedly selecting this group will cause the display to cycle
27502 through all of the available register groups in the reverse order to
27506 Display the general registers.
27508 Display the floating point registers.
27510 Display the system registers.
27512 Display the vector registers.
27514 Display all registers.
27519 Update the source window and the current execution point.
27521 @item winheight @var{name} +@var{count}
27522 @itemx winheight @var{name} -@var{count}
27524 Change the height of the window @var{name} by @var{count}
27525 lines. Positive counts increase the height, while negative counts
27526 decrease it. The @var{name} parameter can be one of @code{src} (the
27527 source window), @code{cmd} (the command window), @code{asm} (the
27528 disassembly window), or @code{regs} (the register display window).
27531 @node TUI Configuration
27532 @section TUI Configuration Variables
27533 @cindex TUI configuration variables
27535 Several configuration variables control the appearance of TUI windows.
27538 @item set tui border-kind @var{kind}
27539 @kindex set tui border-kind
27540 Select the border appearance for the source, assembly and register windows.
27541 The possible values are the following:
27544 Use a space character to draw the border.
27547 Use @sc{ascii} characters @samp{+}, @samp{-} and @samp{|} to draw the border.
27550 Use the Alternate Character Set to draw the border. The border is
27551 drawn using character line graphics if the terminal supports them.
27554 @item set tui border-mode @var{mode}
27555 @kindex set tui border-mode
27556 @itemx set tui active-border-mode @var{mode}
27557 @kindex set tui active-border-mode
27558 Select the display attributes for the borders of the inactive windows
27559 or the active window. The @var{mode} can be one of the following:
27562 Use normal attributes to display the border.
27568 Use reverse video mode.
27571 Use half bright mode.
27573 @item half-standout
27574 Use half bright and standout mode.
27577 Use extra bright or bold mode.
27579 @item bold-standout
27580 Use extra bright or bold and standout mode.
27583 @item set tui tab-width @var{nchars}
27584 @kindex set tui tab-width
27586 Set the width of tab stops to be @var{nchars} characters. This
27587 setting affects the display of TAB characters in the source and
27592 @chapter Using @value{GDBN} under @sc{gnu} Emacs
27595 @cindex @sc{gnu} Emacs
27596 A special interface allows you to use @sc{gnu} Emacs to view (and
27597 edit) the source files for the program you are debugging with
27600 To use this interface, use the command @kbd{M-x gdb} in Emacs. Give the
27601 executable file you want to debug as an argument. This command starts
27602 @value{GDBN} as a subprocess of Emacs, with input and output through a newly
27603 created Emacs buffer.
27604 @c (Do not use the @code{-tui} option to run @value{GDBN} from Emacs.)
27606 Running @value{GDBN} under Emacs can be just like running @value{GDBN} normally except for two
27611 All ``terminal'' input and output goes through an Emacs buffer, called
27614 This applies both to @value{GDBN} commands and their output, and to the input
27615 and output done by the program you are debugging.
27617 This is useful because it means that you can copy the text of previous
27618 commands and input them again; you can even use parts of the output
27621 All the facilities of Emacs' Shell mode are available for interacting
27622 with your program. In particular, you can send signals the usual
27623 way---for example, @kbd{C-c C-c} for an interrupt, @kbd{C-c C-z} for a
27627 @value{GDBN} displays source code through Emacs.
27629 Each time @value{GDBN} displays a stack frame, Emacs automatically finds the
27630 source file for that frame and puts an arrow (@samp{=>}) at the
27631 left margin of the current line. Emacs uses a separate buffer for
27632 source display, and splits the screen to show both your @value{GDBN} session
27635 Explicit @value{GDBN} @code{list} or search commands still produce output as
27636 usual, but you probably have no reason to use them from Emacs.
27639 We call this @dfn{text command mode}. Emacs 22.1, and later, also uses
27640 a graphical mode, enabled by default, which provides further buffers
27641 that can control the execution and describe the state of your program.
27642 @xref{GDB Graphical Interface,,, Emacs, The @sc{gnu} Emacs Manual}.
27644 If you specify an absolute file name when prompted for the @kbd{M-x
27645 gdb} argument, then Emacs sets your current working directory to where
27646 your program resides. If you only specify the file name, then Emacs
27647 sets your current working directory to the directory associated
27648 with the previous buffer. In this case, @value{GDBN} may find your
27649 program by searching your environment's @code{PATH} variable, but on
27650 some operating systems it might not find the source. So, although the
27651 @value{GDBN} input and output session proceeds normally, the auxiliary
27652 buffer does not display the current source and line of execution.
27654 The initial working directory of @value{GDBN} is printed on the top
27655 line of the GUD buffer and this serves as a default for the commands
27656 that specify files for @value{GDBN} to operate on. @xref{Files,
27657 ,Commands to Specify Files}.
27659 By default, @kbd{M-x gdb} calls the program called @file{gdb}. If you
27660 need to call @value{GDBN} by a different name (for example, if you
27661 keep several configurations around, with different names) you can
27662 customize the Emacs variable @code{gud-gdb-command-name} to run the
27665 In the GUD buffer, you can use these special Emacs commands in
27666 addition to the standard Shell mode commands:
27670 Describe the features of Emacs' GUD Mode.
27673 Execute to another source line, like the @value{GDBN} @code{step} command; also
27674 update the display window to show the current file and location.
27677 Execute to next source line in this function, skipping all function
27678 calls, like the @value{GDBN} @code{next} command. Then update the display window
27679 to show the current file and location.
27682 Execute one instruction, like the @value{GDBN} @code{stepi} command; update
27683 display window accordingly.
27686 Execute until exit from the selected stack frame, like the @value{GDBN}
27687 @code{finish} command.
27690 Continue execution of your program, like the @value{GDBN} @code{continue}
27694 Go up the number of frames indicated by the numeric argument
27695 (@pxref{Arguments, , Numeric Arguments, Emacs, The @sc{gnu} Emacs Manual}),
27696 like the @value{GDBN} @code{up} command.
27699 Go down the number of frames indicated by the numeric argument, like the
27700 @value{GDBN} @code{down} command.
27703 In any source file, the Emacs command @kbd{C-x @key{SPC}} (@code{gud-break})
27704 tells @value{GDBN} to set a breakpoint on the source line point is on.
27706 In text command mode, if you type @kbd{M-x speedbar}, Emacs displays a
27707 separate frame which shows a backtrace when the GUD buffer is current.
27708 Move point to any frame in the stack and type @key{RET} to make it
27709 become the current frame and display the associated source in the
27710 source buffer. Alternatively, click @kbd{Mouse-2} to make the
27711 selected frame become the current one. In graphical mode, the
27712 speedbar displays watch expressions.
27714 If you accidentally delete the source-display buffer, an easy way to get
27715 it back is to type the command @code{f} in the @value{GDBN} buffer, to
27716 request a frame display; when you run under Emacs, this recreates
27717 the source buffer if necessary to show you the context of the current
27720 The source files displayed in Emacs are in ordinary Emacs buffers
27721 which are visiting the source files in the usual way. You can edit
27722 the files with these buffers if you wish; but keep in mind that @value{GDBN}
27723 communicates with Emacs in terms of line numbers. If you add or
27724 delete lines from the text, the line numbers that @value{GDBN} knows cease
27725 to correspond properly with the code.
27727 A more detailed description of Emacs' interaction with @value{GDBN} is
27728 given in the Emacs manual (@pxref{Debuggers,,, Emacs, The @sc{gnu}
27732 @chapter The @sc{gdb/mi} Interface
27734 @unnumberedsec Function and Purpose
27736 @cindex @sc{gdb/mi}, its purpose
27737 @sc{gdb/mi} is a line based machine oriented text interface to
27738 @value{GDBN} and is activated by specifying using the
27739 @option{--interpreter} command line option (@pxref{Mode Options}). It
27740 is specifically intended to support the development of systems which
27741 use the debugger as just one small component of a larger system.
27743 This chapter is a specification of the @sc{gdb/mi} interface. It is written
27744 in the form of a reference manual.
27746 Note that @sc{gdb/mi} is still under construction, so some of the
27747 features described below are incomplete and subject to change
27748 (@pxref{GDB/MI Development and Front Ends, , @sc{gdb/mi} Development and Front Ends}).
27750 @unnumberedsec Notation and Terminology
27752 @cindex notational conventions, for @sc{gdb/mi}
27753 This chapter uses the following notation:
27757 @code{|} separates two alternatives.
27760 @code{[ @var{something} ]} indicates that @var{something} is optional:
27761 it may or may not be given.
27764 @code{( @var{group} )*} means that @var{group} inside the parentheses
27765 may repeat zero or more times.
27768 @code{( @var{group} )+} means that @var{group} inside the parentheses
27769 may repeat one or more times.
27772 @code{"@var{string}"} means a literal @var{string}.
27776 @heading Dependencies
27780 * GDB/MI General Design::
27781 * GDB/MI Command Syntax::
27782 * GDB/MI Compatibility with CLI::
27783 * GDB/MI Development and Front Ends::
27784 * GDB/MI Output Records::
27785 * GDB/MI Simple Examples::
27786 * GDB/MI Command Description Format::
27787 * GDB/MI Breakpoint Commands::
27788 * GDB/MI Catchpoint Commands::
27789 * GDB/MI Program Context::
27790 * GDB/MI Thread Commands::
27791 * GDB/MI Ada Tasking Commands::
27792 * GDB/MI Program Execution::
27793 * GDB/MI Stack Manipulation::
27794 * GDB/MI Variable Objects::
27795 * GDB/MI Data Manipulation::
27796 * GDB/MI Tracepoint Commands::
27797 * GDB/MI Symbol Query::
27798 * GDB/MI File Commands::
27800 * GDB/MI Kod Commands::
27801 * GDB/MI Memory Overlay Commands::
27802 * GDB/MI Signal Handling Commands::
27804 * GDB/MI Target Manipulation::
27805 * GDB/MI File Transfer Commands::
27806 * GDB/MI Ada Exceptions Commands::
27807 * GDB/MI Support Commands::
27808 * GDB/MI Miscellaneous Commands::
27811 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
27812 @node GDB/MI General Design
27813 @section @sc{gdb/mi} General Design
27814 @cindex GDB/MI General Design
27816 Interaction of a @sc{GDB/MI} frontend with @value{GDBN} involves three
27817 parts---commands sent to @value{GDBN}, responses to those commands
27818 and notifications. Each command results in exactly one response,
27819 indicating either successful completion of the command, or an error.
27820 For the commands that do not resume the target, the response contains the
27821 requested information. For the commands that resume the target, the
27822 response only indicates whether the target was successfully resumed.
27823 Notifications is the mechanism for reporting changes in the state of the
27824 target, or in @value{GDBN} state, that cannot conveniently be associated with
27825 a command and reported as part of that command response.
27827 The important examples of notifications are:
27831 Exec notifications. These are used to report changes in
27832 target state---when a target is resumed, or stopped. It would not
27833 be feasible to include this information in response of resuming
27834 commands, because one resume commands can result in multiple events in
27835 different threads. Also, quite some time may pass before any event
27836 happens in the target, while a frontend needs to know whether the resuming
27837 command itself was successfully executed.
27840 Console output, and status notifications. Console output
27841 notifications are used to report output of CLI commands, as well as
27842 diagnostics for other commands. Status notifications are used to
27843 report the progress of a long-running operation. Naturally, including
27844 this information in command response would mean no output is produced
27845 until the command is finished, which is undesirable.
27848 General notifications. Commands may have various side effects on
27849 the @value{GDBN} or target state beyond their official purpose. For example,
27850 a command may change the selected thread. Although such changes can
27851 be included in command response, using notification allows for more
27852 orthogonal frontend design.
27856 There's no guarantee that whenever an MI command reports an error,
27857 @value{GDBN} or the target are in any specific state, and especially,
27858 the state is not reverted to the state before the MI command was
27859 processed. Therefore, whenever an MI command results in an error,
27860 we recommend that the frontend refreshes all the information shown in
27861 the user interface.
27865 * Context management::
27866 * Asynchronous and non-stop modes::
27870 @node Context management
27871 @subsection Context management
27873 @subsubsection Threads and Frames
27875 In most cases when @value{GDBN} accesses the target, this access is
27876 done in context of a specific thread and frame (@pxref{Frames}).
27877 Often, even when accessing global data, the target requires that a thread
27878 be specified. The CLI interface maintains the selected thread and frame,
27879 and supplies them to target on each command. This is convenient,
27880 because a command line user would not want to specify that information
27881 explicitly on each command, and because user interacts with
27882 @value{GDBN} via a single terminal, so no confusion is possible as
27883 to what thread and frame are the current ones.
27885 In the case of MI, the concept of selected thread and frame is less
27886 useful. First, a frontend can easily remember this information
27887 itself. Second, a graphical frontend can have more than one window,
27888 each one used for debugging a different thread, and the frontend might
27889 want to access additional threads for internal purposes. This
27890 increases the risk that by relying on implicitly selected thread, the
27891 frontend may be operating on a wrong one. Therefore, each MI command
27892 should explicitly specify which thread and frame to operate on. To
27893 make it possible, each MI command accepts the @samp{--thread} and
27894 @samp{--frame} options, the value to each is @value{GDBN} global
27895 identifier for thread and frame to operate on.
27897 Usually, each top-level window in a frontend allows the user to select
27898 a thread and a frame, and remembers the user selection for further
27899 operations. However, in some cases @value{GDBN} may suggest that the
27900 current thread or frame be changed. For example, when stopping on a
27901 breakpoint it is reasonable to switch to the thread where breakpoint is
27902 hit. For another example, if the user issues the CLI @samp{thread} or
27903 @samp{frame} commands via the frontend, it is desirable to change the
27904 frontend's selection to the one specified by user. @value{GDBN}
27905 communicates the suggestion to change current thread and frame using the
27906 @samp{=thread-selected} notification.
27908 Note that historically, MI shares the selected thread with CLI, so
27909 frontends used the @code{-thread-select} to execute commands in the
27910 right context. However, getting this to work right is cumbersome. The
27911 simplest way is for frontend to emit @code{-thread-select} command
27912 before every command. This doubles the number of commands that need
27913 to be sent. The alternative approach is to suppress @code{-thread-select}
27914 if the selected thread in @value{GDBN} is supposed to be identical to the
27915 thread the frontend wants to operate on. However, getting this
27916 optimization right can be tricky. In particular, if the frontend
27917 sends several commands to @value{GDBN}, and one of the commands changes the
27918 selected thread, then the behaviour of subsequent commands will
27919 change. So, a frontend should either wait for response from such
27920 problematic commands, or explicitly add @code{-thread-select} for
27921 all subsequent commands. No frontend is known to do this exactly
27922 right, so it is suggested to just always pass the @samp{--thread} and
27923 @samp{--frame} options.
27925 @subsubsection Language
27927 The execution of several commands depends on which language is selected.
27928 By default, the current language (@pxref{show language}) is used.
27929 But for commands known to be language-sensitive, it is recommended
27930 to use the @samp{--language} option. This option takes one argument,
27931 which is the name of the language to use while executing the command.
27935 -data-evaluate-expression --language c "sizeof (void*)"
27940 The valid language names are the same names accepted by the
27941 @samp{set language} command (@pxref{Manually}), excluding @samp{auto},
27942 @samp{local} or @samp{unknown}.
27944 @node Asynchronous and non-stop modes
27945 @subsection Asynchronous command execution and non-stop mode
27947 On some targets, @value{GDBN} is capable of processing MI commands
27948 even while the target is running. This is called @dfn{asynchronous
27949 command execution} (@pxref{Background Execution}). The frontend may
27950 specify a preferrence for asynchronous execution using the
27951 @code{-gdb-set mi-async 1} command, which should be emitted before
27952 either running the executable or attaching to the target. After the
27953 frontend has started the executable or attached to the target, it can
27954 find if asynchronous execution is enabled using the
27955 @code{-list-target-features} command.
27958 @item -gdb-set mi-async on
27959 @item -gdb-set mi-async off
27960 Set whether MI is in asynchronous mode.
27962 When @code{off}, which is the default, MI execution commands (e.g.,
27963 @code{-exec-continue}) are foreground commands, and @value{GDBN} waits
27964 for the program to stop before processing further commands.
27966 When @code{on}, MI execution commands are background execution
27967 commands (e.g., @code{-exec-continue} becomes the equivalent of the
27968 @code{c&} CLI command), and so @value{GDBN} is capable of processing
27969 MI commands even while the target is running.
27971 @item -gdb-show mi-async
27972 Show whether MI asynchronous mode is enabled.
27975 Note: In @value{GDBN} version 7.7 and earlier, this option was called
27976 @code{target-async} instead of @code{mi-async}, and it had the effect
27977 of both putting MI in asynchronous mode and making CLI background
27978 commands possible. CLI background commands are now always possible
27979 ``out of the box'' if the target supports them. The old spelling is
27980 kept as a deprecated alias for backwards compatibility.
27982 Even if @value{GDBN} can accept a command while target is running,
27983 many commands that access the target do not work when the target is
27984 running. Therefore, asynchronous command execution is most useful
27985 when combined with non-stop mode (@pxref{Non-Stop Mode}). Then,
27986 it is possible to examine the state of one thread, while other threads
27989 When a given thread is running, MI commands that try to access the
27990 target in the context of that thread may not work, or may work only on
27991 some targets. In particular, commands that try to operate on thread's
27992 stack will not work, on any target. Commands that read memory, or
27993 modify breakpoints, may work or not work, depending on the target. Note
27994 that even commands that operate on global state, such as @code{print},
27995 @code{set}, and breakpoint commands, still access the target in the
27996 context of a specific thread, so frontend should try to find a
27997 stopped thread and perform the operation on that thread (using the
27998 @samp{--thread} option).
28000 Which commands will work in the context of a running thread is
28001 highly target dependent. However, the two commands
28002 @code{-exec-interrupt}, to stop a thread, and @code{-thread-info},
28003 to find the state of a thread, will always work.
28005 @node Thread groups
28006 @subsection Thread groups
28007 @value{GDBN} may be used to debug several processes at the same time.
28008 On some platfroms, @value{GDBN} may support debugging of several
28009 hardware systems, each one having several cores with several different
28010 processes running on each core. This section describes the MI
28011 mechanism to support such debugging scenarios.
28013 The key observation is that regardless of the structure of the
28014 target, MI can have a global list of threads, because most commands that
28015 accept the @samp{--thread} option do not need to know what process that
28016 thread belongs to. Therefore, it is not necessary to introduce
28017 neither additional @samp{--process} option, nor an notion of the
28018 current process in the MI interface. The only strictly new feature
28019 that is required is the ability to find how the threads are grouped
28022 To allow the user to discover such grouping, and to support arbitrary
28023 hierarchy of machines/cores/processes, MI introduces the concept of a
28024 @dfn{thread group}. Thread group is a collection of threads and other
28025 thread groups. A thread group always has a string identifier, a type,
28026 and may have additional attributes specific to the type. A new
28027 command, @code{-list-thread-groups}, returns the list of top-level
28028 thread groups, which correspond to processes that @value{GDBN} is
28029 debugging at the moment. By passing an identifier of a thread group
28030 to the @code{-list-thread-groups} command, it is possible to obtain
28031 the members of specific thread group.
28033 To allow the user to easily discover processes, and other objects, he
28034 wishes to debug, a concept of @dfn{available thread group} is
28035 introduced. Available thread group is an thread group that
28036 @value{GDBN} is not debugging, but that can be attached to, using the
28037 @code{-target-attach} command. The list of available top-level thread
28038 groups can be obtained using @samp{-list-thread-groups --available}.
28039 In general, the content of a thread group may be only retrieved only
28040 after attaching to that thread group.
28042 Thread groups are related to inferiors (@pxref{Inferiors and
28043 Programs}). Each inferior corresponds to a thread group of a special
28044 type @samp{process}, and some additional operations are permitted on
28045 such thread groups.
28047 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28048 @node GDB/MI Command Syntax
28049 @section @sc{gdb/mi} Command Syntax
28052 * GDB/MI Input Syntax::
28053 * GDB/MI Output Syntax::
28056 @node GDB/MI Input Syntax
28057 @subsection @sc{gdb/mi} Input Syntax
28059 @cindex input syntax for @sc{gdb/mi}
28060 @cindex @sc{gdb/mi}, input syntax
28062 @item @var{command} @expansion{}
28063 @code{@var{cli-command} | @var{mi-command}}
28065 @item @var{cli-command} @expansion{}
28066 @code{[ @var{token} ] @var{cli-command} @var{nl}}, where
28067 @var{cli-command} is any existing @value{GDBN} CLI command.
28069 @item @var{mi-command} @expansion{}
28070 @code{[ @var{token} ] "-" @var{operation} ( " " @var{option} )*
28071 @code{[} " --" @code{]} ( " " @var{parameter} )* @var{nl}}
28073 @item @var{token} @expansion{}
28074 "any sequence of digits"
28076 @item @var{option} @expansion{}
28077 @code{"-" @var{parameter} [ " " @var{parameter} ]}
28079 @item @var{parameter} @expansion{}
28080 @code{@var{non-blank-sequence} | @var{c-string}}
28082 @item @var{operation} @expansion{}
28083 @emph{any of the operations described in this chapter}
28085 @item @var{non-blank-sequence} @expansion{}
28086 @emph{anything, provided it doesn't contain special characters such as
28087 "-", @var{nl}, """ and of course " "}
28089 @item @var{c-string} @expansion{}
28090 @code{""" @var{seven-bit-iso-c-string-content} """}
28092 @item @var{nl} @expansion{}
28101 The CLI commands are still handled by the @sc{mi} interpreter; their
28102 output is described below.
28105 The @code{@var{token}}, when present, is passed back when the command
28109 Some @sc{mi} commands accept optional arguments as part of the parameter
28110 list. Each option is identified by a leading @samp{-} (dash) and may be
28111 followed by an optional argument parameter. Options occur first in the
28112 parameter list and can be delimited from normal parameters using
28113 @samp{--} (this is useful when some parameters begin with a dash).
28120 We want easy access to the existing CLI syntax (for debugging).
28123 We want it to be easy to spot a @sc{mi} operation.
28126 @node GDB/MI Output Syntax
28127 @subsection @sc{gdb/mi} Output Syntax
28129 @cindex output syntax of @sc{gdb/mi}
28130 @cindex @sc{gdb/mi}, output syntax
28131 The output from @sc{gdb/mi} consists of zero or more out-of-band records
28132 followed, optionally, by a single result record. This result record
28133 is for the most recent command. The sequence of output records is
28134 terminated by @samp{(gdb)}.
28136 If an input command was prefixed with a @code{@var{token}} then the
28137 corresponding output for that command will also be prefixed by that same
28141 @item @var{output} @expansion{}
28142 @code{( @var{out-of-band-record} )* [ @var{result-record} ] "(gdb)" @var{nl}}
28144 @item @var{result-record} @expansion{}
28145 @code{ [ @var{token} ] "^" @var{result-class} ( "," @var{result} )* @var{nl}}
28147 @item @var{out-of-band-record} @expansion{}
28148 @code{@var{async-record} | @var{stream-record}}
28150 @item @var{async-record} @expansion{}
28151 @code{@var{exec-async-output} | @var{status-async-output} | @var{notify-async-output}}
28153 @item @var{exec-async-output} @expansion{}
28154 @code{[ @var{token} ] "*" @var{async-output nl}}
28156 @item @var{status-async-output} @expansion{}
28157 @code{[ @var{token} ] "+" @var{async-output nl}}
28159 @item @var{notify-async-output} @expansion{}
28160 @code{[ @var{token} ] "=" @var{async-output nl}}
28162 @item @var{async-output} @expansion{}
28163 @code{@var{async-class} ( "," @var{result} )*}
28165 @item @var{result-class} @expansion{}
28166 @code{"done" | "running" | "connected" | "error" | "exit"}
28168 @item @var{async-class} @expansion{}
28169 @code{"stopped" | @var{others}} (where @var{others} will be added
28170 depending on the needs---this is still in development).
28172 @item @var{result} @expansion{}
28173 @code{ @var{variable} "=" @var{value}}
28175 @item @var{variable} @expansion{}
28176 @code{ @var{string} }
28178 @item @var{value} @expansion{}
28179 @code{ @var{const} | @var{tuple} | @var{list} }
28181 @item @var{const} @expansion{}
28182 @code{@var{c-string}}
28184 @item @var{tuple} @expansion{}
28185 @code{ "@{@}" | "@{" @var{result} ( "," @var{result} )* "@}" }
28187 @item @var{list} @expansion{}
28188 @code{ "[]" | "[" @var{value} ( "," @var{value} )* "]" | "["
28189 @var{result} ( "," @var{result} )* "]" }
28191 @item @var{stream-record} @expansion{}
28192 @code{@var{console-stream-output} | @var{target-stream-output} | @var{log-stream-output}}
28194 @item @var{console-stream-output} @expansion{}
28195 @code{"~" @var{c-string nl}}
28197 @item @var{target-stream-output} @expansion{}
28198 @code{"@@" @var{c-string nl}}
28200 @item @var{log-stream-output} @expansion{}
28201 @code{"&" @var{c-string nl}}
28203 @item @var{nl} @expansion{}
28206 @item @var{token} @expansion{}
28207 @emph{any sequence of digits}.
28215 All output sequences end in a single line containing a period.
28218 The @code{@var{token}} is from the corresponding request. Note that
28219 for all async output, while the token is allowed by the grammar and
28220 may be output by future versions of @value{GDBN} for select async
28221 output messages, it is generally omitted. Frontends should treat
28222 all async output as reporting general changes in the state of the
28223 target and there should be no need to associate async output to any
28227 @cindex status output in @sc{gdb/mi}
28228 @var{status-async-output} contains on-going status information about the
28229 progress of a slow operation. It can be discarded. All status output is
28230 prefixed by @samp{+}.
28233 @cindex async output in @sc{gdb/mi}
28234 @var{exec-async-output} contains asynchronous state change on the target
28235 (stopped, started, disappeared). All async output is prefixed by
28239 @cindex notify output in @sc{gdb/mi}
28240 @var{notify-async-output} contains supplementary information that the
28241 client should handle (e.g., a new breakpoint information). All notify
28242 output is prefixed by @samp{=}.
28245 @cindex console output in @sc{gdb/mi}
28246 @var{console-stream-output} is output that should be displayed as is in the
28247 console. It is the textual response to a CLI command. All the console
28248 output is prefixed by @samp{~}.
28251 @cindex target output in @sc{gdb/mi}
28252 @var{target-stream-output} is the output produced by the target program.
28253 All the target output is prefixed by @samp{@@}.
28256 @cindex log output in @sc{gdb/mi}
28257 @var{log-stream-output} is output text coming from @value{GDBN}'s internals, for
28258 instance messages that should be displayed as part of an error log. All
28259 the log output is prefixed by @samp{&}.
28262 @cindex list output in @sc{gdb/mi}
28263 New @sc{gdb/mi} commands should only output @var{lists} containing
28269 @xref{GDB/MI Stream Records, , @sc{gdb/mi} Stream Records}, for more
28270 details about the various output records.
28272 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28273 @node GDB/MI Compatibility with CLI
28274 @section @sc{gdb/mi} Compatibility with CLI
28276 @cindex compatibility, @sc{gdb/mi} and CLI
28277 @cindex @sc{gdb/mi}, compatibility with CLI
28279 For the developers convenience CLI commands can be entered directly,
28280 but there may be some unexpected behaviour. For example, commands
28281 that query the user will behave as if the user replied yes, breakpoint
28282 command lists are not executed and some CLI commands, such as
28283 @code{if}, @code{when} and @code{define}, prompt for further input with
28284 @samp{>}, which is not valid MI output.
28286 This feature may be removed at some stage in the future and it is
28287 recommended that front ends use the @code{-interpreter-exec} command
28288 (@pxref{-interpreter-exec}).
28290 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28291 @node GDB/MI Development and Front Ends
28292 @section @sc{gdb/mi} Development and Front Ends
28293 @cindex @sc{gdb/mi} development
28295 The application which takes the MI output and presents the state of the
28296 program being debugged to the user is called a @dfn{front end}.
28298 Since @sc{gdb/mi} is used by a variety of front ends to @value{GDBN}, changes
28299 to the MI interface may break existing usage. This section describes how the
28300 protocol changes and how to request previous version of the protocol when it
28303 Some changes in MI need not break a carefully designed front end, and
28304 for these the MI version will remain unchanged. The following is a
28305 list of changes that may occur within one level, so front ends should
28306 parse MI output in a way that can handle them:
28310 New MI commands may be added.
28313 New fields may be added to the output of any MI command.
28316 The range of values for fields with specified values, e.g.,
28317 @code{in_scope} (@pxref{-var-update}) may be extended.
28319 @c The format of field's content e.g type prefix, may change so parse it
28320 @c at your own risk. Yes, in general?
28322 @c The order of fields may change? Shouldn't really matter but it might
28323 @c resolve inconsistencies.
28326 If the changes are likely to break front ends, the MI version level
28327 will be increased by one. The new versions of the MI protocol are not compatible
28328 with the old versions. Old versions of MI remain available, allowing front ends
28329 to keep using them until they are modified to use the latest MI version.
28331 Since @code{--interpreter=mi} always points to the latest MI version, it is
28332 recommended that front ends request a specific version of MI when launching
28333 @value{GDBN} (e.g. @code{--interpreter=mi2}) to make sure they get an
28334 interpreter with the MI version they expect.
28336 The following table gives a summary of the the released versions of the MI
28337 interface: the version number, the version of GDB in which it first appeared
28338 and the breaking changes compared to the previous version.
28340 @multitable @columnfractions .05 .05 .9
28341 @headitem MI version @tab GDB version @tab Breaking changes
28358 The @code{-environment-pwd}, @code{-environment-directory} and
28359 @code{-environment-path} commands now returns values using the MI output
28360 syntax, rather than CLI output syntax.
28363 @code{-var-list-children}'s @code{children} result field is now a list, rather
28367 @code{-var-update}'s @code{changelist} result field is now a list, rather than
28379 The output of information about multi-location breakpoints has changed in the
28380 responses to the @code{-break-insert} and @code{-break-info} commands, as well
28381 as in the @code{=breakpoint-created} and @code{=breakpoint-modified} events.
28382 The multiple locations are now placed in a @code{locations} field, whose value
28388 If your front end cannot yet migrate to a more recent version of the
28389 MI protocol, you can nevertheless selectively enable specific features
28390 available in those recent MI versions, using the following commands:
28394 @item -fix-multi-location-breakpoint-output
28395 Use the output for multi-location breakpoints which was introduced by
28396 MI 3, even when using MI versions 2 or 1. This command has no
28397 effect when using MI version 3 or later.
28401 The best way to avoid unexpected changes in MI that might break your front
28402 end is to make your project known to @value{GDBN} developers and
28403 follow development on @email{gdb@@sourceware.org} and
28404 @email{gdb-patches@@sourceware.org}.
28405 @cindex mailing lists
28407 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28408 @node GDB/MI Output Records
28409 @section @sc{gdb/mi} Output Records
28412 * GDB/MI Result Records::
28413 * GDB/MI Stream Records::
28414 * GDB/MI Async Records::
28415 * GDB/MI Breakpoint Information::
28416 * GDB/MI Frame Information::
28417 * GDB/MI Thread Information::
28418 * GDB/MI Ada Exception Information::
28421 @node GDB/MI Result Records
28422 @subsection @sc{gdb/mi} Result Records
28424 @cindex result records in @sc{gdb/mi}
28425 @cindex @sc{gdb/mi}, result records
28426 In addition to a number of out-of-band notifications, the response to a
28427 @sc{gdb/mi} command includes one of the following result indications:
28431 @item "^done" [ "," @var{results} ]
28432 The synchronous operation was successful, @code{@var{results}} are the return
28437 This result record is equivalent to @samp{^done}. Historically, it
28438 was output instead of @samp{^done} if the command has resumed the
28439 target. This behaviour is maintained for backward compatibility, but
28440 all frontends should treat @samp{^done} and @samp{^running}
28441 identically and rely on the @samp{*running} output record to determine
28442 which threads are resumed.
28446 @value{GDBN} has connected to a remote target.
28448 @item "^error" "," "msg=" @var{c-string} [ "," "code=" @var{c-string} ]
28450 The operation failed. The @code{msg=@var{c-string}} variable contains
28451 the corresponding error message.
28453 If present, the @code{code=@var{c-string}} variable provides an error
28454 code on which consumers can rely on to detect the corresponding
28455 error condition. At present, only one error code is defined:
28458 @item "undefined-command"
28459 Indicates that the command causing the error does not exist.
28464 @value{GDBN} has terminated.
28468 @node GDB/MI Stream Records
28469 @subsection @sc{gdb/mi} Stream Records
28471 @cindex @sc{gdb/mi}, stream records
28472 @cindex stream records in @sc{gdb/mi}
28473 @value{GDBN} internally maintains a number of output streams: the console, the
28474 target, and the log. The output intended for each of these streams is
28475 funneled through the @sc{gdb/mi} interface using @dfn{stream records}.
28477 Each stream record begins with a unique @dfn{prefix character} which
28478 identifies its stream (@pxref{GDB/MI Output Syntax, , @sc{gdb/mi} Output
28479 Syntax}). In addition to the prefix, each stream record contains a
28480 @code{@var{string-output}}. This is either raw text (with an implicit new
28481 line) or a quoted C string (which does not contain an implicit newline).
28484 @item "~" @var{string-output}
28485 The console output stream contains text that should be displayed in the
28486 CLI console window. It contains the textual responses to CLI commands.
28488 @item "@@" @var{string-output}
28489 The target output stream contains any textual output from the running
28490 target. This is only present when GDB's event loop is truly
28491 asynchronous, which is currently only the case for remote targets.
28493 @item "&" @var{string-output}
28494 The log stream contains debugging messages being produced by @value{GDBN}'s
28498 @node GDB/MI Async Records
28499 @subsection @sc{gdb/mi} Async Records
28501 @cindex async records in @sc{gdb/mi}
28502 @cindex @sc{gdb/mi}, async records
28503 @dfn{Async} records are used to notify the @sc{gdb/mi} client of
28504 additional changes that have occurred. Those changes can either be a
28505 consequence of @sc{gdb/mi} commands (e.g., a breakpoint modified) or a result of
28506 target activity (e.g., target stopped).
28508 The following is the list of possible async records:
28512 @item *running,thread-id="@var{thread}"
28513 The target is now running. The @var{thread} field can be the global
28514 thread ID of the the thread that is now running, and it can be
28515 @samp{all} if all threads are running. The frontend should assume
28516 that no interaction with a running thread is possible after this
28517 notification is produced. The frontend should not assume that this
28518 notification is output only once for any command. @value{GDBN} may
28519 emit this notification several times, either for different threads,
28520 because it cannot resume all threads together, or even for a single
28521 thread, if the thread must be stepped though some code before letting
28524 @item *stopped,reason="@var{reason}",thread-id="@var{id}",stopped-threads="@var{stopped}",core="@var{core}"
28525 The target has stopped. The @var{reason} field can have one of the
28529 @item breakpoint-hit
28530 A breakpoint was reached.
28531 @item watchpoint-trigger
28532 A watchpoint was triggered.
28533 @item read-watchpoint-trigger
28534 A read watchpoint was triggered.
28535 @item access-watchpoint-trigger
28536 An access watchpoint was triggered.
28537 @item function-finished
28538 An -exec-finish or similar CLI command was accomplished.
28539 @item location-reached
28540 An -exec-until or similar CLI command was accomplished.
28541 @item watchpoint-scope
28542 A watchpoint has gone out of scope.
28543 @item end-stepping-range
28544 An -exec-next, -exec-next-instruction, -exec-step, -exec-step-instruction or
28545 similar CLI command was accomplished.
28546 @item exited-signalled
28547 The inferior exited because of a signal.
28549 The inferior exited.
28550 @item exited-normally
28551 The inferior exited normally.
28552 @item signal-received
28553 A signal was received by the inferior.
28555 The inferior has stopped due to a library being loaded or unloaded.
28556 This can happen when @code{stop-on-solib-events} (@pxref{Files}) is
28557 set or when a @code{catch load} or @code{catch unload} catchpoint is
28558 in use (@pxref{Set Catchpoints}).
28560 The inferior has forked. This is reported when @code{catch fork}
28561 (@pxref{Set Catchpoints}) has been used.
28563 The inferior has vforked. This is reported in when @code{catch vfork}
28564 (@pxref{Set Catchpoints}) has been used.
28565 @item syscall-entry
28566 The inferior entered a system call. This is reported when @code{catch
28567 syscall} (@pxref{Set Catchpoints}) has been used.
28568 @item syscall-return
28569 The inferior returned from a system call. This is reported when
28570 @code{catch syscall} (@pxref{Set Catchpoints}) has been used.
28572 The inferior called @code{exec}. This is reported when @code{catch exec}
28573 (@pxref{Set Catchpoints}) has been used.
28576 The @var{id} field identifies the global thread ID of the thread
28577 that directly caused the stop -- for example by hitting a breakpoint.
28578 Depending on whether all-stop
28579 mode is in effect (@pxref{All-Stop Mode}), @value{GDBN} may either
28580 stop all threads, or only the thread that directly triggered the stop.
28581 If all threads are stopped, the @var{stopped} field will have the
28582 value of @code{"all"}. Otherwise, the value of the @var{stopped}
28583 field will be a list of thread identifiers. Presently, this list will
28584 always include a single thread, but frontend should be prepared to see
28585 several threads in the list. The @var{core} field reports the
28586 processor core on which the stop event has happened. This field may be absent
28587 if such information is not available.
28589 @item =thread-group-added,id="@var{id}"
28590 @itemx =thread-group-removed,id="@var{id}"
28591 A thread group was either added or removed. The @var{id} field
28592 contains the @value{GDBN} identifier of the thread group. When a thread
28593 group is added, it generally might not be associated with a running
28594 process. When a thread group is removed, its id becomes invalid and
28595 cannot be used in any way.
28597 @item =thread-group-started,id="@var{id}",pid="@var{pid}"
28598 A thread group became associated with a running program,
28599 either because the program was just started or the thread group
28600 was attached to a program. The @var{id} field contains the
28601 @value{GDBN} identifier of the thread group. The @var{pid} field
28602 contains process identifier, specific to the operating system.
28604 @item =thread-group-exited,id="@var{id}"[,exit-code="@var{code}"]
28605 A thread group is no longer associated with a running program,
28606 either because the program has exited, or because it was detached
28607 from. The @var{id} field contains the @value{GDBN} identifier of the
28608 thread group. The @var{code} field is the exit code of the inferior; it exists
28609 only when the inferior exited with some code.
28611 @item =thread-created,id="@var{id}",group-id="@var{gid}"
28612 @itemx =thread-exited,id="@var{id}",group-id="@var{gid}"
28613 A thread either was created, or has exited. The @var{id} field
28614 contains the global @value{GDBN} identifier of the thread. The @var{gid}
28615 field identifies the thread group this thread belongs to.
28617 @item =thread-selected,id="@var{id}"[,frame="@var{frame}"]
28618 Informs that the selected thread or frame were changed. This notification
28619 is not emitted as result of the @code{-thread-select} or
28620 @code{-stack-select-frame} commands, but is emitted whenever an MI command
28621 that is not documented to change the selected thread and frame actually
28622 changes them. In particular, invoking, directly or indirectly
28623 (via user-defined command), the CLI @code{thread} or @code{frame} commands,
28624 will generate this notification. Changing the thread or frame from another
28625 user interface (see @ref{Interpreters}) will also generate this notification.
28627 The @var{frame} field is only present if the newly selected thread is
28628 stopped. See @ref{GDB/MI Frame Information} for the format of its value.
28630 We suggest that in response to this notification, front ends
28631 highlight the selected thread and cause subsequent commands to apply to
28634 @item =library-loaded,...
28635 Reports that a new library file was loaded by the program. This
28636 notification has 5 fields---@var{id}, @var{target-name},
28637 @var{host-name}, @var{symbols-loaded} and @var{ranges}. The @var{id} field is an
28638 opaque identifier of the library. For remote debugging case,
28639 @var{target-name} and @var{host-name} fields give the name of the
28640 library file on the target, and on the host respectively. For native
28641 debugging, both those fields have the same value. The
28642 @var{symbols-loaded} field is emitted only for backward compatibility
28643 and should not be relied on to convey any useful information. The
28644 @var{thread-group} field, if present, specifies the id of the thread
28645 group in whose context the library was loaded. If the field is
28646 absent, it means the library was loaded in the context of all present
28647 thread groups. The @var{ranges} field specifies the ranges of addresses belonging
28650 @item =library-unloaded,...
28651 Reports that a library was unloaded by the program. This notification
28652 has 3 fields---@var{id}, @var{target-name} and @var{host-name} with
28653 the same meaning as for the @code{=library-loaded} notification.
28654 The @var{thread-group} field, if present, specifies the id of the
28655 thread group in whose context the library was unloaded. If the field is
28656 absent, it means the library was unloaded in the context of all present
28659 @item =traceframe-changed,num=@var{tfnum},tracepoint=@var{tpnum}
28660 @itemx =traceframe-changed,end
28661 Reports that the trace frame was changed and its new number is
28662 @var{tfnum}. The number of the tracepoint associated with this trace
28663 frame is @var{tpnum}.
28665 @item =tsv-created,name=@var{name},initial=@var{initial}
28666 Reports that the new trace state variable @var{name} is created with
28667 initial value @var{initial}.
28669 @item =tsv-deleted,name=@var{name}
28670 @itemx =tsv-deleted
28671 Reports that the trace state variable @var{name} is deleted or all
28672 trace state variables are deleted.
28674 @item =tsv-modified,name=@var{name},initial=@var{initial}[,current=@var{current}]
28675 Reports that the trace state variable @var{name} is modified with
28676 the initial value @var{initial}. The current value @var{current} of
28677 trace state variable is optional and is reported if the current
28678 value of trace state variable is known.
28680 @item =breakpoint-created,bkpt=@{...@}
28681 @itemx =breakpoint-modified,bkpt=@{...@}
28682 @itemx =breakpoint-deleted,id=@var{number}
28683 Reports that a breakpoint was created, modified, or deleted,
28684 respectively. Only user-visible breakpoints are reported to the MI
28687 The @var{bkpt} argument is of the same form as returned by the various
28688 breakpoint commands; @xref{GDB/MI Breakpoint Commands}. The
28689 @var{number} is the ordinal number of the breakpoint.
28691 Note that if a breakpoint is emitted in the result record of a
28692 command, then it will not also be emitted in an async record.
28694 @item =record-started,thread-group="@var{id}",method="@var{method}"[,format="@var{format}"]
28695 @itemx =record-stopped,thread-group="@var{id}"
28696 Execution log recording was either started or stopped on an
28697 inferior. The @var{id} is the @value{GDBN} identifier of the thread
28698 group corresponding to the affected inferior.
28700 The @var{method} field indicates the method used to record execution. If the
28701 method in use supports multiple recording formats, @var{format} will be present
28702 and contain the currently used format. @xref{Process Record and Replay},
28703 for existing method and format values.
28705 @item =cmd-param-changed,param=@var{param},value=@var{value}
28706 Reports that a parameter of the command @code{set @var{param}} is
28707 changed to @var{value}. In the multi-word @code{set} command,
28708 the @var{param} is the whole parameter list to @code{set} command.
28709 For example, In command @code{set check type on}, @var{param}
28710 is @code{check type} and @var{value} is @code{on}.
28712 @item =memory-changed,thread-group=@var{id},addr=@var{addr},len=@var{len}[,type="code"]
28713 Reports that bytes from @var{addr} to @var{data} + @var{len} were
28714 written in an inferior. The @var{id} is the identifier of the
28715 thread group corresponding to the affected inferior. The optional
28716 @code{type="code"} part is reported if the memory written to holds
28720 @node GDB/MI Breakpoint Information
28721 @subsection @sc{gdb/mi} Breakpoint Information
28723 When @value{GDBN} reports information about a breakpoint, a
28724 tracepoint, a watchpoint, or a catchpoint, it uses a tuple with the
28729 The breakpoint number.
28732 The type of the breakpoint. For ordinary breakpoints this will be
28733 @samp{breakpoint}, but many values are possible.
28736 If the type of the breakpoint is @samp{catchpoint}, then this
28737 indicates the exact type of catchpoint.
28740 This is the breakpoint disposition---either @samp{del}, meaning that
28741 the breakpoint will be deleted at the next stop, or @samp{keep},
28742 meaning that the breakpoint will not be deleted.
28745 This indicates whether the breakpoint is enabled, in which case the
28746 value is @samp{y}, or disabled, in which case the value is @samp{n}.
28747 Note that this is not the same as the field @code{enable}.
28750 The address of the breakpoint. This may be a hexidecimal number,
28751 giving the address; or the string @samp{<PENDING>}, for a pending
28752 breakpoint; or the string @samp{<MULTIPLE>}, for a breakpoint with
28753 multiple locations. This field will not be present if no address can
28754 be determined. For example, a watchpoint does not have an address.
28757 If known, the function in which the breakpoint appears.
28758 If not known, this field is not present.
28761 The name of the source file which contains this function, if known.
28762 If not known, this field is not present.
28765 The full file name of the source file which contains this function, if
28766 known. If not known, this field is not present.
28769 The line number at which this breakpoint appears, if known.
28770 If not known, this field is not present.
28773 If the source file is not known, this field may be provided. If
28774 provided, this holds the address of the breakpoint, possibly followed
28778 If this breakpoint is pending, this field is present and holds the
28779 text used to set the breakpoint, as entered by the user.
28782 Where this breakpoint's condition is evaluated, either @samp{host} or
28786 If this is a thread-specific breakpoint, then this identifies the
28787 thread in which the breakpoint can trigger.
28790 If this breakpoint is restricted to a particular Ada task, then this
28791 field will hold the task identifier.
28794 If the breakpoint is conditional, this is the condition expression.
28797 The ignore count of the breakpoint.
28800 The enable count of the breakpoint.
28802 @item traceframe-usage
28805 @item static-tracepoint-marker-string-id
28806 For a static tracepoint, the name of the static tracepoint marker.
28809 For a masked watchpoint, this is the mask.
28812 A tracepoint's pass count.
28814 @item original-location
28815 The location of the breakpoint as originally specified by the user.
28816 This field is optional.
28819 The number of times the breakpoint has been hit.
28822 This field is only given for tracepoints. This is either @samp{y},
28823 meaning that the tracepoint is installed, or @samp{n}, meaning that it
28827 Some extra data, the exact contents of which are type-dependent.
28830 This field is present if the breakpoint has multiple locations. It is also
28831 exceptionally present if the breakpoint is enabled and has a single, disabled
28834 The value is a list of locations. The format of a location is decribed below.
28838 A location in a multi-location breakpoint is represented as a tuple with the
28844 The location number as a dotted pair, like @samp{1.2}. The first digit is the
28845 number of the parent breakpoint. The second digit is the number of the
28846 location within that breakpoint.
28849 This indicates whether the location is enabled, in which case the
28850 value is @samp{y}, or disabled, in which case the value is @samp{n}.
28851 Note that this is not the same as the field @code{enable}.
28854 The address of this location as an hexidecimal number.
28857 If known, the function in which the location appears.
28858 If not known, this field is not present.
28861 The name of the source file which contains this location, if known.
28862 If not known, this field is not present.
28865 The full file name of the source file which contains this location, if
28866 known. If not known, this field is not present.
28869 The line number at which this location appears, if known.
28870 If not known, this field is not present.
28872 @item thread-groups
28873 The thread groups this location is in.
28877 For example, here is what the output of @code{-break-insert}
28878 (@pxref{GDB/MI Breakpoint Commands}) might be:
28881 -> -break-insert main
28882 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
28883 enabled="y",addr="0x08048564",func="main",file="myprog.c",
28884 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
28889 @node GDB/MI Frame Information
28890 @subsection @sc{gdb/mi} Frame Information
28892 Response from many MI commands includes an information about stack
28893 frame. This information is a tuple that may have the following
28898 The level of the stack frame. The innermost frame has the level of
28899 zero. This field is always present.
28902 The name of the function corresponding to the frame. This field may
28903 be absent if @value{GDBN} is unable to determine the function name.
28906 The code address for the frame. This field is always present.
28909 The name of the source files that correspond to the frame's code
28910 address. This field may be absent.
28913 The source line corresponding to the frames' code address. This field
28917 The name of the binary file (either executable or shared library) the
28918 corresponds to the frame's code address. This field may be absent.
28922 @node GDB/MI Thread Information
28923 @subsection @sc{gdb/mi} Thread Information
28925 Whenever @value{GDBN} has to report an information about a thread, it
28926 uses a tuple with the following fields. The fields are always present unless
28931 The global numeric id assigned to the thread by @value{GDBN}.
28934 The target-specific string identifying the thread.
28937 Additional information about the thread provided by the target.
28938 It is supposed to be human-readable and not interpreted by the
28939 frontend. This field is optional.
28942 The name of the thread. If the user specified a name using the
28943 @code{thread name} command, then this name is given. Otherwise, if
28944 @value{GDBN} can extract the thread name from the target, then that
28945 name is given. If @value{GDBN} cannot find the thread name, then this
28949 The execution state of the thread, either @samp{stopped} or @samp{running},
28950 depending on whether the thread is presently running.
28953 The stack frame currently executing in the thread. This field is only present
28954 if the thread is stopped. Its format is documented in
28955 @ref{GDB/MI Frame Information}.
28958 The value of this field is an integer number of the processor core the
28959 thread was last seen on. This field is optional.
28962 @node GDB/MI Ada Exception Information
28963 @subsection @sc{gdb/mi} Ada Exception Information
28965 Whenever a @code{*stopped} record is emitted because the program
28966 stopped after hitting an exception catchpoint (@pxref{Set Catchpoints}),
28967 @value{GDBN} provides the name of the exception that was raised via
28968 the @code{exception-name} field. Also, for exceptions that were raised
28969 with an exception message, @value{GDBN} provides that message via
28970 the @code{exception-message} field.
28972 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
28973 @node GDB/MI Simple Examples
28974 @section Simple Examples of @sc{gdb/mi} Interaction
28975 @cindex @sc{gdb/mi}, simple examples
28977 This subsection presents several simple examples of interaction using
28978 the @sc{gdb/mi} interface. In these examples, @samp{->} means that the
28979 following line is passed to @sc{gdb/mi} as input, while @samp{<-} means
28980 the output received from @sc{gdb/mi}.
28982 Note the line breaks shown in the examples are here only for
28983 readability, they don't appear in the real output.
28985 @subheading Setting a Breakpoint
28987 Setting a breakpoint generates synchronous output which contains detailed
28988 information of the breakpoint.
28991 -> -break-insert main
28992 <- ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
28993 enabled="y",addr="0x08048564",func="main",file="myprog.c",
28994 fullname="/home/nickrob/myprog.c",line="68",thread-groups=["i1"],
28999 @subheading Program Execution
29001 Program execution generates asynchronous records and MI gives the
29002 reason that execution stopped.
29008 <- *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
29009 frame=@{addr="0x08048564",func="main",
29010 args=[@{name="argc",value="1"@},@{name="argv",value="0xbfc4d4d4"@}],
29011 file="myprog.c",fullname="/home/nickrob/myprog.c",line="68",
29012 arch="i386:x86_64"@}
29017 <- *stopped,reason="exited-normally"
29021 @subheading Quitting @value{GDBN}
29023 Quitting @value{GDBN} just prints the result class @samp{^exit}.
29031 Please note that @samp{^exit} is printed immediately, but it might
29032 take some time for @value{GDBN} to actually exit. During that time, @value{GDBN}
29033 performs necessary cleanups, including killing programs being debugged
29034 or disconnecting from debug hardware, so the frontend should wait till
29035 @value{GDBN} exits and should only forcibly kill @value{GDBN} if it
29036 fails to exit in reasonable time.
29038 @subheading A Bad Command
29040 Here's what happens if you pass a non-existent command:
29044 <- ^error,msg="Undefined MI command: rubbish"
29049 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29050 @node GDB/MI Command Description Format
29051 @section @sc{gdb/mi} Command Description Format
29053 The remaining sections describe blocks of commands. Each block of
29054 commands is laid out in a fashion similar to this section.
29056 @subheading Motivation
29058 The motivation for this collection of commands.
29060 @subheading Introduction
29062 A brief introduction to this collection of commands as a whole.
29064 @subheading Commands
29066 For each command in the block, the following is described:
29068 @subsubheading Synopsis
29071 -command @var{args}@dots{}
29074 @subsubheading Result
29076 @subsubheading @value{GDBN} Command
29078 The corresponding @value{GDBN} CLI command(s), if any.
29080 @subsubheading Example
29082 Example(s) formatted for readability. Some of the described commands have
29083 not been implemented yet and these are labeled N.A.@: (not available).
29086 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29087 @node GDB/MI Breakpoint Commands
29088 @section @sc{gdb/mi} Breakpoint Commands
29090 @cindex breakpoint commands for @sc{gdb/mi}
29091 @cindex @sc{gdb/mi}, breakpoint commands
29092 This section documents @sc{gdb/mi} commands for manipulating
29095 @subheading The @code{-break-after} Command
29096 @findex -break-after
29098 @subsubheading Synopsis
29101 -break-after @var{number} @var{count}
29104 The breakpoint number @var{number} is not in effect until it has been
29105 hit @var{count} times. To see how this is reflected in the output of
29106 the @samp{-break-list} command, see the description of the
29107 @samp{-break-list} command below.
29109 @subsubheading @value{GDBN} Command
29111 The corresponding @value{GDBN} command is @samp{ignore}.
29113 @subsubheading Example
29118 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29119 enabled="y",addr="0x000100d0",func="main",file="hello.c",
29120 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
29128 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29129 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29130 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29131 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29132 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29133 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29134 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29135 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29136 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29137 line="5",thread-groups=["i1"],times="0",ignore="3"@}]@}
29142 @subheading The @code{-break-catch} Command
29143 @findex -break-catch
29146 @subheading The @code{-break-commands} Command
29147 @findex -break-commands
29149 @subsubheading Synopsis
29152 -break-commands @var{number} [ @var{command1} ... @var{commandN} ]
29155 Specifies the CLI commands that should be executed when breakpoint
29156 @var{number} is hit. The parameters @var{command1} to @var{commandN}
29157 are the commands. If no command is specified, any previously-set
29158 commands are cleared. @xref{Break Commands}. Typical use of this
29159 functionality is tracing a program, that is, printing of values of
29160 some variables whenever breakpoint is hit and then continuing.
29162 @subsubheading @value{GDBN} Command
29164 The corresponding @value{GDBN} command is @samp{commands}.
29166 @subsubheading Example
29171 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",
29172 enabled="y",addr="0x000100d0",func="main",file="hello.c",
29173 fullname="/home/foo/hello.c",line="5",thread-groups=["i1"],
29176 -break-commands 1 "print v" "continue"
29181 @subheading The @code{-break-condition} Command
29182 @findex -break-condition
29184 @subsubheading Synopsis
29187 -break-condition @var{number} @var{expr}
29190 Breakpoint @var{number} will stop the program only if the condition in
29191 @var{expr} is true. The condition becomes part of the
29192 @samp{-break-list} output (see the description of the @samp{-break-list}
29195 @subsubheading @value{GDBN} Command
29197 The corresponding @value{GDBN} command is @samp{condition}.
29199 @subsubheading Example
29203 -break-condition 1 1
29207 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29208 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29209 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29210 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29211 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29212 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29213 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29214 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29215 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29216 line="5",cond="1",thread-groups=["i1"],times="0",ignore="3"@}]@}
29220 @subheading The @code{-break-delete} Command
29221 @findex -break-delete
29223 @subsubheading Synopsis
29226 -break-delete ( @var{breakpoint} )+
29229 Delete the breakpoint(s) whose number(s) are specified in the argument
29230 list. This is obviously reflected in the breakpoint list.
29232 @subsubheading @value{GDBN} Command
29234 The corresponding @value{GDBN} command is @samp{delete}.
29236 @subsubheading Example
29244 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29245 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29246 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29247 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29248 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29249 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29250 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29255 @subheading The @code{-break-disable} Command
29256 @findex -break-disable
29258 @subsubheading Synopsis
29261 -break-disable ( @var{breakpoint} )+
29264 Disable the named @var{breakpoint}(s). The field @samp{enabled} in the
29265 break list is now set to @samp{n} for the named @var{breakpoint}(s).
29267 @subsubheading @value{GDBN} Command
29269 The corresponding @value{GDBN} command is @samp{disable}.
29271 @subsubheading Example
29279 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29280 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29281 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29282 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29283 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29284 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29285 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29286 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="n",
29287 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29288 line="5",thread-groups=["i1"],times="0"@}]@}
29292 @subheading The @code{-break-enable} Command
29293 @findex -break-enable
29295 @subsubheading Synopsis
29298 -break-enable ( @var{breakpoint} )+
29301 Enable (previously disabled) @var{breakpoint}(s).
29303 @subsubheading @value{GDBN} Command
29305 The corresponding @value{GDBN} command is @samp{enable}.
29307 @subsubheading Example
29315 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29316 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29317 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29318 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29319 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29320 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29321 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29322 body=[bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29323 addr="0x000100d0",func="main",file="hello.c",fullname="/home/foo/hello.c",
29324 line="5",thread-groups=["i1"],times="0"@}]@}
29328 @subheading The @code{-break-info} Command
29329 @findex -break-info
29331 @subsubheading Synopsis
29334 -break-info @var{breakpoint}
29338 Get information about a single breakpoint.
29340 The result is a table of breakpoints. @xref{GDB/MI Breakpoint
29341 Information}, for details on the format of each breakpoint in the
29344 @subsubheading @value{GDBN} Command
29346 The corresponding @value{GDBN} command is @samp{info break @var{breakpoint}}.
29348 @subsubheading Example
29351 @subheading The @code{-break-insert} Command
29352 @findex -break-insert
29353 @anchor{-break-insert}
29355 @subsubheading Synopsis
29358 -break-insert [ -t ] [ -h ] [ -f ] [ -d ] [ -a ]
29359 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29360 [ -p @var{thread-id} ] [ @var{location} ]
29364 If specified, @var{location}, can be one of:
29367 @item linespec location
29368 A linespec location. @xref{Linespec Locations}.
29370 @item explicit location
29371 An explicit location. @sc{gdb/mi} explicit locations are
29372 analogous to the CLI's explicit locations using the option names
29373 listed below. @xref{Explicit Locations}.
29376 @item --source @var{filename}
29377 The source file name of the location. This option requires the use
29378 of either @samp{--function} or @samp{--line}.
29380 @item --function @var{function}
29381 The name of a function or method.
29383 @item --label @var{label}
29384 The name of a label.
29386 @item --line @var{lineoffset}
29387 An absolute or relative line offset from the start of the location.
29390 @item address location
29391 An address location, *@var{address}. @xref{Address Locations}.
29395 The possible optional parameters of this command are:
29399 Insert a temporary breakpoint.
29401 Insert a hardware breakpoint.
29403 If @var{location} cannot be parsed (for example if it
29404 refers to unknown files or functions), create a pending
29405 breakpoint. Without this flag, @value{GDBN} will report
29406 an error, and won't create a breakpoint, if @var{location}
29409 Create a disabled breakpoint.
29411 Create a tracepoint. @xref{Tracepoints}. When this parameter
29412 is used together with @samp{-h}, a fast tracepoint is created.
29413 @item -c @var{condition}
29414 Make the breakpoint conditional on @var{condition}.
29415 @item -i @var{ignore-count}
29416 Initialize the @var{ignore-count}.
29417 @item -p @var{thread-id}
29418 Restrict the breakpoint to the thread with the specified global
29422 @subsubheading Result
29424 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29425 resulting breakpoint.
29427 Note: this format is open to change.
29428 @c An out-of-band breakpoint instead of part of the result?
29430 @subsubheading @value{GDBN} Command
29432 The corresponding @value{GDBN} commands are @samp{break}, @samp{tbreak},
29433 @samp{hbreak}, and @samp{thbreak}. @c and @samp{rbreak}.
29435 @subsubheading Example
29440 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",
29441 fullname="/home/foo/recursive2.c,line="4",thread-groups=["i1"],
29444 -break-insert -t foo
29445 ^done,bkpt=@{number="2",addr="0x00010774",file="recursive2.c",
29446 fullname="/home/foo/recursive2.c,line="11",thread-groups=["i1"],
29450 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29451 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29452 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29453 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29454 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29455 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29456 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29457 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29458 addr="0x0001072c", func="main",file="recursive2.c",
29459 fullname="/home/foo/recursive2.c,"line="4",thread-groups=["i1"],
29461 bkpt=@{number="2",type="breakpoint",disp="del",enabled="y",
29462 addr="0x00010774",func="foo",file="recursive2.c",
29463 fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29466 @c -break-insert -r foo.*
29467 @c ~int foo(int, int);
29468 @c ^done,bkpt=@{number="3",addr="0x00010774",file="recursive2.c,
29469 @c "fullname="/home/foo/recursive2.c",line="11",thread-groups=["i1"],
29474 @subheading The @code{-dprintf-insert} Command
29475 @findex -dprintf-insert
29477 @subsubheading Synopsis
29480 -dprintf-insert [ -t ] [ -f ] [ -d ]
29481 [ -c @var{condition} ] [ -i @var{ignore-count} ]
29482 [ -p @var{thread-id} ] [ @var{location} ] [ @var{format} ]
29487 If supplied, @var{location} may be specified the same way as for
29488 the @code{-break-insert} command. @xref{-break-insert}.
29490 The possible optional parameters of this command are:
29494 Insert a temporary breakpoint.
29496 If @var{location} cannot be parsed (for example, if it
29497 refers to unknown files or functions), create a pending
29498 breakpoint. Without this flag, @value{GDBN} will report
29499 an error, and won't create a breakpoint, if @var{location}
29502 Create a disabled breakpoint.
29503 @item -c @var{condition}
29504 Make the breakpoint conditional on @var{condition}.
29505 @item -i @var{ignore-count}
29506 Set the ignore count of the breakpoint (@pxref{Conditions, ignore count})
29507 to @var{ignore-count}.
29508 @item -p @var{thread-id}
29509 Restrict the breakpoint to the thread with the specified global
29513 @subsubheading Result
29515 @xref{GDB/MI Breakpoint Information}, for details on the format of the
29516 resulting breakpoint.
29518 @c An out-of-band breakpoint instead of part of the result?
29520 @subsubheading @value{GDBN} Command
29522 The corresponding @value{GDBN} command is @samp{dprintf}.
29524 @subsubheading Example
29528 4-dprintf-insert foo "At foo entry\n"
29529 4^done,bkpt=@{number="1",type="dprintf",disp="keep",enabled="y",
29530 addr="0x000000000040061b",func="foo",file="mi-dprintf.c",
29531 fullname="mi-dprintf.c",line="25",thread-groups=["i1"],
29532 times="0",script=@{"printf \"At foo entry\\n\"","continue"@},
29533 original-location="foo"@}
29535 5-dprintf-insert 26 "arg=%d, g=%d\n" arg g
29536 5^done,bkpt=@{number="2",type="dprintf",disp="keep",enabled="y",
29537 addr="0x000000000040062a",func="foo",file="mi-dprintf.c",
29538 fullname="mi-dprintf.c",line="26",thread-groups=["i1"],
29539 times="0",script=@{"printf \"arg=%d, g=%d\\n\", arg, g","continue"@},
29540 original-location="mi-dprintf.c:26"@}
29544 @subheading The @code{-break-list} Command
29545 @findex -break-list
29547 @subsubheading Synopsis
29553 Displays the list of inserted breakpoints, showing the following fields:
29557 number of the breakpoint
29559 type of the breakpoint: @samp{breakpoint} or @samp{watchpoint}
29561 should the breakpoint be deleted or disabled when it is hit: @samp{keep}
29564 is the breakpoint enabled or no: @samp{y} or @samp{n}
29566 memory location at which the breakpoint is set
29568 logical location of the breakpoint, expressed by function name, file
29570 @item Thread-groups
29571 list of thread groups to which this breakpoint applies
29573 number of times the breakpoint has been hit
29576 If there are no breakpoints or watchpoints, the @code{BreakpointTable}
29577 @code{body} field is an empty list.
29579 @subsubheading @value{GDBN} Command
29581 The corresponding @value{GDBN} command is @samp{info break}.
29583 @subsubheading Example
29588 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29589 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29590 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29591 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29592 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29593 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29594 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29595 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29596 addr="0x000100d0",func="main",file="hello.c",line="5",thread-groups=["i1"],
29598 bkpt=@{number="2",type="breakpoint",disp="keep",enabled="y",
29599 addr="0x00010114",func="foo",file="hello.c",fullname="/home/foo/hello.c",
29600 line="13",thread-groups=["i1"],times="0"@}]@}
29604 Here's an example of the result when there are no breakpoints:
29609 ^done,BreakpointTable=@{nr_rows="0",nr_cols="6",
29610 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29611 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29612 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29613 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29614 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29615 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29620 @subheading The @code{-break-passcount} Command
29621 @findex -break-passcount
29623 @subsubheading Synopsis
29626 -break-passcount @var{tracepoint-number} @var{passcount}
29629 Set the passcount for tracepoint @var{tracepoint-number} to
29630 @var{passcount}. If the breakpoint referred to by @var{tracepoint-number}
29631 is not a tracepoint, error is emitted. This corresponds to CLI
29632 command @samp{passcount}.
29634 @subheading The @code{-break-watch} Command
29635 @findex -break-watch
29637 @subsubheading Synopsis
29640 -break-watch [ -a | -r ]
29643 Create a watchpoint. With the @samp{-a} option it will create an
29644 @dfn{access} watchpoint, i.e., a watchpoint that triggers either on a
29645 read from or on a write to the memory location. With the @samp{-r}
29646 option, the watchpoint created is a @dfn{read} watchpoint, i.e., it will
29647 trigger only when the memory location is accessed for reading. Without
29648 either of the options, the watchpoint created is a regular watchpoint,
29649 i.e., it will trigger when the memory location is accessed for writing.
29650 @xref{Set Watchpoints, , Setting Watchpoints}.
29652 Note that @samp{-break-list} will report a single list of watchpoints and
29653 breakpoints inserted.
29655 @subsubheading @value{GDBN} Command
29657 The corresponding @value{GDBN} commands are @samp{watch}, @samp{awatch}, and
29660 @subsubheading Example
29662 Setting a watchpoint on a variable in the @code{main} function:
29667 ^done,wpt=@{number="2",exp="x"@}
29672 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="x"@},
29673 value=@{old="-268439212",new="55"@},
29674 frame=@{func="main",args=[],file="recursive2.c",
29675 fullname="/home/foo/bar/recursive2.c",line="5",arch="i386:x86_64"@}
29679 Setting a watchpoint on a variable local to a function. @value{GDBN} will stop
29680 the program execution twice: first for the variable changing value, then
29681 for the watchpoint going out of scope.
29686 ^done,wpt=@{number="5",exp="C"@}
29691 *stopped,reason="watchpoint-trigger",
29692 wpt=@{number="5",exp="C"@},value=@{old="-276895068",new="3"@},
29693 frame=@{func="callee4",args=[],
29694 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29695 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
29696 arch="i386:x86_64"@}
29701 *stopped,reason="watchpoint-scope",wpnum="5",
29702 frame=@{func="callee3",args=[@{name="strarg",
29703 value="0x11940 \"A string argument.\""@}],
29704 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29705 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
29706 arch="i386:x86_64"@}
29710 Listing breakpoints and watchpoints, at different points in the program
29711 execution. Note that once the watchpoint goes out of scope, it is
29717 ^done,wpt=@{number="2",exp="C"@}
29720 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29721 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29722 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29723 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29724 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29725 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29726 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29727 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29728 addr="0x00010734",func="callee4",
29729 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29730 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c"line="8",thread-groups=["i1"],
29732 bkpt=@{number="2",type="watchpoint",disp="keep",
29733 enabled="y",addr="",what="C",thread-groups=["i1"],times="0"@}]@}
29738 *stopped,reason="watchpoint-trigger",wpt=@{number="2",exp="C"@},
29739 value=@{old="-276895068",new="3"@},
29740 frame=@{func="callee4",args=[],
29741 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29742 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="13",
29743 arch="i386:x86_64"@}
29746 ^done,BreakpointTable=@{nr_rows="2",nr_cols="6",
29747 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29748 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29749 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29750 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29751 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29752 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29753 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29754 addr="0x00010734",func="callee4",
29755 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29756 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",thread-groups=["i1"],
29758 bkpt=@{number="2",type="watchpoint",disp="keep",
29759 enabled="y",addr="",what="C",thread-groups=["i1"],times="-5"@}]@}
29763 ^done,reason="watchpoint-scope",wpnum="2",
29764 frame=@{func="callee3",args=[@{name="strarg",
29765 value="0x11940 \"A string argument.\""@}],
29766 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29767 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
29768 arch="i386:x86_64"@}
29771 ^done,BreakpointTable=@{nr_rows="1",nr_cols="6",
29772 hdr=[@{width="3",alignment="-1",col_name="number",colhdr="Num"@},
29773 @{width="14",alignment="-1",col_name="type",colhdr="Type"@},
29774 @{width="4",alignment="-1",col_name="disp",colhdr="Disp"@},
29775 @{width="3",alignment="-1",col_name="enabled",colhdr="Enb"@},
29776 @{width="10",alignment="-1",col_name="addr",colhdr="Address"@},
29777 @{width="40",alignment="2",col_name="what",colhdr="What"@}],
29778 body=[bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
29779 addr="0x00010734",func="callee4",
29780 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
29781 fullname="/home/foo/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
29782 thread-groups=["i1"],times="1"@}]@}
29787 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
29788 @node GDB/MI Catchpoint Commands
29789 @section @sc{gdb/mi} Catchpoint Commands
29791 This section documents @sc{gdb/mi} commands for manipulating
29795 * Shared Library GDB/MI Catchpoint Commands::
29796 * Ada Exception GDB/MI Catchpoint Commands::
29797 * C++ Exception GDB/MI Catchpoint Commands::
29800 @node Shared Library GDB/MI Catchpoint Commands
29801 @subsection Shared Library @sc{gdb/mi} Catchpoints
29803 @subheading The @code{-catch-load} Command
29804 @findex -catch-load
29806 @subsubheading Synopsis
29809 -catch-load [ -t ] [ -d ] @var{regexp}
29812 Add a catchpoint for library load events. If the @samp{-t} option is used,
29813 the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
29814 Breakpoints}). If the @samp{-d} option is used, the catchpoint is created
29815 in a disabled state. The @samp{regexp} argument is a regular
29816 expression used to match the name of the loaded library.
29819 @subsubheading @value{GDBN} Command
29821 The corresponding @value{GDBN} command is @samp{catch load}.
29823 @subsubheading Example
29826 -catch-load -t foo.so
29827 ^done,bkpt=@{number="1",type="catchpoint",disp="del",enabled="y",
29828 what="load of library matching foo.so",catch-type="load",times="0"@}
29833 @subheading The @code{-catch-unload} Command
29834 @findex -catch-unload
29836 @subsubheading Synopsis
29839 -catch-unload [ -t ] [ -d ] @var{regexp}
29842 Add a catchpoint for library unload events. If the @samp{-t} option is
29843 used, the catchpoint is a temporary one (@pxref{Set Breaks, ,Setting
29844 Breakpoints}). If the @samp{-d} option is used, the catchpoint is
29845 created in a disabled state. The @samp{regexp} argument is a regular
29846 expression used to match the name of the unloaded library.
29848 @subsubheading @value{GDBN} Command
29850 The corresponding @value{GDBN} command is @samp{catch unload}.
29852 @subsubheading Example
29855 -catch-unload -d bar.so
29856 ^done,bkpt=@{number="2",type="catchpoint",disp="keep",enabled="n",
29857 what="load of library matching bar.so",catch-type="unload",times="0"@}
29861 @node Ada Exception GDB/MI Catchpoint Commands
29862 @subsection Ada Exception @sc{gdb/mi} Catchpoints
29864 The following @sc{gdb/mi} commands can be used to create catchpoints
29865 that stop the execution when Ada exceptions are being raised.
29867 @subheading The @code{-catch-assert} Command
29868 @findex -catch-assert
29870 @subsubheading Synopsis
29873 -catch-assert [ -c @var{condition}] [ -d ] [ -t ]
29876 Add a catchpoint for failed Ada assertions.
29878 The possible optional parameters for this command are:
29881 @item -c @var{condition}
29882 Make the catchpoint conditional on @var{condition}.
29884 Create a disabled catchpoint.
29886 Create a temporary catchpoint.
29889 @subsubheading @value{GDBN} Command
29891 The corresponding @value{GDBN} command is @samp{catch assert}.
29893 @subsubheading Example
29897 ^done,bkptno="5",bkpt=@{number="5",type="breakpoint",disp="keep",
29898 enabled="y",addr="0x0000000000404888",what="failed Ada assertions",
29899 thread-groups=["i1"],times="0",
29900 original-location="__gnat_debug_raise_assert_failure"@}
29904 @subheading The @code{-catch-exception} Command
29905 @findex -catch-exception
29907 @subsubheading Synopsis
29910 -catch-exception [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
29914 Add a catchpoint stopping when Ada exceptions are raised.
29915 By default, the command stops the program when any Ada exception
29916 gets raised. But it is also possible, by using some of the
29917 optional parameters described below, to create more selective
29920 The possible optional parameters for this command are:
29923 @item -c @var{condition}
29924 Make the catchpoint conditional on @var{condition}.
29926 Create a disabled catchpoint.
29927 @item -e @var{exception-name}
29928 Only stop when @var{exception-name} is raised. This option cannot
29929 be used combined with @samp{-u}.
29931 Create a temporary catchpoint.
29933 Stop only when an unhandled exception gets raised. This option
29934 cannot be used combined with @samp{-e}.
29937 @subsubheading @value{GDBN} Command
29939 The corresponding @value{GDBN} commands are @samp{catch exception}
29940 and @samp{catch exception unhandled}.
29942 @subsubheading Example
29945 -catch-exception -e Program_Error
29946 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
29947 enabled="y",addr="0x0000000000404874",
29948 what="`Program_Error' Ada exception", thread-groups=["i1"],
29949 times="0",original-location="__gnat_debug_raise_exception"@}
29953 @subheading The @code{-catch-handlers} Command
29954 @findex -catch-handlers
29956 @subsubheading Synopsis
29959 -catch-handlers [ -c @var{condition}] [ -d ] [ -e @var{exception-name} ]
29963 Add a catchpoint stopping when Ada exceptions are handled.
29964 By default, the command stops the program when any Ada exception
29965 gets handled. But it is also possible, by using some of the
29966 optional parameters described below, to create more selective
29969 The possible optional parameters for this command are:
29972 @item -c @var{condition}
29973 Make the catchpoint conditional on @var{condition}.
29975 Create a disabled catchpoint.
29976 @item -e @var{exception-name}
29977 Only stop when @var{exception-name} is handled.
29979 Create a temporary catchpoint.
29982 @subsubheading @value{GDBN} Command
29984 The corresponding @value{GDBN} command is @samp{catch handlers}.
29986 @subsubheading Example
29989 -catch-handlers -e Constraint_Error
29990 ^done,bkptno="4",bkpt=@{number="4",type="breakpoint",disp="keep",
29991 enabled="y",addr="0x0000000000402f68",
29992 what="`Constraint_Error' Ada exception handlers",thread-groups=["i1"],
29993 times="0",original-location="__gnat_begin_handler"@}
29997 @node C++ Exception GDB/MI Catchpoint Commands
29998 @subsection C@t{++} Exception @sc{gdb/mi} Catchpoints
30000 The following @sc{gdb/mi} commands can be used to create catchpoints
30001 that stop the execution when C@t{++} exceptions are being throw, rethrown,
30004 @subheading The @code{-catch-throw} Command
30005 @findex -catch-throw
30007 @subsubheading Synopsis
30010 -catch-throw [ -t ] [ -r @var{regexp}]
30013 Stop when the debuggee throws a C@t{++} exception. If @var{regexp} is
30014 given, then only exceptions whose type matches the regular expression
30017 If @samp{-t} is given, then the catchpoint is enabled only for one
30018 stop, the catchpoint is automatically deleted after stopping once for
30021 @subsubheading @value{GDBN} Command
30023 The corresponding @value{GDBN} commands are @samp{catch throw}
30024 and @samp{tcatch throw} (@pxref{Set Catchpoints}).
30026 @subsubheading Example
30029 -catch-throw -r exception_type
30030 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30031 addr="0x00000000004006c0",what="exception throw",
30032 catch-type="throw",thread-groups=["i1"],
30033 regexp="exception_type",times="0"@}
30039 ~"Catchpoint 1 (exception thrown), 0x00007ffff7ae00ed
30040 in __cxa_throw () from /lib64/libstdc++.so.6\n"
30041 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
30042 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_throw",
30043 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
30044 thread-id="1",stopped-threads="all",core="6"
30048 @subheading The @code{-catch-rethrow} Command
30049 @findex -catch-rethrow
30051 @subsubheading Synopsis
30054 -catch-rethrow [ -t ] [ -r @var{regexp}]
30057 Stop when a C@t{++} exception is re-thrown. If @var{regexp} is given,
30058 then only exceptions whose type matches the regular expression will be
30061 If @samp{-t} is given, then the catchpoint is enabled only for one
30062 stop, the catchpoint is automatically deleted after the first event is
30065 @subsubheading @value{GDBN} Command
30067 The corresponding @value{GDBN} commands are @samp{catch rethrow}
30068 and @samp{tcatch rethrow} (@pxref{Set Catchpoints}).
30070 @subsubheading Example
30073 -catch-rethrow -r exception_type
30074 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30075 addr="0x00000000004006c0",what="exception rethrow",
30076 catch-type="rethrow",thread-groups=["i1"],
30077 regexp="exception_type",times="0"@}
30083 ~"Catchpoint 1 (exception rethrown), 0x00007ffff7ae00ed
30084 in __cxa_rethrow () from /lib64/libstdc++.so.6\n"
30085 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
30086 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_rethrow",
30087 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
30088 thread-id="1",stopped-threads="all",core="6"
30092 @subheading The @code{-catch-catch} Command
30093 @findex -catch-catch
30095 @subsubheading Synopsis
30098 -catch-catch [ -t ] [ -r @var{regexp}]
30101 Stop when the debuggee catches a C@t{++} exception. If @var{regexp}
30102 is given, then only exceptions whose type matches the regular
30103 expression will be caught.
30105 If @samp{-t} is given, then the catchpoint is enabled only for one
30106 stop, the catchpoint is automatically deleted after the first event is
30109 @subsubheading @value{GDBN} Command
30111 The corresponding @value{GDBN} commands are @samp{catch catch}
30112 and @samp{tcatch catch} (@pxref{Set Catchpoints}).
30114 @subsubheading Example
30117 -catch-catch -r exception_type
30118 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
30119 addr="0x00000000004006c0",what="exception catch",
30120 catch-type="catch",thread-groups=["i1"],
30121 regexp="exception_type",times="0"@}
30127 ~"Catchpoint 1 (exception caught), 0x00007ffff7ae00ed
30128 in __cxa_begin_catch () from /lib64/libstdc++.so.6\n"
30129 *stopped,bkptno="1",reason="breakpoint-hit",disp="keep",
30130 frame=@{addr="0x00007ffff7ae00ed",func="__cxa_begin_catch",
30131 args=[],from="/lib64/libstdc++.so.6",arch="i386:x86-64"@},
30132 thread-id="1",stopped-threads="all",core="6"
30136 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30137 @node GDB/MI Program Context
30138 @section @sc{gdb/mi} Program Context
30140 @subheading The @code{-exec-arguments} Command
30141 @findex -exec-arguments
30144 @subsubheading Synopsis
30147 -exec-arguments @var{args}
30150 Set the inferior program arguments, to be used in the next
30153 @subsubheading @value{GDBN} Command
30155 The corresponding @value{GDBN} command is @samp{set args}.
30157 @subsubheading Example
30161 -exec-arguments -v word
30168 @subheading The @code{-exec-show-arguments} Command
30169 @findex -exec-show-arguments
30171 @subsubheading Synopsis
30174 -exec-show-arguments
30177 Print the arguments of the program.
30179 @subsubheading @value{GDBN} Command
30181 The corresponding @value{GDBN} command is @samp{show args}.
30183 @subsubheading Example
30188 @subheading The @code{-environment-cd} Command
30189 @findex -environment-cd
30191 @subsubheading Synopsis
30194 -environment-cd @var{pathdir}
30197 Set @value{GDBN}'s working directory.
30199 @subsubheading @value{GDBN} Command
30201 The corresponding @value{GDBN} command is @samp{cd}.
30203 @subsubheading Example
30207 -environment-cd /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
30213 @subheading The @code{-environment-directory} Command
30214 @findex -environment-directory
30216 @subsubheading Synopsis
30219 -environment-directory [ -r ] [ @var{pathdir} ]+
30222 Add directories @var{pathdir} to beginning of search path for source files.
30223 If the @samp{-r} option is used, the search path is reset to the default
30224 search path. If directories @var{pathdir} are supplied in addition to the
30225 @samp{-r} option, the search path is first reset and then addition
30227 Multiple directories may be specified, separated by blanks. Specifying
30228 multiple directories in a single command
30229 results in the directories added to the beginning of the
30230 search path in the same order they were presented in the command.
30231 If blanks are needed as
30232 part of a directory name, double-quotes should be used around
30233 the name. In the command output, the path will show up separated
30234 by the system directory-separator character. The directory-separator
30235 character must not be used
30236 in any directory name.
30237 If no directories are specified, the current search path is displayed.
30239 @subsubheading @value{GDBN} Command
30241 The corresponding @value{GDBN} command is @samp{dir}.
30243 @subsubheading Example
30247 -environment-directory /kwikemart/marge/ezannoni/flathead-dev/devo/gdb
30248 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
30250 -environment-directory ""
30251 ^done,source-path="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb:$cdir:$cwd"
30253 -environment-directory -r /home/jjohnstn/src/gdb /usr/src
30254 ^done,source-path="/home/jjohnstn/src/gdb:/usr/src:$cdir:$cwd"
30256 -environment-directory -r
30257 ^done,source-path="$cdir:$cwd"
30262 @subheading The @code{-environment-path} Command
30263 @findex -environment-path
30265 @subsubheading Synopsis
30268 -environment-path [ -r ] [ @var{pathdir} ]+
30271 Add directories @var{pathdir} to beginning of search path for object files.
30272 If the @samp{-r} option is used, the search path is reset to the original
30273 search path that existed at gdb start-up. If directories @var{pathdir} are
30274 supplied in addition to the
30275 @samp{-r} option, the search path is first reset and then addition
30277 Multiple directories may be specified, separated by blanks. Specifying
30278 multiple directories in a single command
30279 results in the directories added to the beginning of the
30280 search path in the same order they were presented in the command.
30281 If blanks are needed as
30282 part of a directory name, double-quotes should be used around
30283 the name. In the command output, the path will show up separated
30284 by the system directory-separator character. The directory-separator
30285 character must not be used
30286 in any directory name.
30287 If no directories are specified, the current path is displayed.
30290 @subsubheading @value{GDBN} Command
30292 The corresponding @value{GDBN} command is @samp{path}.
30294 @subsubheading Example
30299 ^done,path="/usr/bin"
30301 -environment-path /kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb /bin
30302 ^done,path="/kwikemart/marge/ezannoni/flathead-dev/ppc-eabi/gdb:/bin:/usr/bin"
30304 -environment-path -r /usr/local/bin
30305 ^done,path="/usr/local/bin:/usr/bin"
30310 @subheading The @code{-environment-pwd} Command
30311 @findex -environment-pwd
30313 @subsubheading Synopsis
30319 Show the current working directory.
30321 @subsubheading @value{GDBN} Command
30323 The corresponding @value{GDBN} command is @samp{pwd}.
30325 @subsubheading Example
30330 ^done,cwd="/kwikemart/marge/ezannoni/flathead-dev/devo/gdb"
30334 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30335 @node GDB/MI Thread Commands
30336 @section @sc{gdb/mi} Thread Commands
30339 @subheading The @code{-thread-info} Command
30340 @findex -thread-info
30342 @subsubheading Synopsis
30345 -thread-info [ @var{thread-id} ]
30348 Reports information about either a specific thread, if the
30349 @var{thread-id} parameter is present, or about all threads.
30350 @var{thread-id} is the thread's global thread ID. When printing
30351 information about all threads, also reports the global ID of the
30354 @subsubheading @value{GDBN} Command
30356 The @samp{info thread} command prints the same information
30359 @subsubheading Result
30361 The result contains the following attributes:
30365 A list of threads. The format of the elements of the list is described in
30366 @ref{GDB/MI Thread Information}.
30368 @item current-thread-id
30369 The global id of the currently selected thread. This field is omitted if there
30370 is no selected thread (for example, when the selected inferior is not running,
30371 and therefore has no threads) or if a @var{thread-id} argument was passed to
30376 @subsubheading Example
30381 @{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
30382 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",
30383 args=[]@},state="running"@},
30384 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
30385 frame=@{level="0",addr="0x0804891f",func="foo",
30386 args=[@{name="i",value="10"@}],
30387 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},
30388 state="running"@}],
30389 current-thread-id="1"
30393 @subheading The @code{-thread-list-ids} Command
30394 @findex -thread-list-ids
30396 @subsubheading Synopsis
30402 Produces a list of the currently known global @value{GDBN} thread ids.
30403 At the end of the list it also prints the total number of such
30406 This command is retained for historical reasons, the
30407 @code{-thread-info} command should be used instead.
30409 @subsubheading @value{GDBN} Command
30411 Part of @samp{info threads} supplies the same information.
30413 @subsubheading Example
30418 ^done,thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30419 current-thread-id="1",number-of-threads="3"
30424 @subheading The @code{-thread-select} Command
30425 @findex -thread-select
30427 @subsubheading Synopsis
30430 -thread-select @var{thread-id}
30433 Make thread with global thread number @var{thread-id} the current
30434 thread. It prints the number of the new current thread, and the
30435 topmost frame for that thread.
30437 This command is deprecated in favor of explicitly using the
30438 @samp{--thread} option to each command.
30440 @subsubheading @value{GDBN} Command
30442 The corresponding @value{GDBN} command is @samp{thread}.
30444 @subsubheading Example
30451 *stopped,reason="end-stepping-range",thread-id="2",line="187",
30452 file="../../../devo/gdb/testsuite/gdb.threads/linux-dp.c"
30456 thread-ids=@{thread-id="3",thread-id="2",thread-id="1"@},
30457 number-of-threads="3"
30460 ^done,new-thread-id="3",
30461 frame=@{level="0",func="vprintf",
30462 args=[@{name="format",value="0x8048e9c \"%*s%c %d %c\\n\""@},
30463 @{name="arg",value="0x2"@}],file="vprintf.c",line="31",arch="i386:x86_64"@}
30467 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30468 @node GDB/MI Ada Tasking Commands
30469 @section @sc{gdb/mi} Ada Tasking Commands
30471 @subheading The @code{-ada-task-info} Command
30472 @findex -ada-task-info
30474 @subsubheading Synopsis
30477 -ada-task-info [ @var{task-id} ]
30480 Reports information about either a specific Ada task, if the
30481 @var{task-id} parameter is present, or about all Ada tasks.
30483 @subsubheading @value{GDBN} Command
30485 The @samp{info tasks} command prints the same information
30486 about all Ada tasks (@pxref{Ada Tasks}).
30488 @subsubheading Result
30490 The result is a table of Ada tasks. The following columns are
30491 defined for each Ada task:
30495 This field exists only for the current thread. It has the value @samp{*}.
30498 The identifier that @value{GDBN} uses to refer to the Ada task.
30501 The identifier that the target uses to refer to the Ada task.
30504 The global thread identifier of the thread corresponding to the Ada
30507 This field should always exist, as Ada tasks are always implemented
30508 on top of a thread. But if @value{GDBN} cannot find this corresponding
30509 thread for any reason, the field is omitted.
30512 This field exists only when the task was created by another task.
30513 In this case, it provides the ID of the parent task.
30516 The base priority of the task.
30519 The current state of the task. For a detailed description of the
30520 possible states, see @ref{Ada Tasks}.
30523 The name of the task.
30527 @subsubheading Example
30531 ^done,tasks=@{nr_rows="3",nr_cols="8",
30532 hdr=[@{width="1",alignment="-1",col_name="current",colhdr=""@},
30533 @{width="3",alignment="1",col_name="id",colhdr="ID"@},
30534 @{width="9",alignment="1",col_name="task-id",colhdr="TID"@},
30535 @{width="4",alignment="1",col_name="thread-id",colhdr=""@},
30536 @{width="4",alignment="1",col_name="parent-id",colhdr="P-ID"@},
30537 @{width="3",alignment="1",col_name="priority",colhdr="Pri"@},
30538 @{width="22",alignment="-1",col_name="state",colhdr="State"@},
30539 @{width="1",alignment="2",col_name="name",colhdr="Name"@}],
30540 body=[@{current="*",id="1",task-id=" 644010",thread-id="1",priority="48",
30541 state="Child Termination Wait",name="main_task"@}]@}
30545 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
30546 @node GDB/MI Program Execution
30547 @section @sc{gdb/mi} Program Execution
30549 These are the asynchronous commands which generate the out-of-band
30550 record @samp{*stopped}. Currently @value{GDBN} only really executes
30551 asynchronously with remote targets and this interaction is mimicked in
30554 @subheading The @code{-exec-continue} Command
30555 @findex -exec-continue
30557 @subsubheading Synopsis
30560 -exec-continue [--reverse] [--all|--thread-group N]
30563 Resumes the execution of the inferior program, which will continue
30564 to execute until it reaches a debugger stop event. If the
30565 @samp{--reverse} option is specified, execution resumes in reverse until
30566 it reaches a stop event. Stop events may include
30569 breakpoints or watchpoints
30571 signals or exceptions
30573 the end of the process (or its beginning under @samp{--reverse})
30575 the end or beginning of a replay log if one is being used.
30577 In all-stop mode (@pxref{All-Stop
30578 Mode}), may resume only one thread, or all threads, depending on the
30579 value of the @samp{scheduler-locking} variable. If @samp{--all} is
30580 specified, all threads (in all inferiors) will be resumed. The @samp{--all} option is
30581 ignored in all-stop mode. If the @samp{--thread-group} options is
30582 specified, then all threads in that thread group are resumed.
30584 @subsubheading @value{GDBN} Command
30586 The corresponding @value{GDBN} corresponding is @samp{continue}.
30588 @subsubheading Example
30595 *stopped,reason="breakpoint-hit",disp="keep",bkptno="2",frame=@{
30596 func="foo",args=[],file="hello.c",fullname="/home/foo/bar/hello.c",
30597 line="13",arch="i386:x86_64"@}
30602 @subheading The @code{-exec-finish} Command
30603 @findex -exec-finish
30605 @subsubheading Synopsis
30608 -exec-finish [--reverse]
30611 Resumes the execution of the inferior program until the current
30612 function is exited. Displays the results returned by the function.
30613 If the @samp{--reverse} option is specified, resumes the reverse
30614 execution of the inferior program until the point where current
30615 function was called.
30617 @subsubheading @value{GDBN} Command
30619 The corresponding @value{GDBN} command is @samp{finish}.
30621 @subsubheading Example
30623 Function returning @code{void}.
30630 *stopped,reason="function-finished",frame=@{func="main",args=[],
30631 file="hello.c",fullname="/home/foo/bar/hello.c",line="7",arch="i386:x86_64"@}
30635 Function returning other than @code{void}. The name of the internal
30636 @value{GDBN} variable storing the result is printed, together with the
30643 *stopped,reason="function-finished",frame=@{addr="0x000107b0",func="foo",
30644 args=[@{name="a",value="1"],@{name="b",value="9"@}@},
30645 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
30646 arch="i386:x86_64"@},
30647 gdb-result-var="$1",return-value="0"
30652 @subheading The @code{-exec-interrupt} Command
30653 @findex -exec-interrupt
30655 @subsubheading Synopsis
30658 -exec-interrupt [--all|--thread-group N]
30661 Interrupts the background execution of the target. Note how the token
30662 associated with the stop message is the one for the execution command
30663 that has been interrupted. The token for the interrupt itself only
30664 appears in the @samp{^done} output. If the user is trying to
30665 interrupt a non-running program, an error message will be printed.
30667 Note that when asynchronous execution is enabled, this command is
30668 asynchronous just like other execution commands. That is, first the
30669 @samp{^done} response will be printed, and the target stop will be
30670 reported after that using the @samp{*stopped} notification.
30672 In non-stop mode, only the context thread is interrupted by default.
30673 All threads (in all inferiors) will be interrupted if the
30674 @samp{--all} option is specified. If the @samp{--thread-group}
30675 option is specified, all threads in that group will be interrupted.
30677 @subsubheading @value{GDBN} Command
30679 The corresponding @value{GDBN} command is @samp{interrupt}.
30681 @subsubheading Example
30692 111*stopped,signal-name="SIGINT",signal-meaning="Interrupt",
30693 frame=@{addr="0x00010140",func="foo",args=[],file="try.c",
30694 fullname="/home/foo/bar/try.c",line="13",arch="i386:x86_64"@}
30699 ^error,msg="mi_cmd_exec_interrupt: Inferior not executing."
30703 @subheading The @code{-exec-jump} Command
30706 @subsubheading Synopsis
30709 -exec-jump @var{location}
30712 Resumes execution of the inferior program at the location specified by
30713 parameter. @xref{Specify Location}, for a description of the
30714 different forms of @var{location}.
30716 @subsubheading @value{GDBN} Command
30718 The corresponding @value{GDBN} command is @samp{jump}.
30720 @subsubheading Example
30723 -exec-jump foo.c:10
30724 *running,thread-id="all"
30729 @subheading The @code{-exec-next} Command
30732 @subsubheading Synopsis
30735 -exec-next [--reverse]
30738 Resumes execution of the inferior program, stopping when the beginning
30739 of the next source line is reached.
30741 If the @samp{--reverse} option is specified, resumes reverse execution
30742 of the inferior program, stopping at the beginning of the previous
30743 source line. If you issue this command on the first line of a
30744 function, it will take you back to the caller of that function, to the
30745 source line where the function was called.
30748 @subsubheading @value{GDBN} Command
30750 The corresponding @value{GDBN} command is @samp{next}.
30752 @subsubheading Example
30758 *stopped,reason="end-stepping-range",line="8",file="hello.c"
30763 @subheading The @code{-exec-next-instruction} Command
30764 @findex -exec-next-instruction
30766 @subsubheading Synopsis
30769 -exec-next-instruction [--reverse]
30772 Executes one machine instruction. If the instruction is a function
30773 call, continues until the function returns. If the program stops at an
30774 instruction in the middle of a source line, the address will be
30777 If the @samp{--reverse} option is specified, resumes reverse execution
30778 of the inferior program, stopping at the previous instruction. If the
30779 previously executed instruction was a return from another function,
30780 it will continue to execute in reverse until the call to that function
30781 (from the current stack frame) is reached.
30783 @subsubheading @value{GDBN} Command
30785 The corresponding @value{GDBN} command is @samp{nexti}.
30787 @subsubheading Example
30791 -exec-next-instruction
30795 *stopped,reason="end-stepping-range",
30796 addr="0x000100d4",line="5",file="hello.c"
30801 @subheading The @code{-exec-return} Command
30802 @findex -exec-return
30804 @subsubheading Synopsis
30810 Makes current function return immediately. Doesn't execute the inferior.
30811 Displays the new current frame.
30813 @subsubheading @value{GDBN} Command
30815 The corresponding @value{GDBN} command is @samp{return}.
30817 @subsubheading Example
30821 200-break-insert callee4
30822 200^done,bkpt=@{number="1",addr="0x00010734",
30823 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",line="8"@}
30828 000*stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
30829 frame=@{func="callee4",args=[],
30830 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30831 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
30832 arch="i386:x86_64"@}
30838 111^done,frame=@{level="0",func="callee3",
30839 args=[@{name="strarg",
30840 value="0x11940 \"A string argument.\""@}],
30841 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
30842 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="18",
30843 arch="i386:x86_64"@}
30848 @subheading The @code{-exec-run} Command
30851 @subsubheading Synopsis
30854 -exec-run [ --all | --thread-group N ] [ --start ]
30857 Starts execution of the inferior from the beginning. The inferior
30858 executes until either a breakpoint is encountered or the program
30859 exits. In the latter case the output will include an exit code, if
30860 the program has exited exceptionally.
30862 When neither the @samp{--all} nor the @samp{--thread-group} option
30863 is specified, the current inferior is started. If the
30864 @samp{--thread-group} option is specified, it should refer to a thread
30865 group of type @samp{process}, and that thread group will be started.
30866 If the @samp{--all} option is specified, then all inferiors will be started.
30868 Using the @samp{--start} option instructs the debugger to stop
30869 the execution at the start of the inferior's main subprogram,
30870 following the same behavior as the @code{start} command
30871 (@pxref{Starting}).
30873 @subsubheading @value{GDBN} Command
30875 The corresponding @value{GDBN} command is @samp{run}.
30877 @subsubheading Examples
30882 ^done,bkpt=@{number="1",addr="0x0001072c",file="recursive2.c",line="4"@}
30887 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",
30888 frame=@{func="main",args=[],file="recursive2.c",
30889 fullname="/home/foo/bar/recursive2.c",line="4",arch="i386:x86_64"@}
30894 Program exited normally:
30902 *stopped,reason="exited-normally"
30907 Program exited exceptionally:
30915 *stopped,reason="exited",exit-code="01"
30919 Another way the program can terminate is if it receives a signal such as
30920 @code{SIGINT}. In this case, @sc{gdb/mi} displays this:
30924 *stopped,reason="exited-signalled",signal-name="SIGINT",
30925 signal-meaning="Interrupt"
30929 @c @subheading -exec-signal
30932 @subheading The @code{-exec-step} Command
30935 @subsubheading Synopsis
30938 -exec-step [--reverse]
30941 Resumes execution of the inferior program, stopping when the beginning
30942 of the next source line is reached, if the next source line is not a
30943 function call. If it is, stop at the first instruction of the called
30944 function. If the @samp{--reverse} option is specified, resumes reverse
30945 execution of the inferior program, stopping at the beginning of the
30946 previously executed source line.
30948 @subsubheading @value{GDBN} Command
30950 The corresponding @value{GDBN} command is @samp{step}.
30952 @subsubheading Example
30954 Stepping into a function:
30960 *stopped,reason="end-stepping-range",
30961 frame=@{func="foo",args=[@{name="a",value="10"@},
30962 @{name="b",value="0"@}],file="recursive2.c",
30963 fullname="/home/foo/bar/recursive2.c",line="11",arch="i386:x86_64"@}
30973 *stopped,reason="end-stepping-range",line="14",file="recursive2.c"
30978 @subheading The @code{-exec-step-instruction} Command
30979 @findex -exec-step-instruction
30981 @subsubheading Synopsis
30984 -exec-step-instruction [--reverse]
30987 Resumes the inferior which executes one machine instruction. If the
30988 @samp{--reverse} option is specified, resumes reverse execution of the
30989 inferior program, stopping at the previously executed instruction.
30990 The output, once @value{GDBN} has stopped, will vary depending on
30991 whether we have stopped in the middle of a source line or not. In the
30992 former case, the address at which the program stopped will be printed
30995 @subsubheading @value{GDBN} Command
30997 The corresponding @value{GDBN} command is @samp{stepi}.
30999 @subsubheading Example
31003 -exec-step-instruction
31007 *stopped,reason="end-stepping-range",
31008 frame=@{func="foo",args=[],file="try.c",
31009 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
31011 -exec-step-instruction
31015 *stopped,reason="end-stepping-range",
31016 frame=@{addr="0x000100f4",func="foo",args=[],file="try.c",
31017 fullname="/home/foo/bar/try.c",line="10",arch="i386:x86_64"@}
31022 @subheading The @code{-exec-until} Command
31023 @findex -exec-until
31025 @subsubheading Synopsis
31028 -exec-until [ @var{location} ]
31031 Executes the inferior until the @var{location} specified in the
31032 argument is reached. If there is no argument, the inferior executes
31033 until a source line greater than the current one is reached. The
31034 reason for stopping in this case will be @samp{location-reached}.
31036 @subsubheading @value{GDBN} Command
31038 The corresponding @value{GDBN} command is @samp{until}.
31040 @subsubheading Example
31044 -exec-until recursive2.c:6
31048 *stopped,reason="location-reached",frame=@{func="main",args=[],
31049 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="6",
31050 arch="i386:x86_64"@}
31055 @subheading -file-clear
31056 Is this going away????
31059 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31060 @node GDB/MI Stack Manipulation
31061 @section @sc{gdb/mi} Stack Manipulation Commands
31063 @subheading The @code{-enable-frame-filters} Command
31064 @findex -enable-frame-filters
31067 -enable-frame-filters
31070 @value{GDBN} allows Python-based frame filters to affect the output of
31071 the MI commands relating to stack traces. As there is no way to
31072 implement this in a fully backward-compatible way, a front end must
31073 request that this functionality be enabled.
31075 Once enabled, this feature cannot be disabled.
31077 Note that if Python support has not been compiled into @value{GDBN},
31078 this command will still succeed (and do nothing).
31080 @subheading The @code{-stack-info-frame} Command
31081 @findex -stack-info-frame
31083 @subsubheading Synopsis
31089 Get info on the selected frame.
31091 @subsubheading @value{GDBN} Command
31093 The corresponding @value{GDBN} command is @samp{info frame} or @samp{frame}
31094 (without arguments).
31096 @subsubheading Example
31101 ^done,frame=@{level="1",addr="0x0001076c",func="callee3",
31102 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31103 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
31104 arch="i386:x86_64"@}
31108 @subheading The @code{-stack-info-depth} Command
31109 @findex -stack-info-depth
31111 @subsubheading Synopsis
31114 -stack-info-depth [ @var{max-depth} ]
31117 Return the depth of the stack. If the integer argument @var{max-depth}
31118 is specified, do not count beyond @var{max-depth} frames.
31120 @subsubheading @value{GDBN} Command
31122 There's no equivalent @value{GDBN} command.
31124 @subsubheading Example
31126 For a stack with frame levels 0 through 11:
31133 -stack-info-depth 4
31136 -stack-info-depth 12
31139 -stack-info-depth 11
31142 -stack-info-depth 13
31147 @anchor{-stack-list-arguments}
31148 @subheading The @code{-stack-list-arguments} Command
31149 @findex -stack-list-arguments
31151 @subsubheading Synopsis
31154 -stack-list-arguments [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31155 [ @var{low-frame} @var{high-frame} ]
31158 Display a list of the arguments for the frames between @var{low-frame}
31159 and @var{high-frame} (inclusive). If @var{low-frame} and
31160 @var{high-frame} are not provided, list the arguments for the whole
31161 call stack. If the two arguments are equal, show the single frame
31162 at the corresponding level. It is an error if @var{low-frame} is
31163 larger than the actual number of frames. On the other hand,
31164 @var{high-frame} may be larger than the actual number of frames, in
31165 which case only existing frames will be returned.
31167 If @var{print-values} is 0 or @code{--no-values}, print only the names of
31168 the variables; if it is 1 or @code{--all-values}, print also their
31169 values; and if it is 2 or @code{--simple-values}, print the name,
31170 type and value for simple data types, and the name and type for arrays,
31171 structures and unions. If the option @code{--no-frame-filters} is
31172 supplied, then Python frame filters will not be executed.
31174 If the @code{--skip-unavailable} option is specified, arguments that
31175 are not available are not listed. Partially available arguments
31176 are still displayed, however.
31178 Use of this command to obtain arguments in a single frame is
31179 deprecated in favor of the @samp{-stack-list-variables} command.
31181 @subsubheading @value{GDBN} Command
31183 @value{GDBN} does not have an equivalent command. @code{gdbtk} has a
31184 @samp{gdb_get_args} command which partially overlaps with the
31185 functionality of @samp{-stack-list-arguments}.
31187 @subsubheading Example
31194 frame=@{level="0",addr="0x00010734",func="callee4",
31195 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31196 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="8",
31197 arch="i386:x86_64"@},
31198 frame=@{level="1",addr="0x0001076c",func="callee3",
31199 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31200 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="17",
31201 arch="i386:x86_64"@},
31202 frame=@{level="2",addr="0x0001078c",func="callee2",
31203 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31204 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="22",
31205 arch="i386:x86_64"@},
31206 frame=@{level="3",addr="0x000107b4",func="callee1",
31207 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31208 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="27",
31209 arch="i386:x86_64"@},
31210 frame=@{level="4",addr="0x000107e0",func="main",
31211 file="../../../devo/gdb/testsuite/gdb.mi/basics.c",
31212 fullname="/home/foo/bar/devo/gdb/testsuite/gdb.mi/basics.c",line="32",
31213 arch="i386:x86_64"@}]
31215 -stack-list-arguments 0
31218 frame=@{level="0",args=[]@},
31219 frame=@{level="1",args=[name="strarg"]@},
31220 frame=@{level="2",args=[name="intarg",name="strarg"]@},
31221 frame=@{level="3",args=[name="intarg",name="strarg",name="fltarg"]@},
31222 frame=@{level="4",args=[]@}]
31224 -stack-list-arguments 1
31227 frame=@{level="0",args=[]@},
31229 args=[@{name="strarg",value="0x11940 \"A string argument.\""@}]@},
31230 frame=@{level="2",args=[
31231 @{name="intarg",value="2"@},
31232 @{name="strarg",value="0x11940 \"A string argument.\""@}]@},
31233 @{frame=@{level="3",args=[
31234 @{name="intarg",value="2"@},
31235 @{name="strarg",value="0x11940 \"A string argument.\""@},
31236 @{name="fltarg",value="3.5"@}]@},
31237 frame=@{level="4",args=[]@}]
31239 -stack-list-arguments 0 2 2
31240 ^done,stack-args=[frame=@{level="2",args=[name="intarg",name="strarg"]@}]
31242 -stack-list-arguments 1 2 2
31243 ^done,stack-args=[frame=@{level="2",
31244 args=[@{name="intarg",value="2"@},
31245 @{name="strarg",value="0x11940 \"A string argument.\""@}]@}]
31249 @c @subheading -stack-list-exception-handlers
31252 @anchor{-stack-list-frames}
31253 @subheading The @code{-stack-list-frames} Command
31254 @findex -stack-list-frames
31256 @subsubheading Synopsis
31259 -stack-list-frames [ --no-frame-filters @var{low-frame} @var{high-frame} ]
31262 List the frames currently on the stack. For each frame it displays the
31267 The frame number, 0 being the topmost frame, i.e., the innermost function.
31269 The @code{$pc} value for that frame.
31273 File name of the source file where the function lives.
31274 @item @var{fullname}
31275 The full file name of the source file where the function lives.
31277 Line number corresponding to the @code{$pc}.
31279 The shared library where this function is defined. This is only given
31280 if the frame's function is not known.
31282 Frame's architecture.
31285 If invoked without arguments, this command prints a backtrace for the
31286 whole stack. If given two integer arguments, it shows the frames whose
31287 levels are between the two arguments (inclusive). If the two arguments
31288 are equal, it shows the single frame at the corresponding level. It is
31289 an error if @var{low-frame} is larger than the actual number of
31290 frames. On the other hand, @var{high-frame} may be larger than the
31291 actual number of frames, in which case only existing frames will be
31292 returned. If the option @code{--no-frame-filters} is supplied, then
31293 Python frame filters will not be executed.
31295 @subsubheading @value{GDBN} Command
31297 The corresponding @value{GDBN} commands are @samp{backtrace} and @samp{where}.
31299 @subsubheading Example
31301 Full stack backtrace:
31307 [frame=@{level="0",addr="0x0001076c",func="foo",
31308 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="11",
31309 arch="i386:x86_64"@},
31310 frame=@{level="1",addr="0x000107a4",func="foo",
31311 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31312 arch="i386:x86_64"@},
31313 frame=@{level="2",addr="0x000107a4",func="foo",
31314 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31315 arch="i386:x86_64"@},
31316 frame=@{level="3",addr="0x000107a4",func="foo",
31317 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31318 arch="i386:x86_64"@},
31319 frame=@{level="4",addr="0x000107a4",func="foo",
31320 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31321 arch="i386:x86_64"@},
31322 frame=@{level="5",addr="0x000107a4",func="foo",
31323 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31324 arch="i386:x86_64"@},
31325 frame=@{level="6",addr="0x000107a4",func="foo",
31326 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31327 arch="i386:x86_64"@},
31328 frame=@{level="7",addr="0x000107a4",func="foo",
31329 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31330 arch="i386:x86_64"@},
31331 frame=@{level="8",addr="0x000107a4",func="foo",
31332 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31333 arch="i386:x86_64"@},
31334 frame=@{level="9",addr="0x000107a4",func="foo",
31335 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31336 arch="i386:x86_64"@},
31337 frame=@{level="10",addr="0x000107a4",func="foo",
31338 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31339 arch="i386:x86_64"@},
31340 frame=@{level="11",addr="0x00010738",func="main",
31341 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="4",
31342 arch="i386:x86_64"@}]
31346 Show frames between @var{low_frame} and @var{high_frame}:
31350 -stack-list-frames 3 5
31352 [frame=@{level="3",addr="0x000107a4",func="foo",
31353 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31354 arch="i386:x86_64"@},
31355 frame=@{level="4",addr="0x000107a4",func="foo",
31356 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31357 arch="i386:x86_64"@},
31358 frame=@{level="5",addr="0x000107a4",func="foo",
31359 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31360 arch="i386:x86_64"@}]
31364 Show a single frame:
31368 -stack-list-frames 3 3
31370 [frame=@{level="3",addr="0x000107a4",func="foo",
31371 file="recursive2.c",fullname="/home/foo/bar/recursive2.c",line="14",
31372 arch="i386:x86_64"@}]
31377 @subheading The @code{-stack-list-locals} Command
31378 @findex -stack-list-locals
31379 @anchor{-stack-list-locals}
31381 @subsubheading Synopsis
31384 -stack-list-locals [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31387 Display the local variable names for the selected frame. If
31388 @var{print-values} is 0 or @code{--no-values}, print only the names of
31389 the variables; if it is 1 or @code{--all-values}, print also their
31390 values; and if it is 2 or @code{--simple-values}, print the name,
31391 type and value for simple data types, and the name and type for arrays,
31392 structures and unions. In this last case, a frontend can immediately
31393 display the value of simple data types and create variable objects for
31394 other data types when the user wishes to explore their values in
31395 more detail. If the option @code{--no-frame-filters} is supplied, then
31396 Python frame filters will not be executed.
31398 If the @code{--skip-unavailable} option is specified, local variables
31399 that are not available are not listed. Partially available local
31400 variables are still displayed, however.
31402 This command is deprecated in favor of the
31403 @samp{-stack-list-variables} command.
31405 @subsubheading @value{GDBN} Command
31407 @samp{info locals} in @value{GDBN}, @samp{gdb_get_locals} in @code{gdbtk}.
31409 @subsubheading Example
31413 -stack-list-locals 0
31414 ^done,locals=[name="A",name="B",name="C"]
31416 -stack-list-locals --all-values
31417 ^done,locals=[@{name="A",value="1"@},@{name="B",value="2"@},
31418 @{name="C",value="@{1, 2, 3@}"@}]
31419 -stack-list-locals --simple-values
31420 ^done,locals=[@{name="A",type="int",value="1"@},
31421 @{name="B",type="int",value="2"@},@{name="C",type="int [3]"@}]
31425 @anchor{-stack-list-variables}
31426 @subheading The @code{-stack-list-variables} Command
31427 @findex -stack-list-variables
31429 @subsubheading Synopsis
31432 -stack-list-variables [ --no-frame-filters ] [ --skip-unavailable ] @var{print-values}
31435 Display the names of local variables and function arguments for the selected frame. If
31436 @var{print-values} is 0 or @code{--no-values}, print only the names of
31437 the variables; if it is 1 or @code{--all-values}, print also their
31438 values; and if it is 2 or @code{--simple-values}, print the name,
31439 type and value for simple data types, and the name and type for arrays,
31440 structures and unions. If the option @code{--no-frame-filters} is
31441 supplied, then Python frame filters will not be executed.
31443 If the @code{--skip-unavailable} option is specified, local variables
31444 and arguments that are not available are not listed. Partially
31445 available arguments and local variables are still displayed, however.
31447 @subsubheading Example
31451 -stack-list-variables --thread 1 --frame 0 --all-values
31452 ^done,variables=[@{name="x",value="11"@},@{name="s",value="@{a = 1, b = 2@}"@}]
31457 @subheading The @code{-stack-select-frame} Command
31458 @findex -stack-select-frame
31460 @subsubheading Synopsis
31463 -stack-select-frame @var{framenum}
31466 Change the selected frame. Select a different frame @var{framenum} on
31469 This command in deprecated in favor of passing the @samp{--frame}
31470 option to every command.
31472 @subsubheading @value{GDBN} Command
31474 The corresponding @value{GDBN} commands are @samp{frame}, @samp{up},
31475 @samp{down}, @samp{select-frame}, @samp{up-silent}, and @samp{down-silent}.
31477 @subsubheading Example
31481 -stack-select-frame 2
31486 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
31487 @node GDB/MI Variable Objects
31488 @section @sc{gdb/mi} Variable Objects
31492 @subheading Motivation for Variable Objects in @sc{gdb/mi}
31494 For the implementation of a variable debugger window (locals, watched
31495 expressions, etc.), we are proposing the adaptation of the existing code
31496 used by @code{Insight}.
31498 The two main reasons for that are:
31502 It has been proven in practice (it is already on its second generation).
31505 It will shorten development time (needless to say how important it is
31509 The original interface was designed to be used by Tcl code, so it was
31510 slightly changed so it could be used through @sc{gdb/mi}. This section
31511 describes the @sc{gdb/mi} operations that will be available and gives some
31512 hints about their use.
31514 @emph{Note}: In addition to the set of operations described here, we
31515 expect the @sc{gui} implementation of a variable window to require, at
31516 least, the following operations:
31519 @item @code{-gdb-show} @code{output-radix}
31520 @item @code{-stack-list-arguments}
31521 @item @code{-stack-list-locals}
31522 @item @code{-stack-select-frame}
31527 @subheading Introduction to Variable Objects
31529 @cindex variable objects in @sc{gdb/mi}
31531 Variable objects are "object-oriented" MI interface for examining and
31532 changing values of expressions. Unlike some other MI interfaces that
31533 work with expressions, variable objects are specifically designed for
31534 simple and efficient presentation in the frontend. A variable object
31535 is identified by string name. When a variable object is created, the
31536 frontend specifies the expression for that variable object. The
31537 expression can be a simple variable, or it can be an arbitrary complex
31538 expression, and can even involve CPU registers. After creating a
31539 variable object, the frontend can invoke other variable object
31540 operations---for example to obtain or change the value of a variable
31541 object, or to change display format.
31543 Variable objects have hierarchical tree structure. Any variable object
31544 that corresponds to a composite type, such as structure in C, has
31545 a number of child variable objects, for example corresponding to each
31546 element of a structure. A child variable object can itself have
31547 children, recursively. Recursion ends when we reach
31548 leaf variable objects, which always have built-in types. Child variable
31549 objects are created only by explicit request, so if a frontend
31550 is not interested in the children of a particular variable object, no
31551 child will be created.
31553 For a leaf variable object it is possible to obtain its value as a
31554 string, or set the value from a string. String value can be also
31555 obtained for a non-leaf variable object, but it's generally a string
31556 that only indicates the type of the object, and does not list its
31557 contents. Assignment to a non-leaf variable object is not allowed.
31559 A frontend does not need to read the values of all variable objects each time
31560 the program stops. Instead, MI provides an update command that lists all
31561 variable objects whose values has changed since the last update
31562 operation. This considerably reduces the amount of data that must
31563 be transferred to the frontend. As noted above, children variable
31564 objects are created on demand, and only leaf variable objects have a
31565 real value. As result, gdb will read target memory only for leaf
31566 variables that frontend has created.
31568 The automatic update is not always desirable. For example, a frontend
31569 might want to keep a value of some expression for future reference,
31570 and never update it. For another example, fetching memory is
31571 relatively slow for embedded targets, so a frontend might want
31572 to disable automatic update for the variables that are either not
31573 visible on the screen, or ``closed''. This is possible using so
31574 called ``frozen variable objects''. Such variable objects are never
31575 implicitly updated.
31577 Variable objects can be either @dfn{fixed} or @dfn{floating}. For the
31578 fixed variable object, the expression is parsed when the variable
31579 object is created, including associating identifiers to specific
31580 variables. The meaning of expression never changes. For a floating
31581 variable object the values of variables whose names appear in the
31582 expressions are re-evaluated every time in the context of the current
31583 frame. Consider this example:
31588 struct work_state state;
31595 If a fixed variable object for the @code{state} variable is created in
31596 this function, and we enter the recursive call, the variable
31597 object will report the value of @code{state} in the top-level
31598 @code{do_work} invocation. On the other hand, a floating variable
31599 object will report the value of @code{state} in the current frame.
31601 If an expression specified when creating a fixed variable object
31602 refers to a local variable, the variable object becomes bound to the
31603 thread and frame in which the variable object is created. When such
31604 variable object is updated, @value{GDBN} makes sure that the
31605 thread/frame combination the variable object is bound to still exists,
31606 and re-evaluates the variable object in context of that thread/frame.
31608 The following is the complete set of @sc{gdb/mi} operations defined to
31609 access this functionality:
31611 @multitable @columnfractions .4 .6
31612 @item @strong{Operation}
31613 @tab @strong{Description}
31615 @item @code{-enable-pretty-printing}
31616 @tab enable Python-based pretty-printing
31617 @item @code{-var-create}
31618 @tab create a variable object
31619 @item @code{-var-delete}
31620 @tab delete the variable object and/or its children
31621 @item @code{-var-set-format}
31622 @tab set the display format of this variable
31623 @item @code{-var-show-format}
31624 @tab show the display format of this variable
31625 @item @code{-var-info-num-children}
31626 @tab tells how many children this object has
31627 @item @code{-var-list-children}
31628 @tab return a list of the object's children
31629 @item @code{-var-info-type}
31630 @tab show the type of this variable object
31631 @item @code{-var-info-expression}
31632 @tab print parent-relative expression that this variable object represents
31633 @item @code{-var-info-path-expression}
31634 @tab print full expression that this variable object represents
31635 @item @code{-var-show-attributes}
31636 @tab is this variable editable? does it exist here?
31637 @item @code{-var-evaluate-expression}
31638 @tab get the value of this variable
31639 @item @code{-var-assign}
31640 @tab set the value of this variable
31641 @item @code{-var-update}
31642 @tab update the variable and its children
31643 @item @code{-var-set-frozen}
31644 @tab set frozeness attribute
31645 @item @code{-var-set-update-range}
31646 @tab set range of children to display on update
31649 In the next subsection we describe each operation in detail and suggest
31650 how it can be used.
31652 @subheading Description And Use of Operations on Variable Objects
31654 @subheading The @code{-enable-pretty-printing} Command
31655 @findex -enable-pretty-printing
31658 -enable-pretty-printing
31661 @value{GDBN} allows Python-based visualizers to affect the output of the
31662 MI variable object commands. However, because there was no way to
31663 implement this in a fully backward-compatible way, a front end must
31664 request that this functionality be enabled.
31666 Once enabled, this feature cannot be disabled.
31668 Note that if Python support has not been compiled into @value{GDBN},
31669 this command will still succeed (and do nothing).
31671 This feature is currently (as of @value{GDBN} 7.0) experimental, and
31672 may work differently in future versions of @value{GDBN}.
31674 @subheading The @code{-var-create} Command
31675 @findex -var-create
31677 @subsubheading Synopsis
31680 -var-create @{@var{name} | "-"@}
31681 @{@var{frame-addr} | "*" | "@@"@} @var{expression}
31684 This operation creates a variable object, which allows the monitoring of
31685 a variable, the result of an expression, a memory cell or a CPU
31688 The @var{name} parameter is the string by which the object can be
31689 referenced. It must be unique. If @samp{-} is specified, the varobj
31690 system will generate a string ``varNNNNNN'' automatically. It will be
31691 unique provided that one does not specify @var{name} of that format.
31692 The command fails if a duplicate name is found.
31694 The frame under which the expression should be evaluated can be
31695 specified by @var{frame-addr}. A @samp{*} indicates that the current
31696 frame should be used. A @samp{@@} indicates that a floating variable
31697 object must be created.
31699 @var{expression} is any expression valid on the current language set (must not
31700 begin with a @samp{*}), or one of the following:
31704 @samp{*@var{addr}}, where @var{addr} is the address of a memory cell
31707 @samp{*@var{addr}-@var{addr}} --- a memory address range (TBD)
31710 @samp{$@var{regname}} --- a CPU register name
31713 @cindex dynamic varobj
31714 A varobj's contents may be provided by a Python-based pretty-printer. In this
31715 case the varobj is known as a @dfn{dynamic varobj}. Dynamic varobjs
31716 have slightly different semantics in some cases. If the
31717 @code{-enable-pretty-printing} command is not sent, then @value{GDBN}
31718 will never create a dynamic varobj. This ensures backward
31719 compatibility for existing clients.
31721 @subsubheading Result
31723 This operation returns attributes of the newly-created varobj. These
31728 The name of the varobj.
31731 The number of children of the varobj. This number is not necessarily
31732 reliable for a dynamic varobj. Instead, you must examine the
31733 @samp{has_more} attribute.
31736 The varobj's scalar value. For a varobj whose type is some sort of
31737 aggregate (e.g., a @code{struct}), or for a dynamic varobj, this value
31738 will not be interesting.
31741 The varobj's type. This is a string representation of the type, as
31742 would be printed by the @value{GDBN} CLI. If @samp{print object}
31743 (@pxref{Print Settings, set print object}) is set to @code{on}, the
31744 @emph{actual} (derived) type of the object is shown rather than the
31745 @emph{declared} one.
31748 If a variable object is bound to a specific thread, then this is the
31749 thread's global identifier.
31752 For a dynamic varobj, this indicates whether there appear to be any
31753 children available. For a non-dynamic varobj, this will be 0.
31756 This attribute will be present and have the value @samp{1} if the
31757 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
31758 then this attribute will not be present.
31761 A dynamic varobj can supply a display hint to the front end. The
31762 value comes directly from the Python pretty-printer object's
31763 @code{display_hint} method. @xref{Pretty Printing API}.
31766 Typical output will look like this:
31769 name="@var{name}",numchild="@var{N}",type="@var{type}",thread-id="@var{M}",
31770 has_more="@var{has_more}"
31774 @subheading The @code{-var-delete} Command
31775 @findex -var-delete
31777 @subsubheading Synopsis
31780 -var-delete [ -c ] @var{name}
31783 Deletes a previously created variable object and all of its children.
31784 With the @samp{-c} option, just deletes the children.
31786 Returns an error if the object @var{name} is not found.
31789 @subheading The @code{-var-set-format} Command
31790 @findex -var-set-format
31792 @subsubheading Synopsis
31795 -var-set-format @var{name} @var{format-spec}
31798 Sets the output format for the value of the object @var{name} to be
31801 @anchor{-var-set-format}
31802 The syntax for the @var{format-spec} is as follows:
31805 @var{format-spec} @expansion{}
31806 @{binary | decimal | hexadecimal | octal | natural | zero-hexadecimal@}
31809 The natural format is the default format choosen automatically
31810 based on the variable type (like decimal for an @code{int}, hex
31811 for pointers, etc.).
31813 The zero-hexadecimal format has a representation similar to hexadecimal
31814 but with padding zeroes to the left of the value. For example, a 32-bit
31815 hexadecimal value of 0x1234 would be represented as 0x00001234 in the
31816 zero-hexadecimal format.
31818 For a variable with children, the format is set only on the
31819 variable itself, and the children are not affected.
31821 @subheading The @code{-var-show-format} Command
31822 @findex -var-show-format
31824 @subsubheading Synopsis
31827 -var-show-format @var{name}
31830 Returns the format used to display the value of the object @var{name}.
31833 @var{format} @expansion{}
31838 @subheading The @code{-var-info-num-children} Command
31839 @findex -var-info-num-children
31841 @subsubheading Synopsis
31844 -var-info-num-children @var{name}
31847 Returns the number of children of a variable object @var{name}:
31853 Note that this number is not completely reliable for a dynamic varobj.
31854 It will return the current number of children, but more children may
31858 @subheading The @code{-var-list-children} Command
31859 @findex -var-list-children
31861 @subsubheading Synopsis
31864 -var-list-children [@var{print-values}] @var{name} [@var{from} @var{to}]
31866 @anchor{-var-list-children}
31868 Return a list of the children of the specified variable object and
31869 create variable objects for them, if they do not already exist. With
31870 a single argument or if @var{print-values} has a value of 0 or
31871 @code{--no-values}, print only the names of the variables; if
31872 @var{print-values} is 1 or @code{--all-values}, also print their
31873 values; and if it is 2 or @code{--simple-values} print the name and
31874 value for simple data types and just the name for arrays, structures
31877 @var{from} and @var{to}, if specified, indicate the range of children
31878 to report. If @var{from} or @var{to} is less than zero, the range is
31879 reset and all children will be reported. Otherwise, children starting
31880 at @var{from} (zero-based) and up to and excluding @var{to} will be
31883 If a child range is requested, it will only affect the current call to
31884 @code{-var-list-children}, but not future calls to @code{-var-update}.
31885 For this, you must instead use @code{-var-set-update-range}. The
31886 intent of this approach is to enable a front end to implement any
31887 update approach it likes; for example, scrolling a view may cause the
31888 front end to request more children with @code{-var-list-children}, and
31889 then the front end could call @code{-var-set-update-range} with a
31890 different range to ensure that future updates are restricted to just
31893 For each child the following results are returned:
31898 Name of the variable object created for this child.
31901 The expression to be shown to the user by the front end to designate this child.
31902 For example this may be the name of a structure member.
31904 For a dynamic varobj, this value cannot be used to form an
31905 expression. There is no way to do this at all with a dynamic varobj.
31907 For C/C@t{++} structures there are several pseudo children returned to
31908 designate access qualifiers. For these pseudo children @var{exp} is
31909 @samp{public}, @samp{private}, or @samp{protected}. In this case the
31910 type and value are not present.
31912 A dynamic varobj will not report the access qualifying
31913 pseudo-children, regardless of the language. This information is not
31914 available at all with a dynamic varobj.
31917 Number of children this child has. For a dynamic varobj, this will be
31921 The type of the child. If @samp{print object}
31922 (@pxref{Print Settings, set print object}) is set to @code{on}, the
31923 @emph{actual} (derived) type of the object is shown rather than the
31924 @emph{declared} one.
31927 If values were requested, this is the value.
31930 If this variable object is associated with a thread, this is the
31931 thread's global thread id. Otherwise this result is not present.
31934 If the variable object is frozen, this variable will be present with a value of 1.
31937 A dynamic varobj can supply a display hint to the front end. The
31938 value comes directly from the Python pretty-printer object's
31939 @code{display_hint} method. @xref{Pretty Printing API}.
31942 This attribute will be present and have the value @samp{1} if the
31943 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
31944 then this attribute will not be present.
31948 The result may have its own attributes:
31952 A dynamic varobj can supply a display hint to the front end. The
31953 value comes directly from the Python pretty-printer object's
31954 @code{display_hint} method. @xref{Pretty Printing API}.
31957 This is an integer attribute which is nonzero if there are children
31958 remaining after the end of the selected range.
31961 @subsubheading Example
31965 -var-list-children n
31966 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
31967 numchild=@var{n},type=@var{type}@},@r{(repeats N times)}]
31969 -var-list-children --all-values n
31970 ^done,numchild=@var{n},children=[child=@{name=@var{name},exp=@var{exp},
31971 numchild=@var{n},value=@var{value},type=@var{type}@},@r{(repeats N times)}]
31975 @subheading The @code{-var-info-type} Command
31976 @findex -var-info-type
31978 @subsubheading Synopsis
31981 -var-info-type @var{name}
31984 Returns the type of the specified variable @var{name}. The type is
31985 returned as a string in the same format as it is output by the
31989 type=@var{typename}
31993 @subheading The @code{-var-info-expression} Command
31994 @findex -var-info-expression
31996 @subsubheading Synopsis
31999 -var-info-expression @var{name}
32002 Returns a string that is suitable for presenting this
32003 variable object in user interface. The string is generally
32004 not valid expression in the current language, and cannot be evaluated.
32006 For example, if @code{a} is an array, and variable object
32007 @code{A} was created for @code{a}, then we'll get this output:
32010 (gdb) -var-info-expression A.1
32011 ^done,lang="C",exp="1"
32015 Here, the value of @code{lang} is the language name, which can be
32016 found in @ref{Supported Languages}.
32018 Note that the output of the @code{-var-list-children} command also
32019 includes those expressions, so the @code{-var-info-expression} command
32022 @subheading The @code{-var-info-path-expression} Command
32023 @findex -var-info-path-expression
32025 @subsubheading Synopsis
32028 -var-info-path-expression @var{name}
32031 Returns an expression that can be evaluated in the current
32032 context and will yield the same value that a variable object has.
32033 Compare this with the @code{-var-info-expression} command, which
32034 result can be used only for UI presentation. Typical use of
32035 the @code{-var-info-path-expression} command is creating a
32036 watchpoint from a variable object.
32038 This command is currently not valid for children of a dynamic varobj,
32039 and will give an error when invoked on one.
32041 For example, suppose @code{C} is a C@t{++} class, derived from class
32042 @code{Base}, and that the @code{Base} class has a member called
32043 @code{m_size}. Assume a variable @code{c} is has the type of
32044 @code{C} and a variable object @code{C} was created for variable
32045 @code{c}. Then, we'll get this output:
32047 (gdb) -var-info-path-expression C.Base.public.m_size
32048 ^done,path_expr=((Base)c).m_size)
32051 @subheading The @code{-var-show-attributes} Command
32052 @findex -var-show-attributes
32054 @subsubheading Synopsis
32057 -var-show-attributes @var{name}
32060 List attributes of the specified variable object @var{name}:
32063 status=@var{attr} [ ( ,@var{attr} )* ]
32067 where @var{attr} is @code{@{ @{ editable | noneditable @} | TBD @}}.
32069 @subheading The @code{-var-evaluate-expression} Command
32070 @findex -var-evaluate-expression
32072 @subsubheading Synopsis
32075 -var-evaluate-expression [-f @var{format-spec}] @var{name}
32078 Evaluates the expression that is represented by the specified variable
32079 object and returns its value as a string. The format of the string
32080 can be specified with the @samp{-f} option. The possible values of
32081 this option are the same as for @code{-var-set-format}
32082 (@pxref{-var-set-format}). If the @samp{-f} option is not specified,
32083 the current display format will be used. The current display format
32084 can be changed using the @code{-var-set-format} command.
32090 Note that one must invoke @code{-var-list-children} for a variable
32091 before the value of a child variable can be evaluated.
32093 @subheading The @code{-var-assign} Command
32094 @findex -var-assign
32096 @subsubheading Synopsis
32099 -var-assign @var{name} @var{expression}
32102 Assigns the value of @var{expression} to the variable object specified
32103 by @var{name}. The object must be @samp{editable}. If the variable's
32104 value is altered by the assign, the variable will show up in any
32105 subsequent @code{-var-update} list.
32107 @subsubheading Example
32115 ^done,changelist=[@{name="var1",in_scope="true",type_changed="false"@}]
32119 @subheading The @code{-var-update} Command
32120 @findex -var-update
32122 @subsubheading Synopsis
32125 -var-update [@var{print-values}] @{@var{name} | "*"@}
32128 Reevaluate the expressions corresponding to the variable object
32129 @var{name} and all its direct and indirect children, and return the
32130 list of variable objects whose values have changed; @var{name} must
32131 be a root variable object. Here, ``changed'' means that the result of
32132 @code{-var-evaluate-expression} before and after the
32133 @code{-var-update} is different. If @samp{*} is used as the variable
32134 object names, all existing variable objects are updated, except
32135 for frozen ones (@pxref{-var-set-frozen}). The option
32136 @var{print-values} determines whether both names and values, or just
32137 names are printed. The possible values of this option are the same
32138 as for @code{-var-list-children} (@pxref{-var-list-children}). It is
32139 recommended to use the @samp{--all-values} option, to reduce the
32140 number of MI commands needed on each program stop.
32142 With the @samp{*} parameter, if a variable object is bound to a
32143 currently running thread, it will not be updated, without any
32146 If @code{-var-set-update-range} was previously used on a varobj, then
32147 only the selected range of children will be reported.
32149 @code{-var-update} reports all the changed varobjs in a tuple named
32152 Each item in the change list is itself a tuple holding:
32156 The name of the varobj.
32159 If values were requested for this update, then this field will be
32160 present and will hold the value of the varobj.
32163 @anchor{-var-update}
32164 This field is a string which may take one of three values:
32168 The variable object's current value is valid.
32171 The variable object does not currently hold a valid value but it may
32172 hold one in the future if its associated expression comes back into
32176 The variable object no longer holds a valid value.
32177 This can occur when the executable file being debugged has changed,
32178 either through recompilation or by using the @value{GDBN} @code{file}
32179 command. The front end should normally choose to delete these variable
32183 In the future new values may be added to this list so the front should
32184 be prepared for this possibility. @xref{GDB/MI Development and Front Ends, ,@sc{GDB/MI} Development and Front Ends}.
32187 This is only present if the varobj is still valid. If the type
32188 changed, then this will be the string @samp{true}; otherwise it will
32191 When a varobj's type changes, its children are also likely to have
32192 become incorrect. Therefore, the varobj's children are automatically
32193 deleted when this attribute is @samp{true}. Also, the varobj's update
32194 range, when set using the @code{-var-set-update-range} command, is
32198 If the varobj's type changed, then this field will be present and will
32201 @item new_num_children
32202 For a dynamic varobj, if the number of children changed, or if the
32203 type changed, this will be the new number of children.
32205 The @samp{numchild} field in other varobj responses is generally not
32206 valid for a dynamic varobj -- it will show the number of children that
32207 @value{GDBN} knows about, but because dynamic varobjs lazily
32208 instantiate their children, this will not reflect the number of
32209 children which may be available.
32211 The @samp{new_num_children} attribute only reports changes to the
32212 number of children known by @value{GDBN}. This is the only way to
32213 detect whether an update has removed children (which necessarily can
32214 only happen at the end of the update range).
32217 The display hint, if any.
32220 This is an integer value, which will be 1 if there are more children
32221 available outside the varobj's update range.
32224 This attribute will be present and have the value @samp{1} if the
32225 varobj is a dynamic varobj. If the varobj is not a dynamic varobj,
32226 then this attribute will not be present.
32229 If new children were added to a dynamic varobj within the selected
32230 update range (as set by @code{-var-set-update-range}), then they will
32231 be listed in this attribute.
32234 @subsubheading Example
32241 -var-update --all-values var1
32242 ^done,changelist=[@{name="var1",value="3",in_scope="true",
32243 type_changed="false"@}]
32247 @subheading The @code{-var-set-frozen} Command
32248 @findex -var-set-frozen
32249 @anchor{-var-set-frozen}
32251 @subsubheading Synopsis
32254 -var-set-frozen @var{name} @var{flag}
32257 Set the frozenness flag on the variable object @var{name}. The
32258 @var{flag} parameter should be either @samp{1} to make the variable
32259 frozen or @samp{0} to make it unfrozen. If a variable object is
32260 frozen, then neither itself, nor any of its children, are
32261 implicitly updated by @code{-var-update} of
32262 a parent variable or by @code{-var-update *}. Only
32263 @code{-var-update} of the variable itself will update its value and
32264 values of its children. After a variable object is unfrozen, it is
32265 implicitly updated by all subsequent @code{-var-update} operations.
32266 Unfreezing a variable does not update it, only subsequent
32267 @code{-var-update} does.
32269 @subsubheading Example
32273 -var-set-frozen V 1
32278 @subheading The @code{-var-set-update-range} command
32279 @findex -var-set-update-range
32280 @anchor{-var-set-update-range}
32282 @subsubheading Synopsis
32285 -var-set-update-range @var{name} @var{from} @var{to}
32288 Set the range of children to be returned by future invocations of
32289 @code{-var-update}.
32291 @var{from} and @var{to} indicate the range of children to report. If
32292 @var{from} or @var{to} is less than zero, the range is reset and all
32293 children will be reported. Otherwise, children starting at @var{from}
32294 (zero-based) and up to and excluding @var{to} will be reported.
32296 @subsubheading Example
32300 -var-set-update-range V 1 2
32304 @subheading The @code{-var-set-visualizer} command
32305 @findex -var-set-visualizer
32306 @anchor{-var-set-visualizer}
32308 @subsubheading Synopsis
32311 -var-set-visualizer @var{name} @var{visualizer}
32314 Set a visualizer for the variable object @var{name}.
32316 @var{visualizer} is the visualizer to use. The special value
32317 @samp{None} means to disable any visualizer in use.
32319 If not @samp{None}, @var{visualizer} must be a Python expression.
32320 This expression must evaluate to a callable object which accepts a
32321 single argument. @value{GDBN} will call this object with the value of
32322 the varobj @var{name} as an argument (this is done so that the same
32323 Python pretty-printing code can be used for both the CLI and MI).
32324 When called, this object must return an object which conforms to the
32325 pretty-printing interface (@pxref{Pretty Printing API}).
32327 The pre-defined function @code{gdb.default_visualizer} may be used to
32328 select a visualizer by following the built-in process
32329 (@pxref{Selecting Pretty-Printers}). This is done automatically when
32330 a varobj is created, and so ordinarily is not needed.
32332 This feature is only available if Python support is enabled. The MI
32333 command @code{-list-features} (@pxref{GDB/MI Support Commands})
32334 can be used to check this.
32336 @subsubheading Example
32338 Resetting the visualizer:
32342 -var-set-visualizer V None
32346 Reselecting the default (type-based) visualizer:
32350 -var-set-visualizer V gdb.default_visualizer
32354 Suppose @code{SomeClass} is a visualizer class. A lambda expression
32355 can be used to instantiate this class for a varobj:
32359 -var-set-visualizer V "lambda val: SomeClass()"
32363 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
32364 @node GDB/MI Data Manipulation
32365 @section @sc{gdb/mi} Data Manipulation
32367 @cindex data manipulation, in @sc{gdb/mi}
32368 @cindex @sc{gdb/mi}, data manipulation
32369 This section describes the @sc{gdb/mi} commands that manipulate data:
32370 examine memory and registers, evaluate expressions, etc.
32372 For details about what an addressable memory unit is,
32373 @pxref{addressable memory unit}.
32375 @c REMOVED FROM THE INTERFACE.
32376 @c @subheading -data-assign
32377 @c Change the value of a program variable. Plenty of side effects.
32378 @c @subsubheading GDB Command
32380 @c @subsubheading Example
32383 @subheading The @code{-data-disassemble} Command
32384 @findex -data-disassemble
32386 @subsubheading Synopsis
32390 [ -s @var{start-addr} -e @var{end-addr} ]
32391 | [ -a @var{addr} ]
32392 | [ -f @var{filename} -l @var{linenum} [ -n @var{lines} ] ]
32400 @item @var{start-addr}
32401 is the beginning address (or @code{$pc})
32402 @item @var{end-addr}
32405 is an address anywhere within (or the name of) the function to
32406 disassemble. If an address is specified, the whole function
32407 surrounding that address will be disassembled. If a name is
32408 specified, the whole function with that name will be disassembled.
32409 @item @var{filename}
32410 is the name of the file to disassemble
32411 @item @var{linenum}
32412 is the line number to disassemble around
32414 is the number of disassembly lines to be produced. If it is -1,
32415 the whole function will be disassembled, in case no @var{end-addr} is
32416 specified. If @var{end-addr} is specified as a non-zero value, and
32417 @var{lines} is lower than the number of disassembly lines between
32418 @var{start-addr} and @var{end-addr}, only @var{lines} lines are
32419 displayed; if @var{lines} is higher than the number of lines between
32420 @var{start-addr} and @var{end-addr}, only the lines up to @var{end-addr}
32425 @item 0 disassembly only
32426 @item 1 mixed source and disassembly (deprecated)
32427 @item 2 disassembly with raw opcodes
32428 @item 3 mixed source and disassembly with raw opcodes (deprecated)
32429 @item 4 mixed source and disassembly
32430 @item 5 mixed source and disassembly with raw opcodes
32433 Modes 1 and 3 are deprecated. The output is ``source centric''
32434 which hasn't proved useful in practice.
32435 @xref{Machine Code}, for a discussion of the difference between
32436 @code{/m} and @code{/s} output of the @code{disassemble} command.
32439 @subsubheading Result
32441 The result of the @code{-data-disassemble} command will be a list named
32442 @samp{asm_insns}, the contents of this list depend on the @var{mode}
32443 used with the @code{-data-disassemble} command.
32445 For modes 0 and 2 the @samp{asm_insns} list contains tuples with the
32450 The address at which this instruction was disassembled.
32453 The name of the function this instruction is within.
32456 The decimal offset in bytes from the start of @samp{func-name}.
32459 The text disassembly for this @samp{address}.
32462 This field is only present for modes 2, 3 and 5. This contains the raw opcode
32463 bytes for the @samp{inst} field.
32467 For modes 1, 3, 4 and 5 the @samp{asm_insns} list contains tuples named
32468 @samp{src_and_asm_line}, each of which has the following fields:
32472 The line number within @samp{file}.
32475 The file name from the compilation unit. This might be an absolute
32476 file name or a relative file name depending on the compile command
32480 Absolute file name of @samp{file}. It is converted to a canonical form
32481 using the source file search path
32482 (@pxref{Source Path, ,Specifying Source Directories})
32483 and after resolving all the symbolic links.
32485 If the source file is not found this field will contain the path as
32486 present in the debug information.
32488 @item line_asm_insn
32489 This is a list of tuples containing the disassembly for @samp{line} in
32490 @samp{file}. The fields of each tuple are the same as for
32491 @code{-data-disassemble} in @var{mode} 0 and 2, so @samp{address},
32492 @samp{func-name}, @samp{offset}, @samp{inst}, and optionally
32497 Note that whatever included in the @samp{inst} field, is not
32498 manipulated directly by @sc{gdb/mi}, i.e., it is not possible to
32501 @subsubheading @value{GDBN} Command
32503 The corresponding @value{GDBN} command is @samp{disassemble}.
32505 @subsubheading Example
32507 Disassemble from the current value of @code{$pc} to @code{$pc + 20}:
32511 -data-disassemble -s $pc -e "$pc + 20" -- 0
32514 @{address="0x000107c0",func-name="main",offset="4",
32515 inst="mov 2, %o0"@},
32516 @{address="0x000107c4",func-name="main",offset="8",
32517 inst="sethi %hi(0x11800), %o2"@},
32518 @{address="0x000107c8",func-name="main",offset="12",
32519 inst="or %o2, 0x140, %o1\t! 0x11940 <_lib_version+8>"@},
32520 @{address="0x000107cc",func-name="main",offset="16",
32521 inst="sethi %hi(0x11800), %o2"@},
32522 @{address="0x000107d0",func-name="main",offset="20",
32523 inst="or %o2, 0x168, %o4\t! 0x11968 <_lib_version+48>"@}]
32527 Disassemble the whole @code{main} function. Line 32 is part of
32531 -data-disassemble -f basics.c -l 32 -- 0
32533 @{address="0x000107bc",func-name="main",offset="0",
32534 inst="save %sp, -112, %sp"@},
32535 @{address="0x000107c0",func-name="main",offset="4",
32536 inst="mov 2, %o0"@},
32537 @{address="0x000107c4",func-name="main",offset="8",
32538 inst="sethi %hi(0x11800), %o2"@},
32540 @{address="0x0001081c",func-name="main",offset="96",inst="ret "@},
32541 @{address="0x00010820",func-name="main",offset="100",inst="restore "@}]
32545 Disassemble 3 instructions from the start of @code{main}:
32549 -data-disassemble -f basics.c -l 32 -n 3 -- 0
32551 @{address="0x000107bc",func-name="main",offset="0",
32552 inst="save %sp, -112, %sp"@},
32553 @{address="0x000107c0",func-name="main",offset="4",
32554 inst="mov 2, %o0"@},
32555 @{address="0x000107c4",func-name="main",offset="8",
32556 inst="sethi %hi(0x11800), %o2"@}]
32560 Disassemble 3 instructions from the start of @code{main} in mixed mode:
32564 -data-disassemble -f basics.c -l 32 -n 3 -- 1
32566 src_and_asm_line=@{line="31",
32567 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32568 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32569 line_asm_insn=[@{address="0x000107bc",
32570 func-name="main",offset="0",inst="save %sp, -112, %sp"@}]@},
32571 src_and_asm_line=@{line="32",
32572 file="../../../src/gdb/testsuite/gdb.mi/basics.c",
32573 fullname="/absolute/path/to/src/gdb/testsuite/gdb.mi/basics.c",
32574 line_asm_insn=[@{address="0x000107c0",
32575 func-name="main",offset="4",inst="mov 2, %o0"@},
32576 @{address="0x000107c4",func-name="main",offset="8",
32577 inst="sethi %hi(0x11800), %o2"@}]@}]
32582 @subheading The @code{-data-evaluate-expression} Command
32583 @findex -data-evaluate-expression
32585 @subsubheading Synopsis
32588 -data-evaluate-expression @var{expr}
32591 Evaluate @var{expr} as an expression. The expression could contain an
32592 inferior function call. The function call will execute synchronously.
32593 If the expression contains spaces, it must be enclosed in double quotes.
32595 @subsubheading @value{GDBN} Command
32597 The corresponding @value{GDBN} commands are @samp{print}, @samp{output}, and
32598 @samp{call}. In @code{gdbtk} only, there's a corresponding
32599 @samp{gdb_eval} command.
32601 @subsubheading Example
32603 In the following example, the numbers that precede the commands are the
32604 @dfn{tokens} described in @ref{GDB/MI Command Syntax, ,@sc{gdb/mi}
32605 Command Syntax}. Notice how @sc{gdb/mi} returns the same tokens in its
32609 211-data-evaluate-expression A
32612 311-data-evaluate-expression &A
32613 311^done,value="0xefffeb7c"
32615 411-data-evaluate-expression A+3
32618 511-data-evaluate-expression "A + 3"
32624 @subheading The @code{-data-list-changed-registers} Command
32625 @findex -data-list-changed-registers
32627 @subsubheading Synopsis
32630 -data-list-changed-registers
32633 Display a list of the registers that have changed.
32635 @subsubheading @value{GDBN} Command
32637 @value{GDBN} doesn't have a direct analog for this command; @code{gdbtk}
32638 has the corresponding command @samp{gdb_changed_register_list}.
32640 @subsubheading Example
32642 On a PPC MBX board:
32650 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",frame=@{
32651 func="main",args=[],file="try.c",fullname="/home/foo/bar/try.c",
32652 line="5",arch="powerpc"@}
32654 -data-list-changed-registers
32655 ^done,changed-registers=["0","1","2","4","5","6","7","8","9",
32656 "10","11","13","14","15","16","17","18","19","20","21","22","23",
32657 "24","25","26","27","28","30","31","64","65","66","67","69"]
32662 @subheading The @code{-data-list-register-names} Command
32663 @findex -data-list-register-names
32665 @subsubheading Synopsis
32668 -data-list-register-names [ ( @var{regno} )+ ]
32671 Show a list of register names for the current target. If no arguments
32672 are given, it shows a list of the names of all the registers. If
32673 integer numbers are given as arguments, it will print a list of the
32674 names of the registers corresponding to the arguments. To ensure
32675 consistency between a register name and its number, the output list may
32676 include empty register names.
32678 @subsubheading @value{GDBN} Command
32680 @value{GDBN} does not have a command which corresponds to
32681 @samp{-data-list-register-names}. In @code{gdbtk} there is a
32682 corresponding command @samp{gdb_regnames}.
32684 @subsubheading Example
32686 For the PPC MBX board:
32689 -data-list-register-names
32690 ^done,register-names=["r0","r1","r2","r3","r4","r5","r6","r7",
32691 "r8","r9","r10","r11","r12","r13","r14","r15","r16","r17","r18",
32692 "r19","r20","r21","r22","r23","r24","r25","r26","r27","r28","r29",
32693 "r30","r31","f0","f1","f2","f3","f4","f5","f6","f7","f8","f9",
32694 "f10","f11","f12","f13","f14","f15","f16","f17","f18","f19","f20",
32695 "f21","f22","f23","f24","f25","f26","f27","f28","f29","f30","f31",
32696 "", "pc","ps","cr","lr","ctr","xer"]
32698 -data-list-register-names 1 2 3
32699 ^done,register-names=["r1","r2","r3"]
32703 @subheading The @code{-data-list-register-values} Command
32704 @findex -data-list-register-values
32706 @subsubheading Synopsis
32709 -data-list-register-values
32710 [ @code{--skip-unavailable} ] @var{fmt} [ ( @var{regno} )*]
32713 Display the registers' contents. The format according to which the
32714 registers' contents are to be returned is given by @var{fmt}, followed
32715 by an optional list of numbers specifying the registers to display. A
32716 missing list of numbers indicates that the contents of all the
32717 registers must be returned. The @code{--skip-unavailable} option
32718 indicates that only the available registers are to be returned.
32720 Allowed formats for @var{fmt} are:
32737 @subsubheading @value{GDBN} Command
32739 The corresponding @value{GDBN} commands are @samp{info reg}, @samp{info
32740 all-reg}, and (in @code{gdbtk}) @samp{gdb_fetch_registers}.
32742 @subsubheading Example
32744 For a PPC MBX board (note: line breaks are for readability only, they
32745 don't appear in the actual output):
32749 -data-list-register-values r 64 65
32750 ^done,register-values=[@{number="64",value="0xfe00a300"@},
32751 @{number="65",value="0x00029002"@}]
32753 -data-list-register-values x
32754 ^done,register-values=[@{number="0",value="0xfe0043c8"@},
32755 @{number="1",value="0x3fff88"@},@{number="2",value="0xfffffffe"@},
32756 @{number="3",value="0x0"@},@{number="4",value="0xa"@},
32757 @{number="5",value="0x3fff68"@},@{number="6",value="0x3fff58"@},
32758 @{number="7",value="0xfe011e98"@},@{number="8",value="0x2"@},
32759 @{number="9",value="0xfa202820"@},@{number="10",value="0xfa202808"@},
32760 @{number="11",value="0x1"@},@{number="12",value="0x0"@},
32761 @{number="13",value="0x4544"@},@{number="14",value="0xffdfffff"@},
32762 @{number="15",value="0xffffffff"@},@{number="16",value="0xfffffeff"@},
32763 @{number="17",value="0xefffffed"@},@{number="18",value="0xfffffffe"@},
32764 @{number="19",value="0xffffffff"@},@{number="20",value="0xffffffff"@},
32765 @{number="21",value="0xffffffff"@},@{number="22",value="0xfffffff7"@},
32766 @{number="23",value="0xffffffff"@},@{number="24",value="0xffffffff"@},
32767 @{number="25",value="0xffffffff"@},@{number="26",value="0xfffffffb"@},
32768 @{number="27",value="0xffffffff"@},@{number="28",value="0xf7bfffff"@},
32769 @{number="29",value="0x0"@},@{number="30",value="0xfe010000"@},
32770 @{number="31",value="0x0"@},@{number="32",value="0x0"@},
32771 @{number="33",value="0x0"@},@{number="34",value="0x0"@},
32772 @{number="35",value="0x0"@},@{number="36",value="0x0"@},
32773 @{number="37",value="0x0"@},@{number="38",value="0x0"@},
32774 @{number="39",value="0x0"@},@{number="40",value="0x0"@},
32775 @{number="41",value="0x0"@},@{number="42",value="0x0"@},
32776 @{number="43",value="0x0"@},@{number="44",value="0x0"@},
32777 @{number="45",value="0x0"@},@{number="46",value="0x0"@},
32778 @{number="47",value="0x0"@},@{number="48",value="0x0"@},
32779 @{number="49",value="0x0"@},@{number="50",value="0x0"@},
32780 @{number="51",value="0x0"@},@{number="52",value="0x0"@},
32781 @{number="53",value="0x0"@},@{number="54",value="0x0"@},
32782 @{number="55",value="0x0"@},@{number="56",value="0x0"@},
32783 @{number="57",value="0x0"@},@{number="58",value="0x0"@},
32784 @{number="59",value="0x0"@},@{number="60",value="0x0"@},
32785 @{number="61",value="0x0"@},@{number="62",value="0x0"@},
32786 @{number="63",value="0x0"@},@{number="64",value="0xfe00a300"@},
32787 @{number="65",value="0x29002"@},@{number="66",value="0x202f04b5"@},
32788 @{number="67",value="0xfe0043b0"@},@{number="68",value="0xfe00b3e4"@},
32789 @{number="69",value="0x20002b03"@}]
32794 @subheading The @code{-data-read-memory} Command
32795 @findex -data-read-memory
32797 This command is deprecated, use @code{-data-read-memory-bytes} instead.
32799 @subsubheading Synopsis
32802 -data-read-memory [ -o @var{byte-offset} ]
32803 @var{address} @var{word-format} @var{word-size}
32804 @var{nr-rows} @var{nr-cols} [ @var{aschar} ]
32811 @item @var{address}
32812 An expression specifying the address of the first memory word to be
32813 read. Complex expressions containing embedded white space should be
32814 quoted using the C convention.
32816 @item @var{word-format}
32817 The format to be used to print the memory words. The notation is the
32818 same as for @value{GDBN}'s @code{print} command (@pxref{Output Formats,
32821 @item @var{word-size}
32822 The size of each memory word in bytes.
32824 @item @var{nr-rows}
32825 The number of rows in the output table.
32827 @item @var{nr-cols}
32828 The number of columns in the output table.
32831 If present, indicates that each row should include an @sc{ascii} dump. The
32832 value of @var{aschar} is used as a padding character when a byte is not a
32833 member of the printable @sc{ascii} character set (printable @sc{ascii}
32834 characters are those whose code is between 32 and 126, inclusively).
32836 @item @var{byte-offset}
32837 An offset to add to the @var{address} before fetching memory.
32840 This command displays memory contents as a table of @var{nr-rows} by
32841 @var{nr-cols} words, each word being @var{word-size} bytes. In total,
32842 @code{@var{nr-rows} * @var{nr-cols} * @var{word-size}} bytes are read
32843 (returned as @samp{total-bytes}). Should less than the requested number
32844 of bytes be returned by the target, the missing words are identified
32845 using @samp{N/A}. The number of bytes read from the target is returned
32846 in @samp{nr-bytes} and the starting address used to read memory in
32849 The address of the next/previous row or page is available in
32850 @samp{next-row} and @samp{prev-row}, @samp{next-page} and
32853 @subsubheading @value{GDBN} Command
32855 The corresponding @value{GDBN} command is @samp{x}. @code{gdbtk} has
32856 @samp{gdb_get_mem} memory read command.
32858 @subsubheading Example
32860 Read six bytes of memory starting at @code{bytes+6} but then offset by
32861 @code{-6} bytes. Format as three rows of two columns. One byte per
32862 word. Display each word in hex.
32866 9-data-read-memory -o -6 -- bytes+6 x 1 3 2
32867 9^done,addr="0x00001390",nr-bytes="6",total-bytes="6",
32868 next-row="0x00001396",prev-row="0x0000138e",next-page="0x00001396",
32869 prev-page="0x0000138a",memory=[
32870 @{addr="0x00001390",data=["0x00","0x01"]@},
32871 @{addr="0x00001392",data=["0x02","0x03"]@},
32872 @{addr="0x00001394",data=["0x04","0x05"]@}]
32876 Read two bytes of memory starting at address @code{shorts + 64} and
32877 display as a single word formatted in decimal.
32881 5-data-read-memory shorts+64 d 2 1 1
32882 5^done,addr="0x00001510",nr-bytes="2",total-bytes="2",
32883 next-row="0x00001512",prev-row="0x0000150e",
32884 next-page="0x00001512",prev-page="0x0000150e",memory=[
32885 @{addr="0x00001510",data=["128"]@}]
32889 Read thirty two bytes of memory starting at @code{bytes+16} and format
32890 as eight rows of four columns. Include a string encoding with @samp{x}
32891 used as the non-printable character.
32895 4-data-read-memory bytes+16 x 1 8 4 x
32896 4^done,addr="0x000013a0",nr-bytes="32",total-bytes="32",
32897 next-row="0x000013c0",prev-row="0x0000139c",
32898 next-page="0x000013c0",prev-page="0x00001380",memory=[
32899 @{addr="0x000013a0",data=["0x10","0x11","0x12","0x13"],ascii="xxxx"@},
32900 @{addr="0x000013a4",data=["0x14","0x15","0x16","0x17"],ascii="xxxx"@},
32901 @{addr="0x000013a8",data=["0x18","0x19","0x1a","0x1b"],ascii="xxxx"@},
32902 @{addr="0x000013ac",data=["0x1c","0x1d","0x1e","0x1f"],ascii="xxxx"@},
32903 @{addr="0x000013b0",data=["0x20","0x21","0x22","0x23"],ascii=" !\"#"@},
32904 @{addr="0x000013b4",data=["0x24","0x25","0x26","0x27"],ascii="$%&'"@},
32905 @{addr="0x000013b8",data=["0x28","0x29","0x2a","0x2b"],ascii="()*+"@},
32906 @{addr="0x000013bc",data=["0x2c","0x2d","0x2e","0x2f"],ascii=",-./"@}]
32910 @subheading The @code{-data-read-memory-bytes} Command
32911 @findex -data-read-memory-bytes
32913 @subsubheading Synopsis
32916 -data-read-memory-bytes [ -o @var{offset} ]
32917 @var{address} @var{count}
32924 @item @var{address}
32925 An expression specifying the address of the first addressable memory unit
32926 to be read. Complex expressions containing embedded white space should be
32927 quoted using the C convention.
32930 The number of addressable memory units to read. This should be an integer
32934 The offset relative to @var{address} at which to start reading. This
32935 should be an integer literal. This option is provided so that a frontend
32936 is not required to first evaluate address and then perform address
32937 arithmetics itself.
32941 This command attempts to read all accessible memory regions in the
32942 specified range. First, all regions marked as unreadable in the memory
32943 map (if one is defined) will be skipped. @xref{Memory Region
32944 Attributes}. Second, @value{GDBN} will attempt to read the remaining
32945 regions. For each one, if reading full region results in an errors,
32946 @value{GDBN} will try to read a subset of the region.
32948 In general, every single memory unit in the region may be readable or not,
32949 and the only way to read every readable unit is to try a read at
32950 every address, which is not practical. Therefore, @value{GDBN} will
32951 attempt to read all accessible memory units at either beginning or the end
32952 of the region, using a binary division scheme. This heuristic works
32953 well for reading accross a memory map boundary. Note that if a region
32954 has a readable range that is neither at the beginning or the end,
32955 @value{GDBN} will not read it.
32957 The result record (@pxref{GDB/MI Result Records}) that is output of
32958 the command includes a field named @samp{memory} whose content is a
32959 list of tuples. Each tuple represent a successfully read memory block
32960 and has the following fields:
32964 The start address of the memory block, as hexadecimal literal.
32967 The end address of the memory block, as hexadecimal literal.
32970 The offset of the memory block, as hexadecimal literal, relative to
32971 the start address passed to @code{-data-read-memory-bytes}.
32974 The contents of the memory block, in hex.
32980 @subsubheading @value{GDBN} Command
32982 The corresponding @value{GDBN} command is @samp{x}.
32984 @subsubheading Example
32988 -data-read-memory-bytes &a 10
32989 ^done,memory=[@{begin="0xbffff154",offset="0x00000000",
32991 contents="01000000020000000300"@}]
32996 @subheading The @code{-data-write-memory-bytes} Command
32997 @findex -data-write-memory-bytes
32999 @subsubheading Synopsis
33002 -data-write-memory-bytes @var{address} @var{contents}
33003 -data-write-memory-bytes @var{address} @var{contents} @r{[}@var{count}@r{]}
33010 @item @var{address}
33011 An expression specifying the address of the first addressable memory unit
33012 to be written. Complex expressions containing embedded white space should
33013 be quoted using the C convention.
33015 @item @var{contents}
33016 The hex-encoded data to write. It is an error if @var{contents} does
33017 not represent an integral number of addressable memory units.
33020 Optional argument indicating the number of addressable memory units to be
33021 written. If @var{count} is greater than @var{contents}' length,
33022 @value{GDBN} will repeatedly write @var{contents} until it fills
33023 @var{count} memory units.
33027 @subsubheading @value{GDBN} Command
33029 There's no corresponding @value{GDBN} command.
33031 @subsubheading Example
33035 -data-write-memory-bytes &a "aabbccdd"
33042 -data-write-memory-bytes &a "aabbccdd" 16e
33047 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33048 @node GDB/MI Tracepoint Commands
33049 @section @sc{gdb/mi} Tracepoint Commands
33051 The commands defined in this section implement MI support for
33052 tracepoints. For detailed introduction, see @ref{Tracepoints}.
33054 @subheading The @code{-trace-find} Command
33055 @findex -trace-find
33057 @subsubheading Synopsis
33060 -trace-find @var{mode} [@var{parameters}@dots{}]
33063 Find a trace frame using criteria defined by @var{mode} and
33064 @var{parameters}. The following table lists permissible
33065 modes and their parameters. For details of operation, see @ref{tfind}.
33070 No parameters are required. Stops examining trace frames.
33073 An integer is required as parameter. Selects tracepoint frame with
33076 @item tracepoint-number
33077 An integer is required as parameter. Finds next
33078 trace frame that corresponds to tracepoint with the specified number.
33081 An address is required as parameter. Finds
33082 next trace frame that corresponds to any tracepoint at the specified
33085 @item pc-inside-range
33086 Two addresses are required as parameters. Finds next trace
33087 frame that corresponds to a tracepoint at an address inside the
33088 specified range. Both bounds are considered to be inside the range.
33090 @item pc-outside-range
33091 Two addresses are required as parameters. Finds
33092 next trace frame that corresponds to a tracepoint at an address outside
33093 the specified range. Both bounds are considered to be inside the range.
33096 Line specification is required as parameter. @xref{Specify Location}.
33097 Finds next trace frame that corresponds to a tracepoint at
33098 the specified location.
33102 If @samp{none} was passed as @var{mode}, the response does not
33103 have fields. Otherwise, the response may have the following fields:
33107 This field has either @samp{0} or @samp{1} as the value, depending
33108 on whether a matching tracepoint was found.
33111 The index of the found traceframe. This field is present iff
33112 the @samp{found} field has value of @samp{1}.
33115 The index of the found tracepoint. This field is present iff
33116 the @samp{found} field has value of @samp{1}.
33119 The information about the frame corresponding to the found trace
33120 frame. This field is present only if a trace frame was found.
33121 @xref{GDB/MI Frame Information}, for description of this field.
33125 @subsubheading @value{GDBN} Command
33127 The corresponding @value{GDBN} command is @samp{tfind}.
33129 @subheading -trace-define-variable
33130 @findex -trace-define-variable
33132 @subsubheading Synopsis
33135 -trace-define-variable @var{name} [ @var{value} ]
33138 Create trace variable @var{name} if it does not exist. If
33139 @var{value} is specified, sets the initial value of the specified
33140 trace variable to that value. Note that the @var{name} should start
33141 with the @samp{$} character.
33143 @subsubheading @value{GDBN} Command
33145 The corresponding @value{GDBN} command is @samp{tvariable}.
33147 @subheading The @code{-trace-frame-collected} Command
33148 @findex -trace-frame-collected
33150 @subsubheading Synopsis
33153 -trace-frame-collected
33154 [--var-print-values @var{var_pval}]
33155 [--comp-print-values @var{comp_pval}]
33156 [--registers-format @var{regformat}]
33157 [--memory-contents]
33160 This command returns the set of collected objects, register names,
33161 trace state variable names, memory ranges and computed expressions
33162 that have been collected at a particular trace frame. The optional
33163 parameters to the command affect the output format in different ways.
33164 See the output description table below for more details.
33166 The reported names can be used in the normal manner to create
33167 varobjs and inspect the objects themselves. The items returned by
33168 this command are categorized so that it is clear which is a variable,
33169 which is a register, which is a trace state variable, which is a
33170 memory range and which is a computed expression.
33172 For instance, if the actions were
33174 collect myVar, myArray[myIndex], myObj.field, myPtr->field, myCount + 2
33175 collect *(int*)0xaf02bef0@@40
33179 the object collected in its entirety would be @code{myVar}. The
33180 object @code{myArray} would be partially collected, because only the
33181 element at index @code{myIndex} would be collected. The remaining
33182 objects would be computed expressions.
33184 An example output would be:
33188 -trace-frame-collected
33190 explicit-variables=[@{name="myVar",value="1"@}],
33191 computed-expressions=[@{name="myArray[myIndex]",value="0"@},
33192 @{name="myObj.field",value="0"@},
33193 @{name="myPtr->field",value="1"@},
33194 @{name="myCount + 2",value="3"@},
33195 @{name="$tvar1 + 1",value="43970027"@}],
33196 registers=[@{number="0",value="0x7fe2c6e79ec8"@},
33197 @{number="1",value="0x0"@},
33198 @{number="2",value="0x4"@},
33200 @{number="125",value="0x0"@}],
33201 tvars=[@{name="$tvar1",current="43970026"@}],
33202 memory=[@{address="0x0000000000602264",length="4"@},
33203 @{address="0x0000000000615bc0",length="4"@}]
33210 @item explicit-variables
33211 The set of objects that have been collected in their entirety (as
33212 opposed to collecting just a few elements of an array or a few struct
33213 members). For each object, its name and value are printed.
33214 The @code{--var-print-values} option affects how or whether the value
33215 field is output. If @var{var_pval} is 0, then print only the names;
33216 if it is 1, print also their values; and if it is 2, print the name,
33217 type and value for simple data types, and the name and type for
33218 arrays, structures and unions.
33220 @item computed-expressions
33221 The set of computed expressions that have been collected at the
33222 current trace frame. The @code{--comp-print-values} option affects
33223 this set like the @code{--var-print-values} option affects the
33224 @code{explicit-variables} set. See above.
33227 The registers that have been collected at the current trace frame.
33228 For each register collected, the name and current value are returned.
33229 The value is formatted according to the @code{--registers-format}
33230 option. See the @command{-data-list-register-values} command for a
33231 list of the allowed formats. The default is @samp{x}.
33234 The trace state variables that have been collected at the current
33235 trace frame. For each trace state variable collected, the name and
33236 current value are returned.
33239 The set of memory ranges that have been collected at the current trace
33240 frame. Its content is a list of tuples. Each tuple represents a
33241 collected memory range and has the following fields:
33245 The start address of the memory range, as hexadecimal literal.
33248 The length of the memory range, as decimal literal.
33251 The contents of the memory block, in hex. This field is only present
33252 if the @code{--memory-contents} option is specified.
33258 @subsubheading @value{GDBN} Command
33260 There is no corresponding @value{GDBN} command.
33262 @subsubheading Example
33264 @subheading -trace-list-variables
33265 @findex -trace-list-variables
33267 @subsubheading Synopsis
33270 -trace-list-variables
33273 Return a table of all defined trace variables. Each element of the
33274 table has the following fields:
33278 The name of the trace variable. This field is always present.
33281 The initial value. This is a 64-bit signed integer. This
33282 field is always present.
33285 The value the trace variable has at the moment. This is a 64-bit
33286 signed integer. This field is absent iff current value is
33287 not defined, for example if the trace was never run, or is
33292 @subsubheading @value{GDBN} Command
33294 The corresponding @value{GDBN} command is @samp{tvariables}.
33296 @subsubheading Example
33300 -trace-list-variables
33301 ^done,trace-variables=@{nr_rows="1",nr_cols="3",
33302 hdr=[@{width="15",alignment="-1",col_name="name",colhdr="Name"@},
33303 @{width="11",alignment="-1",col_name="initial",colhdr="Initial"@},
33304 @{width="11",alignment="-1",col_name="current",colhdr="Current"@}],
33305 body=[variable=@{name="$trace_timestamp",initial="0"@}
33306 variable=@{name="$foo",initial="10",current="15"@}]@}
33310 @subheading -trace-save
33311 @findex -trace-save
33313 @subsubheading Synopsis
33316 -trace-save [ -r ] [ -ctf ] @var{filename}
33319 Saves the collected trace data to @var{filename}. Without the
33320 @samp{-r} option, the data is downloaded from the target and saved
33321 in a local file. With the @samp{-r} option the target is asked
33322 to perform the save.
33324 By default, this command will save the trace in the tfile format. You can
33325 supply the optional @samp{-ctf} argument to save it the CTF format. See
33326 @ref{Trace Files} for more information about CTF.
33328 @subsubheading @value{GDBN} Command
33330 The corresponding @value{GDBN} command is @samp{tsave}.
33333 @subheading -trace-start
33334 @findex -trace-start
33336 @subsubheading Synopsis
33342 Starts a tracing experiment. The result of this command does not
33345 @subsubheading @value{GDBN} Command
33347 The corresponding @value{GDBN} command is @samp{tstart}.
33349 @subheading -trace-status
33350 @findex -trace-status
33352 @subsubheading Synopsis
33358 Obtains the status of a tracing experiment. The result may include
33359 the following fields:
33364 May have a value of either @samp{0}, when no tracing operations are
33365 supported, @samp{1}, when all tracing operations are supported, or
33366 @samp{file} when examining trace file. In the latter case, examining
33367 of trace frame is possible but new tracing experiement cannot be
33368 started. This field is always present.
33371 May have a value of either @samp{0} or @samp{1} depending on whether
33372 tracing experiement is in progress on target. This field is present
33373 if @samp{supported} field is not @samp{0}.
33376 Report the reason why the tracing was stopped last time. This field
33377 may be absent iff tracing was never stopped on target yet. The
33378 value of @samp{request} means the tracing was stopped as result of
33379 the @code{-trace-stop} command. The value of @samp{overflow} means
33380 the tracing buffer is full. The value of @samp{disconnection} means
33381 tracing was automatically stopped when @value{GDBN} has disconnected.
33382 The value of @samp{passcount} means tracing was stopped when a
33383 tracepoint was passed a maximal number of times for that tracepoint.
33384 This field is present if @samp{supported} field is not @samp{0}.
33386 @item stopping-tracepoint
33387 The number of tracepoint whose passcount as exceeded. This field is
33388 present iff the @samp{stop-reason} field has the value of
33392 @itemx frames-created
33393 The @samp{frames} field is a count of the total number of trace frames
33394 in the trace buffer, while @samp{frames-created} is the total created
33395 during the run, including ones that were discarded, such as when a
33396 circular trace buffer filled up. Both fields are optional.
33400 These fields tell the current size of the tracing buffer and the
33401 remaining space. These fields are optional.
33404 The value of the circular trace buffer flag. @code{1} means that the
33405 trace buffer is circular and old trace frames will be discarded if
33406 necessary to make room, @code{0} means that the trace buffer is linear
33410 The value of the disconnected tracing flag. @code{1} means that
33411 tracing will continue after @value{GDBN} disconnects, @code{0} means
33412 that the trace run will stop.
33415 The filename of the trace file being examined. This field is
33416 optional, and only present when examining a trace file.
33420 @subsubheading @value{GDBN} Command
33422 The corresponding @value{GDBN} command is @samp{tstatus}.
33424 @subheading -trace-stop
33425 @findex -trace-stop
33427 @subsubheading Synopsis
33433 Stops a tracing experiment. The result of this command has the same
33434 fields as @code{-trace-status}, except that the @samp{supported} and
33435 @samp{running} fields are not output.
33437 @subsubheading @value{GDBN} Command
33439 The corresponding @value{GDBN} command is @samp{tstop}.
33442 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33443 @node GDB/MI Symbol Query
33444 @section @sc{gdb/mi} Symbol Query Commands
33448 @subheading The @code{-symbol-info-address} Command
33449 @findex -symbol-info-address
33451 @subsubheading Synopsis
33454 -symbol-info-address @var{symbol}
33457 Describe where @var{symbol} is stored.
33459 @subsubheading @value{GDBN} Command
33461 The corresponding @value{GDBN} command is @samp{info address}.
33463 @subsubheading Example
33467 @subheading The @code{-symbol-info-file} Command
33468 @findex -symbol-info-file
33470 @subsubheading Synopsis
33476 Show the file for the symbol.
33478 @subsubheading @value{GDBN} Command
33480 There's no equivalent @value{GDBN} command. @code{gdbtk} has
33481 @samp{gdb_find_file}.
33483 @subsubheading Example
33487 @subheading The @code{-symbol-info-function} Command
33488 @findex -symbol-info-function
33490 @subsubheading Synopsis
33493 -symbol-info-function
33496 Show which function the symbol lives in.
33498 @subsubheading @value{GDBN} Command
33500 @samp{gdb_get_function} in @code{gdbtk}.
33502 @subsubheading Example
33506 @subheading The @code{-symbol-info-line} Command
33507 @findex -symbol-info-line
33509 @subsubheading Synopsis
33515 Show the core addresses of the code for a source line.
33517 @subsubheading @value{GDBN} Command
33519 The corresponding @value{GDBN} command is @samp{info line}.
33520 @code{gdbtk} has the @samp{gdb_get_line} and @samp{gdb_get_file} commands.
33522 @subsubheading Example
33526 @subheading The @code{-symbol-info-symbol} Command
33527 @findex -symbol-info-symbol
33529 @subsubheading Synopsis
33532 -symbol-info-symbol @var{addr}
33535 Describe what symbol is at location @var{addr}.
33537 @subsubheading @value{GDBN} Command
33539 The corresponding @value{GDBN} command is @samp{info symbol}.
33541 @subsubheading Example
33545 @subheading The @code{-symbol-list-functions} Command
33546 @findex -symbol-list-functions
33548 @subsubheading Synopsis
33551 -symbol-list-functions
33554 List the functions in the executable.
33556 @subsubheading @value{GDBN} Command
33558 @samp{info functions} in @value{GDBN}, @samp{gdb_listfunc} and
33559 @samp{gdb_search} in @code{gdbtk}.
33561 @subsubheading Example
33566 @subheading The @code{-symbol-list-lines} Command
33567 @findex -symbol-list-lines
33569 @subsubheading Synopsis
33572 -symbol-list-lines @var{filename}
33575 Print the list of lines that contain code and their associated program
33576 addresses for the given source filename. The entries are sorted in
33577 ascending PC order.
33579 @subsubheading @value{GDBN} Command
33581 There is no corresponding @value{GDBN} command.
33583 @subsubheading Example
33586 -symbol-list-lines basics.c
33587 ^done,lines=[@{pc="0x08048554",line="7"@},@{pc="0x0804855a",line="8"@}]
33593 @subheading The @code{-symbol-list-types} Command
33594 @findex -symbol-list-types
33596 @subsubheading Synopsis
33602 List all the type names.
33604 @subsubheading @value{GDBN} Command
33606 The corresponding commands are @samp{info types} in @value{GDBN},
33607 @samp{gdb_search} in @code{gdbtk}.
33609 @subsubheading Example
33613 @subheading The @code{-symbol-list-variables} Command
33614 @findex -symbol-list-variables
33616 @subsubheading Synopsis
33619 -symbol-list-variables
33622 List all the global and static variable names.
33624 @subsubheading @value{GDBN} Command
33626 @samp{info variables} in @value{GDBN}, @samp{gdb_search} in @code{gdbtk}.
33628 @subsubheading Example
33632 @subheading The @code{-symbol-locate} Command
33633 @findex -symbol-locate
33635 @subsubheading Synopsis
33641 @subsubheading @value{GDBN} Command
33643 @samp{gdb_loc} in @code{gdbtk}.
33645 @subsubheading Example
33649 @subheading The @code{-symbol-type} Command
33650 @findex -symbol-type
33652 @subsubheading Synopsis
33655 -symbol-type @var{variable}
33658 Show type of @var{variable}.
33660 @subsubheading @value{GDBN} Command
33662 The corresponding @value{GDBN} command is @samp{ptype}, @code{gdbtk} has
33663 @samp{gdb_obj_variable}.
33665 @subsubheading Example
33670 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33671 @node GDB/MI File Commands
33672 @section @sc{gdb/mi} File Commands
33674 This section describes the GDB/MI commands to specify executable file names
33675 and to read in and obtain symbol table information.
33677 @subheading The @code{-file-exec-and-symbols} Command
33678 @findex -file-exec-and-symbols
33680 @subsubheading Synopsis
33683 -file-exec-and-symbols @var{file}
33686 Specify the executable file to be debugged. This file is the one from
33687 which the symbol table is also read. If no file is specified, the
33688 command clears the executable and symbol information. If breakpoints
33689 are set when using this command with no arguments, @value{GDBN} will produce
33690 error messages. Otherwise, no output is produced, except a completion
33693 @subsubheading @value{GDBN} Command
33695 The corresponding @value{GDBN} command is @samp{file}.
33697 @subsubheading Example
33701 -file-exec-and-symbols /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33707 @subheading The @code{-file-exec-file} Command
33708 @findex -file-exec-file
33710 @subsubheading Synopsis
33713 -file-exec-file @var{file}
33716 Specify the executable file to be debugged. Unlike
33717 @samp{-file-exec-and-symbols}, the symbol table is @emph{not} read
33718 from this file. If used without argument, @value{GDBN} clears the information
33719 about the executable file. No output is produced, except a completion
33722 @subsubheading @value{GDBN} Command
33724 The corresponding @value{GDBN} command is @samp{exec-file}.
33726 @subsubheading Example
33730 -file-exec-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33737 @subheading The @code{-file-list-exec-sections} Command
33738 @findex -file-list-exec-sections
33740 @subsubheading Synopsis
33743 -file-list-exec-sections
33746 List the sections of the current executable file.
33748 @subsubheading @value{GDBN} Command
33750 The @value{GDBN} command @samp{info file} shows, among the rest, the same
33751 information as this command. @code{gdbtk} has a corresponding command
33752 @samp{gdb_load_info}.
33754 @subsubheading Example
33759 @subheading The @code{-file-list-exec-source-file} Command
33760 @findex -file-list-exec-source-file
33762 @subsubheading Synopsis
33765 -file-list-exec-source-file
33768 List the line number, the current source file, and the absolute path
33769 to the current source file for the current executable. The macro
33770 information field has a value of @samp{1} or @samp{0} depending on
33771 whether or not the file includes preprocessor macro information.
33773 @subsubheading @value{GDBN} Command
33775 The @value{GDBN} equivalent is @samp{info source}
33777 @subsubheading Example
33781 123-file-list-exec-source-file
33782 123^done,line="1",file="foo.c",fullname="/home/bar/foo.c,macro-info="1"
33787 @subheading The @code{-file-list-exec-source-files} Command
33788 @findex -file-list-exec-source-files
33790 @subsubheading Synopsis
33793 -file-list-exec-source-files
33796 List the source files for the current executable.
33798 It will always output both the filename and fullname (absolute file
33799 name) of a source file.
33801 @subsubheading @value{GDBN} Command
33803 The @value{GDBN} equivalent is @samp{info sources}.
33804 @code{gdbtk} has an analogous command @samp{gdb_listfiles}.
33806 @subsubheading Example
33809 -file-list-exec-source-files
33811 @{file=foo.c,fullname=/home/foo.c@},
33812 @{file=/home/bar.c,fullname=/home/bar.c@},
33813 @{file=gdb_could_not_find_fullpath.c@}]
33817 @subheading The @code{-file-list-shared-libraries} Command
33818 @findex -file-list-shared-libraries
33820 @subsubheading Synopsis
33823 -file-list-shared-libraries [ @var{regexp} ]
33826 List the shared libraries in the program.
33827 With a regular expression @var{regexp}, only those libraries whose
33828 names match @var{regexp} are listed.
33830 @subsubheading @value{GDBN} Command
33832 The corresponding @value{GDBN} command is @samp{info shared}. The fields
33833 have a similar meaning to the @code{=library-loaded} notification.
33834 The @code{ranges} field specifies the multiple segments belonging to this
33835 library. Each range has the following fields:
33839 The address defining the inclusive lower bound of the segment.
33841 The address defining the exclusive upper bound of the segment.
33844 @subsubheading Example
33847 -file-list-exec-source-files
33848 ^done,shared-libraries=[
33849 @{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"@}]@},
33850 @{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"@}]@}]
33856 @subheading The @code{-file-list-symbol-files} Command
33857 @findex -file-list-symbol-files
33859 @subsubheading Synopsis
33862 -file-list-symbol-files
33867 @subsubheading @value{GDBN} Command
33869 The corresponding @value{GDBN} command is @samp{info file} (part of it).
33871 @subsubheading Example
33876 @subheading The @code{-file-symbol-file} Command
33877 @findex -file-symbol-file
33879 @subsubheading Synopsis
33882 -file-symbol-file @var{file}
33885 Read symbol table info from the specified @var{file} argument. When
33886 used without arguments, clears @value{GDBN}'s symbol table info. No output is
33887 produced, except for a completion notification.
33889 @subsubheading @value{GDBN} Command
33891 The corresponding @value{GDBN} command is @samp{symbol-file}.
33893 @subsubheading Example
33897 -file-symbol-file /kwikemart/marge/ezannoni/TRUNK/mbx/hello.mbx
33903 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33904 @node GDB/MI Memory Overlay Commands
33905 @section @sc{gdb/mi} Memory Overlay Commands
33907 The memory overlay commands are not implemented.
33909 @c @subheading -overlay-auto
33911 @c @subheading -overlay-list-mapping-state
33913 @c @subheading -overlay-list-overlays
33915 @c @subheading -overlay-map
33917 @c @subheading -overlay-off
33919 @c @subheading -overlay-on
33921 @c @subheading -overlay-unmap
33923 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33924 @node GDB/MI Signal Handling Commands
33925 @section @sc{gdb/mi} Signal Handling Commands
33927 Signal handling commands are not implemented.
33929 @c @subheading -signal-handle
33931 @c @subheading -signal-list-handle-actions
33933 @c @subheading -signal-list-signal-types
33937 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
33938 @node GDB/MI Target Manipulation
33939 @section @sc{gdb/mi} Target Manipulation Commands
33942 @subheading The @code{-target-attach} Command
33943 @findex -target-attach
33945 @subsubheading Synopsis
33948 -target-attach @var{pid} | @var{gid} | @var{file}
33951 Attach to a process @var{pid} or a file @var{file} outside of
33952 @value{GDBN}, or a thread group @var{gid}. If attaching to a thread
33953 group, the id previously returned by
33954 @samp{-list-thread-groups --available} must be used.
33956 @subsubheading @value{GDBN} Command
33958 The corresponding @value{GDBN} command is @samp{attach}.
33960 @subsubheading Example
33964 =thread-created,id="1"
33965 *stopped,thread-id="1",frame=@{addr="0xb7f7e410",func="bar",args=[]@}
33971 @subheading The @code{-target-compare-sections} Command
33972 @findex -target-compare-sections
33974 @subsubheading Synopsis
33977 -target-compare-sections [ @var{section} ]
33980 Compare data of section @var{section} on target to the exec file.
33981 Without the argument, all sections are compared.
33983 @subsubheading @value{GDBN} Command
33985 The @value{GDBN} equivalent is @samp{compare-sections}.
33987 @subsubheading Example
33992 @subheading The @code{-target-detach} Command
33993 @findex -target-detach
33995 @subsubheading Synopsis
33998 -target-detach [ @var{pid} | @var{gid} ]
34001 Detach from the remote target which normally resumes its execution.
34002 If either @var{pid} or @var{gid} is specified, detaches from either
34003 the specified process, or specified thread group. There's no output.
34005 @subsubheading @value{GDBN} Command
34007 The corresponding @value{GDBN} command is @samp{detach}.
34009 @subsubheading Example
34019 @subheading The @code{-target-disconnect} Command
34020 @findex -target-disconnect
34022 @subsubheading Synopsis
34028 Disconnect from the remote target. There's no output and the target is
34029 generally not resumed.
34031 @subsubheading @value{GDBN} Command
34033 The corresponding @value{GDBN} command is @samp{disconnect}.
34035 @subsubheading Example
34045 @subheading The @code{-target-download} Command
34046 @findex -target-download
34048 @subsubheading Synopsis
34054 Loads the executable onto the remote target.
34055 It prints out an update message every half second, which includes the fields:
34059 The name of the section.
34061 The size of what has been sent so far for that section.
34063 The size of the section.
34065 The total size of what was sent so far (the current and the previous sections).
34067 The size of the overall executable to download.
34071 Each message is sent as status record (@pxref{GDB/MI Output Syntax, ,
34072 @sc{gdb/mi} Output Syntax}).
34074 In addition, it prints the name and size of the sections, as they are
34075 downloaded. These messages include the following fields:
34079 The name of the section.
34081 The size of the section.
34083 The size of the overall executable to download.
34087 At the end, a summary is printed.
34089 @subsubheading @value{GDBN} Command
34091 The corresponding @value{GDBN} command is @samp{load}.
34093 @subsubheading Example
34095 Note: each status message appears on a single line. Here the messages
34096 have been broken down so that they can fit onto a page.
34101 +download,@{section=".text",section-size="6668",total-size="9880"@}
34102 +download,@{section=".text",section-sent="512",section-size="6668",
34103 total-sent="512",total-size="9880"@}
34104 +download,@{section=".text",section-sent="1024",section-size="6668",
34105 total-sent="1024",total-size="9880"@}
34106 +download,@{section=".text",section-sent="1536",section-size="6668",
34107 total-sent="1536",total-size="9880"@}
34108 +download,@{section=".text",section-sent="2048",section-size="6668",
34109 total-sent="2048",total-size="9880"@}
34110 +download,@{section=".text",section-sent="2560",section-size="6668",
34111 total-sent="2560",total-size="9880"@}
34112 +download,@{section=".text",section-sent="3072",section-size="6668",
34113 total-sent="3072",total-size="9880"@}
34114 +download,@{section=".text",section-sent="3584",section-size="6668",
34115 total-sent="3584",total-size="9880"@}
34116 +download,@{section=".text",section-sent="4096",section-size="6668",
34117 total-sent="4096",total-size="9880"@}
34118 +download,@{section=".text",section-sent="4608",section-size="6668",
34119 total-sent="4608",total-size="9880"@}
34120 +download,@{section=".text",section-sent="5120",section-size="6668",
34121 total-sent="5120",total-size="9880"@}
34122 +download,@{section=".text",section-sent="5632",section-size="6668",
34123 total-sent="5632",total-size="9880"@}
34124 +download,@{section=".text",section-sent="6144",section-size="6668",
34125 total-sent="6144",total-size="9880"@}
34126 +download,@{section=".text",section-sent="6656",section-size="6668",
34127 total-sent="6656",total-size="9880"@}
34128 +download,@{section=".init",section-size="28",total-size="9880"@}
34129 +download,@{section=".fini",section-size="28",total-size="9880"@}
34130 +download,@{section=".data",section-size="3156",total-size="9880"@}
34131 +download,@{section=".data",section-sent="512",section-size="3156",
34132 total-sent="7236",total-size="9880"@}
34133 +download,@{section=".data",section-sent="1024",section-size="3156",
34134 total-sent="7748",total-size="9880"@}
34135 +download,@{section=".data",section-sent="1536",section-size="3156",
34136 total-sent="8260",total-size="9880"@}
34137 +download,@{section=".data",section-sent="2048",section-size="3156",
34138 total-sent="8772",total-size="9880"@}
34139 +download,@{section=".data",section-sent="2560",section-size="3156",
34140 total-sent="9284",total-size="9880"@}
34141 +download,@{section=".data",section-sent="3072",section-size="3156",
34142 total-sent="9796",total-size="9880"@}
34143 ^done,address="0x10004",load-size="9880",transfer-rate="6586",
34150 @subheading The @code{-target-exec-status} Command
34151 @findex -target-exec-status
34153 @subsubheading Synopsis
34156 -target-exec-status
34159 Provide information on the state of the target (whether it is running or
34160 not, for instance).
34162 @subsubheading @value{GDBN} Command
34164 There's no equivalent @value{GDBN} command.
34166 @subsubheading Example
34170 @subheading The @code{-target-list-available-targets} Command
34171 @findex -target-list-available-targets
34173 @subsubheading Synopsis
34176 -target-list-available-targets
34179 List the possible targets to connect to.
34181 @subsubheading @value{GDBN} Command
34183 The corresponding @value{GDBN} command is @samp{help target}.
34185 @subsubheading Example
34189 @subheading The @code{-target-list-current-targets} Command
34190 @findex -target-list-current-targets
34192 @subsubheading Synopsis
34195 -target-list-current-targets
34198 Describe the current target.
34200 @subsubheading @value{GDBN} Command
34202 The corresponding information is printed by @samp{info file} (among
34205 @subsubheading Example
34209 @subheading The @code{-target-list-parameters} Command
34210 @findex -target-list-parameters
34212 @subsubheading Synopsis
34215 -target-list-parameters
34221 @subsubheading @value{GDBN} Command
34225 @subsubheading Example
34228 @subheading The @code{-target-flash-erase} Command
34229 @findex -target-flash-erase
34231 @subsubheading Synopsis
34234 -target-flash-erase
34237 Erases all known flash memory regions on the target.
34239 The corresponding @value{GDBN} command is @samp{flash-erase}.
34241 The output is a list of flash regions that have been erased, with starting
34242 addresses and memory region sizes.
34246 -target-flash-erase
34247 ^done,erased-regions=@{address="0x0",size="0x40000"@}
34251 @subheading The @code{-target-select} Command
34252 @findex -target-select
34254 @subsubheading Synopsis
34257 -target-select @var{type} @var{parameters @dots{}}
34260 Connect @value{GDBN} to the remote target. This command takes two args:
34264 The type of target, for instance @samp{remote}, etc.
34265 @item @var{parameters}
34266 Device names, host names and the like. @xref{Target Commands, ,
34267 Commands for Managing Targets}, for more details.
34270 The output is a connection notification, followed by the address at
34271 which the target program is, in the following form:
34274 ^connected,addr="@var{address}",func="@var{function name}",
34275 args=[@var{arg list}]
34278 @subsubheading @value{GDBN} Command
34280 The corresponding @value{GDBN} command is @samp{target}.
34282 @subsubheading Example
34286 -target-select remote /dev/ttya
34287 ^connected,addr="0xfe00a300",func="??",args=[]
34291 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34292 @node GDB/MI File Transfer Commands
34293 @section @sc{gdb/mi} File Transfer Commands
34296 @subheading The @code{-target-file-put} Command
34297 @findex -target-file-put
34299 @subsubheading Synopsis
34302 -target-file-put @var{hostfile} @var{targetfile}
34305 Copy file @var{hostfile} from the host system (the machine running
34306 @value{GDBN}) to @var{targetfile} on the target system.
34308 @subsubheading @value{GDBN} Command
34310 The corresponding @value{GDBN} command is @samp{remote put}.
34312 @subsubheading Example
34316 -target-file-put localfile remotefile
34322 @subheading The @code{-target-file-get} Command
34323 @findex -target-file-get
34325 @subsubheading Synopsis
34328 -target-file-get @var{targetfile} @var{hostfile}
34331 Copy file @var{targetfile} from the target system to @var{hostfile}
34332 on the host system.
34334 @subsubheading @value{GDBN} Command
34336 The corresponding @value{GDBN} command is @samp{remote get}.
34338 @subsubheading Example
34342 -target-file-get remotefile localfile
34348 @subheading The @code{-target-file-delete} Command
34349 @findex -target-file-delete
34351 @subsubheading Synopsis
34354 -target-file-delete @var{targetfile}
34357 Delete @var{targetfile} from the target system.
34359 @subsubheading @value{GDBN} Command
34361 The corresponding @value{GDBN} command is @samp{remote delete}.
34363 @subsubheading Example
34367 -target-file-delete remotefile
34373 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34374 @node GDB/MI Ada Exceptions Commands
34375 @section Ada Exceptions @sc{gdb/mi} Commands
34377 @subheading The @code{-info-ada-exceptions} Command
34378 @findex -info-ada-exceptions
34380 @subsubheading Synopsis
34383 -info-ada-exceptions [ @var{regexp}]
34386 List all Ada exceptions defined within the program being debugged.
34387 With a regular expression @var{regexp}, only those exceptions whose
34388 names match @var{regexp} are listed.
34390 @subsubheading @value{GDBN} Command
34392 The corresponding @value{GDBN} command is @samp{info exceptions}.
34394 @subsubheading Result
34396 The result is a table of Ada exceptions. The following columns are
34397 defined for each exception:
34401 The name of the exception.
34404 The address of the exception.
34408 @subsubheading Example
34411 -info-ada-exceptions aint
34412 ^done,ada-exceptions=@{nr_rows="2",nr_cols="2",
34413 hdr=[@{width="1",alignment="-1",col_name="name",colhdr="Name"@},
34414 @{width="1",alignment="-1",col_name="address",colhdr="Address"@}],
34415 body=[@{name="constraint_error",address="0x0000000000613da0"@},
34416 @{name="const.aint_global_e",address="0x0000000000613b00"@}]@}
34419 @subheading Catching Ada Exceptions
34421 The commands describing how to ask @value{GDBN} to stop when a program
34422 raises an exception are described at @ref{Ada Exception GDB/MI
34423 Catchpoint Commands}.
34426 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34427 @node GDB/MI Support Commands
34428 @section @sc{gdb/mi} Support Commands
34430 Since new commands and features get regularly added to @sc{gdb/mi},
34431 some commands are available to help front-ends query the debugger
34432 about support for these capabilities. Similarly, it is also possible
34433 to query @value{GDBN} about target support of certain features.
34435 @subheading The @code{-info-gdb-mi-command} Command
34436 @cindex @code{-info-gdb-mi-command}
34437 @findex -info-gdb-mi-command
34439 @subsubheading Synopsis
34442 -info-gdb-mi-command @var{cmd_name}
34445 Query support for the @sc{gdb/mi} command named @var{cmd_name}.
34447 Note that the dash (@code{-}) starting all @sc{gdb/mi} commands
34448 is technically not part of the command name (@pxref{GDB/MI Input
34449 Syntax}), and thus should be omitted in @var{cmd_name}. However,
34450 for ease of use, this command also accepts the form with the leading
34453 @subsubheading @value{GDBN} Command
34455 There is no corresponding @value{GDBN} command.
34457 @subsubheading Result
34459 The result is a tuple. There is currently only one field:
34463 This field is equal to @code{"true"} if the @sc{gdb/mi} command exists,
34464 @code{"false"} otherwise.
34468 @subsubheading Example
34470 Here is an example where the @sc{gdb/mi} command does not exist:
34473 -info-gdb-mi-command unsupported-command
34474 ^done,command=@{exists="false"@}
34478 And here is an example where the @sc{gdb/mi} command is known
34482 -info-gdb-mi-command symbol-list-lines
34483 ^done,command=@{exists="true"@}
34486 @subheading The @code{-list-features} Command
34487 @findex -list-features
34488 @cindex supported @sc{gdb/mi} features, list
34490 Returns a list of particular features of the MI protocol that
34491 this version of gdb implements. A feature can be a command,
34492 or a new field in an output of some command, or even an
34493 important bugfix. While a frontend can sometimes detect presence
34494 of a feature at runtime, it is easier to perform detection at debugger
34497 The command returns a list of strings, with each string naming an
34498 available feature. Each returned string is just a name, it does not
34499 have any internal structure. The list of possible feature names
34505 (gdb) -list-features
34506 ^done,result=["feature1","feature2"]
34509 The current list of features is:
34512 @item frozen-varobjs
34513 Indicates support for the @code{-var-set-frozen} command, as well
34514 as possible presense of the @code{frozen} field in the output
34515 of @code{-varobj-create}.
34516 @item pending-breakpoints
34517 Indicates support for the @option{-f} option to the @code{-break-insert}
34520 Indicates Python scripting support, Python-based
34521 pretty-printing commands, and possible presence of the
34522 @samp{display_hint} field in the output of @code{-var-list-children}
34524 Indicates support for the @code{-thread-info} command.
34525 @item data-read-memory-bytes
34526 Indicates support for the @code{-data-read-memory-bytes} and the
34527 @code{-data-write-memory-bytes} commands.
34528 @item breakpoint-notifications
34529 Indicates that changes to breakpoints and breakpoints created via the
34530 CLI will be announced via async records.
34531 @item ada-task-info
34532 Indicates support for the @code{-ada-task-info} command.
34533 @item language-option
34534 Indicates that all @sc{gdb/mi} commands accept the @option{--language}
34535 option (@pxref{Context management}).
34536 @item info-gdb-mi-command
34537 Indicates support for the @code{-info-gdb-mi-command} command.
34538 @item undefined-command-error-code
34539 Indicates support for the "undefined-command" error code in error result
34540 records, produced when trying to execute an undefined @sc{gdb/mi} command
34541 (@pxref{GDB/MI Result Records}).
34542 @item exec-run-start-option
34543 Indicates that the @code{-exec-run} command supports the @option{--start}
34544 option (@pxref{GDB/MI Program Execution}).
34545 @item data-disassemble-a-option
34546 Indicates that the @code{-data-disassemble} command supports the @option{-a}
34547 option (@pxref{GDB/MI Data Manipulation}).
34550 @subheading The @code{-list-target-features} Command
34551 @findex -list-target-features
34553 Returns a list of particular features that are supported by the
34554 target. Those features affect the permitted MI commands, but
34555 unlike the features reported by the @code{-list-features} command, the
34556 features depend on which target GDB is using at the moment. Whenever
34557 a target can change, due to commands such as @code{-target-select},
34558 @code{-target-attach} or @code{-exec-run}, the list of target features
34559 may change, and the frontend should obtain it again.
34563 (gdb) -list-target-features
34564 ^done,result=["async"]
34567 The current list of features is:
34571 Indicates that the target is capable of asynchronous command
34572 execution, which means that @value{GDBN} will accept further commands
34573 while the target is running.
34576 Indicates that the target is capable of reverse execution.
34577 @xref{Reverse Execution}, for more information.
34581 @c %%%%%%%%%%%%%%%%%%%%%%%%%%%% SECTION %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
34582 @node GDB/MI Miscellaneous Commands
34583 @section Miscellaneous @sc{gdb/mi} Commands
34585 @c @subheading -gdb-complete
34587 @subheading The @code{-gdb-exit} Command
34590 @subsubheading Synopsis
34596 Exit @value{GDBN} immediately.
34598 @subsubheading @value{GDBN} Command
34600 Approximately corresponds to @samp{quit}.
34602 @subsubheading Example
34612 @subheading The @code{-exec-abort} Command
34613 @findex -exec-abort
34615 @subsubheading Synopsis
34621 Kill the inferior running program.
34623 @subsubheading @value{GDBN} Command
34625 The corresponding @value{GDBN} command is @samp{kill}.
34627 @subsubheading Example
34632 @subheading The @code{-gdb-set} Command
34635 @subsubheading Synopsis
34641 Set an internal @value{GDBN} variable.
34642 @c IS THIS A DOLLAR VARIABLE? OR SOMETHING LIKE ANNOTATE ?????
34644 @subsubheading @value{GDBN} Command
34646 The corresponding @value{GDBN} command is @samp{set}.
34648 @subsubheading Example
34658 @subheading The @code{-gdb-show} Command
34661 @subsubheading Synopsis
34667 Show the current value of a @value{GDBN} variable.
34669 @subsubheading @value{GDBN} Command
34671 The corresponding @value{GDBN} command is @samp{show}.
34673 @subsubheading Example
34682 @c @subheading -gdb-source
34685 @subheading The @code{-gdb-version} Command
34686 @findex -gdb-version
34688 @subsubheading Synopsis
34694 Show version information for @value{GDBN}. Used mostly in testing.
34696 @subsubheading @value{GDBN} Command
34698 The @value{GDBN} equivalent is @samp{show version}. @value{GDBN} by
34699 default shows this information when you start an interactive session.
34701 @subsubheading Example
34703 @c This example modifies the actual output from GDB to avoid overfull
34709 ~Copyright 2000 Free Software Foundation, Inc.
34710 ~GDB is free software, covered by the GNU General Public License, and
34711 ~you are welcome to change it and/or distribute copies of it under
34712 ~ certain conditions.
34713 ~Type "show copying" to see the conditions.
34714 ~There is absolutely no warranty for GDB. Type "show warranty" for
34716 ~This GDB was configured as
34717 "--host=sparc-sun-solaris2.5.1 --target=ppc-eabi".
34722 @subheading The @code{-list-thread-groups} Command
34723 @findex -list-thread-groups
34725 @subheading Synopsis
34728 -list-thread-groups [ --available ] [ --recurse 1 ] [ @var{group} ... ]
34731 Lists thread groups (@pxref{Thread groups}). When a single thread
34732 group is passed as the argument, lists the children of that group.
34733 When several thread group are passed, lists information about those
34734 thread groups. Without any parameters, lists information about all
34735 top-level thread groups.
34737 Normally, thread groups that are being debugged are reported.
34738 With the @samp{--available} option, @value{GDBN} reports thread groups
34739 available on the target.
34741 The output of this command may have either a @samp{threads} result or
34742 a @samp{groups} result. The @samp{thread} result has a list of tuples
34743 as value, with each tuple describing a thread (@pxref{GDB/MI Thread
34744 Information}). The @samp{groups} result has a list of tuples as value,
34745 each tuple describing a thread group. If top-level groups are
34746 requested (that is, no parameter is passed), or when several groups
34747 are passed, the output always has a @samp{groups} result. The format
34748 of the @samp{group} result is described below.
34750 To reduce the number of roundtrips it's possible to list thread groups
34751 together with their children, by passing the @samp{--recurse} option
34752 and the recursion depth. Presently, only recursion depth of 1 is
34753 permitted. If this option is present, then every reported thread group
34754 will also include its children, either as @samp{group} or
34755 @samp{threads} field.
34757 In general, any combination of option and parameters is permitted, with
34758 the following caveats:
34762 When a single thread group is passed, the output will typically
34763 be the @samp{threads} result. Because threads may not contain
34764 anything, the @samp{recurse} option will be ignored.
34767 When the @samp{--available} option is passed, limited information may
34768 be available. In particular, the list of threads of a process might
34769 be inaccessible. Further, specifying specific thread groups might
34770 not give any performance advantage over listing all thread groups.
34771 The frontend should assume that @samp{-list-thread-groups --available}
34772 is always an expensive operation and cache the results.
34776 The @samp{groups} result is a list of tuples, where each tuple may
34777 have the following fields:
34781 Identifier of the thread group. This field is always present.
34782 The identifier is an opaque string; frontends should not try to
34783 convert it to an integer, even though it might look like one.
34786 The type of the thread group. At present, only @samp{process} is a
34790 The target-specific process identifier. This field is only present
34791 for thread groups of type @samp{process} and only if the process exists.
34794 The exit code of this group's last exited thread, formatted in octal.
34795 This field is only present for thread groups of type @samp{process} and
34796 only if the process is not running.
34799 The number of children this thread group has. This field may be
34800 absent for an available thread group.
34803 This field has a list of tuples as value, each tuple describing a
34804 thread. It may be present if the @samp{--recurse} option is
34805 specified, and it's actually possible to obtain the threads.
34808 This field is a list of integers, each identifying a core that one
34809 thread of the group is running on. This field may be absent if
34810 such information is not available.
34813 The name of the executable file that corresponds to this thread group.
34814 The field is only present for thread groups of type @samp{process},
34815 and only if there is a corresponding executable file.
34819 @subheading Example
34823 -list-thread-groups
34824 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2"@}]
34825 -list-thread-groups 17
34826 ^done,threads=[@{id="2",target-id="Thread 0xb7e14b90 (LWP 21257)",
34827 frame=@{level="0",addr="0xffffe410",func="__kernel_vsyscall",args=[]@},state="running"@},
34828 @{id="1",target-id="Thread 0xb7e156b0 (LWP 21254)",
34829 frame=@{level="0",addr="0x0804891f",func="foo",args=[@{name="i",value="10"@}],
34830 file="/tmp/a.c",fullname="/tmp/a.c",line="158",arch="i386:x86_64"@},state="running"@}]]
34831 -list-thread-groups --available
34832 ^done,groups=[@{id="17",type="process",pid="yyy",num_children="2",cores=[1,2]@}]
34833 -list-thread-groups --available --recurse 1
34834 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
34835 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
34836 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},..]
34837 -list-thread-groups --available --recurse 1 17 18
34838 ^done,groups=[@{id="17", types="process",pid="yyy",num_children="2",cores=[1,2],
34839 threads=[@{id="1",target-id="Thread 0xb7e14b90",cores=[1]@},
34840 @{id="2",target-id="Thread 0xb7e14b90",cores=[2]@}]@},...]
34843 @subheading The @code{-info-os} Command
34846 @subsubheading Synopsis
34849 -info-os [ @var{type} ]
34852 If no argument is supplied, the command returns a table of available
34853 operating-system-specific information types. If one of these types is
34854 supplied as an argument @var{type}, then the command returns a table
34855 of data of that type.
34857 The types of information available depend on the target operating
34860 @subsubheading @value{GDBN} Command
34862 The corresponding @value{GDBN} command is @samp{info os}.
34864 @subsubheading Example
34866 When run on a @sc{gnu}/Linux system, the output will look something
34872 ^done,OSDataTable=@{nr_rows="10",nr_cols="3",
34873 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="Type"@},
34874 @{width="10",alignment="-1",col_name="col1",colhdr="Description"@},
34875 @{width="10",alignment="-1",col_name="col2",colhdr="Title"@}],
34876 body=[item=@{col0="cpus",col1="Listing of all cpus/cores on the system",
34878 item=@{col0="files",col1="Listing of all file descriptors",
34879 col2="File descriptors"@},
34880 item=@{col0="modules",col1="Listing of all loaded kernel modules",
34881 col2="Kernel modules"@},
34882 item=@{col0="msg",col1="Listing of all message queues",
34883 col2="Message queues"@},
34884 item=@{col0="processes",col1="Listing of all processes",
34885 col2="Processes"@},
34886 item=@{col0="procgroups",col1="Listing of all process groups",
34887 col2="Process groups"@},
34888 item=@{col0="semaphores",col1="Listing of all semaphores",
34889 col2="Semaphores"@},
34890 item=@{col0="shm",col1="Listing of all shared-memory regions",
34891 col2="Shared-memory regions"@},
34892 item=@{col0="sockets",col1="Listing of all internet-domain sockets",
34894 item=@{col0="threads",col1="Listing of all threads",
34898 ^done,OSDataTable=@{nr_rows="190",nr_cols="4",
34899 hdr=[@{width="10",alignment="-1",col_name="col0",colhdr="pid"@},
34900 @{width="10",alignment="-1",col_name="col1",colhdr="user"@},
34901 @{width="10",alignment="-1",col_name="col2",colhdr="command"@},
34902 @{width="10",alignment="-1",col_name="col3",colhdr="cores"@}],
34903 body=[item=@{col0="1",col1="root",col2="/sbin/init",col3="0"@},
34904 item=@{col0="2",col1="root",col2="[kthreadd]",col3="1"@},
34905 item=@{col0="3",col1="root",col2="[ksoftirqd/0]",col3="0"@},
34907 item=@{col0="26446",col1="stan",col2="bash",col3="0"@},
34908 item=@{col0="28152",col1="stan",col2="bash",col3="1"@}]@}
34912 (Note that the MI output here includes a @code{"Title"} column that
34913 does not appear in command-line @code{info os}; this column is useful
34914 for MI clients that want to enumerate the types of data, such as in a
34915 popup menu, but is needless clutter on the command line, and
34916 @code{info os} omits it.)
34918 @subheading The @code{-add-inferior} Command
34919 @findex -add-inferior
34921 @subheading Synopsis
34927 Creates a new inferior (@pxref{Inferiors and Programs}). The created
34928 inferior is not associated with any executable. Such association may
34929 be established with the @samp{-file-exec-and-symbols} command
34930 (@pxref{GDB/MI File Commands}). The command response has a single
34931 field, @samp{inferior}, whose value is the identifier of the
34932 thread group corresponding to the new inferior.
34934 @subheading Example
34939 ^done,inferior="i3"
34942 @subheading The @code{-interpreter-exec} Command
34943 @findex -interpreter-exec
34945 @subheading Synopsis
34948 -interpreter-exec @var{interpreter} @var{command}
34950 @anchor{-interpreter-exec}
34952 Execute the specified @var{command} in the given @var{interpreter}.
34954 @subheading @value{GDBN} Command
34956 The corresponding @value{GDBN} command is @samp{interpreter-exec}.
34958 @subheading Example
34962 -interpreter-exec console "break main"
34963 &"During symbol reading, couldn't parse type; debugger out of date?.\n"
34964 &"During symbol reading, bad structure-type format.\n"
34965 ~"Breakpoint 1 at 0x8074fc6: file ../../src/gdb/main.c, line 743.\n"
34970 @subheading The @code{-inferior-tty-set} Command
34971 @findex -inferior-tty-set
34973 @subheading Synopsis
34976 -inferior-tty-set /dev/pts/1
34979 Set terminal for future runs of the program being debugged.
34981 @subheading @value{GDBN} Command
34983 The corresponding @value{GDBN} command is @samp{set inferior-tty} /dev/pts/1.
34985 @subheading Example
34989 -inferior-tty-set /dev/pts/1
34994 @subheading The @code{-inferior-tty-show} Command
34995 @findex -inferior-tty-show
34997 @subheading Synopsis
35003 Show terminal for future runs of program being debugged.
35005 @subheading @value{GDBN} Command
35007 The corresponding @value{GDBN} command is @samp{show inferior-tty}.
35009 @subheading Example
35013 -inferior-tty-set /dev/pts/1
35017 ^done,inferior_tty_terminal="/dev/pts/1"
35021 @subheading The @code{-enable-timings} Command
35022 @findex -enable-timings
35024 @subheading Synopsis
35027 -enable-timings [yes | no]
35030 Toggle the printing of the wallclock, user and system times for an MI
35031 command as a field in its output. This command is to help frontend
35032 developers optimize the performance of their code. No argument is
35033 equivalent to @samp{yes}.
35035 @subheading @value{GDBN} Command
35039 @subheading Example
35047 ^done,bkpt=@{number="1",type="breakpoint",disp="keep",enabled="y",
35048 addr="0x080484ed",func="main",file="myprog.c",
35049 fullname="/home/nickrob/myprog.c",line="73",thread-groups=["i1"],
35051 time=@{wallclock="0.05185",user="0.00800",system="0.00000"@}
35059 *stopped,reason="breakpoint-hit",disp="keep",bkptno="1",thread-id="0",
35060 frame=@{addr="0x080484ed",func="main",args=[@{name="argc",value="1"@},
35061 @{name="argv",value="0xbfb60364"@}],file="myprog.c",
35062 fullname="/home/nickrob/myprog.c",line="73",arch="i386:x86_64"@}
35066 @subheading The @code{-complete} Command
35069 @subheading Synopsis
35072 -complete @var{command}
35075 Show a list of completions for partially typed CLI @var{command}.
35077 This command is intended for @sc{gdb/mi} frontends that cannot use two separate
35078 CLI and MI channels --- for example: because of lack of PTYs like on Windows or
35079 because @value{GDBN} is used remotely via a SSH connection.
35083 The result consists of two or three fields:
35087 This field contains the completed @var{command}. If @var{command}
35088 has no known completions, this field is omitted.
35091 This field contains a (possibly empty) array of matches. It is always present.
35093 @item max_completions_reached
35094 This field contains @code{1} if number of known completions is above
35095 @code{max-completions} limit (@pxref{Completion}), otherwise it contains
35096 @code{0}. It is always present.
35100 @subheading @value{GDBN} Command
35102 The corresponding @value{GDBN} command is @samp{complete}.
35104 @subheading Example
35109 ^done,completion="break",
35110 matches=["break","break-range"],
35111 max_completions_reached="0"
35114 ^done,completion="b ma",
35115 matches=["b madvise","b main"],max_completions_reached="0"
35117 -complete "b push_b"
35118 ^done,completion="b push_back(",
35120 "b A::push_back(void*)",
35121 "b std::string::push_back(char)",
35122 "b std::vector<int, std::allocator<int> >::push_back(int&&)"],
35123 max_completions_reached="0"
35125 -complete "nonexist"
35126 ^done,matches=[],max_completions_reached="0"
35132 @chapter @value{GDBN} Annotations
35134 This chapter describes annotations in @value{GDBN}. Annotations were
35135 designed to interface @value{GDBN} to graphical user interfaces or other
35136 similar programs which want to interact with @value{GDBN} at a
35137 relatively high level.
35139 The annotation mechanism has largely been superseded by @sc{gdb/mi}
35143 This is Edition @value{EDITION}, @value{DATE}.
35147 * Annotations Overview:: What annotations are; the general syntax.
35148 * Server Prefix:: Issuing a command without affecting user state.
35149 * Prompting:: Annotations marking @value{GDBN}'s need for input.
35150 * Errors:: Annotations for error messages.
35151 * Invalidation:: Some annotations describe things now invalid.
35152 * Annotations for Running::
35153 Whether the program is running, how it stopped, etc.
35154 * Source Annotations:: Annotations describing source code.
35157 @node Annotations Overview
35158 @section What is an Annotation?
35159 @cindex annotations
35161 Annotations start with a newline character, two @samp{control-z}
35162 characters, and the name of the annotation. If there is no additional
35163 information associated with this annotation, the name of the annotation
35164 is followed immediately by a newline. If there is additional
35165 information, the name of the annotation is followed by a space, the
35166 additional information, and a newline. The additional information
35167 cannot contain newline characters.
35169 Any output not beginning with a newline and two @samp{control-z}
35170 characters denotes literal output from @value{GDBN}. Currently there is
35171 no need for @value{GDBN} to output a newline followed by two
35172 @samp{control-z} characters, but if there was such a need, the
35173 annotations could be extended with an @samp{escape} annotation which
35174 means those three characters as output.
35176 The annotation @var{level}, which is specified using the
35177 @option{--annotate} command line option (@pxref{Mode Options}), controls
35178 how much information @value{GDBN} prints together with its prompt,
35179 values of expressions, source lines, and other types of output. Level 0
35180 is for no annotations, level 1 is for use when @value{GDBN} is run as a
35181 subprocess of @sc{gnu} Emacs, level 3 is the maximum annotation suitable
35182 for programs that control @value{GDBN}, and level 2 annotations have
35183 been made obsolete (@pxref{Limitations, , Limitations of the Annotation
35184 Interface, annotate, GDB's Obsolete Annotations}).
35187 @kindex set annotate
35188 @item set annotate @var{level}
35189 The @value{GDBN} command @code{set annotate} sets the level of
35190 annotations to the specified @var{level}.
35192 @item show annotate
35193 @kindex show annotate
35194 Show the current annotation level.
35197 This chapter describes level 3 annotations.
35199 A simple example of starting up @value{GDBN} with annotations is:
35202 $ @kbd{gdb --annotate=3}
35204 Copyright 2003 Free Software Foundation, Inc.
35205 GDB is free software, covered by the GNU General Public License,
35206 and you are welcome to change it and/or distribute copies of it
35207 under certain conditions.
35208 Type "show copying" to see the conditions.
35209 There is absolutely no warranty for GDB. Type "show warranty"
35211 This GDB was configured as "i386-pc-linux-gnu"
35222 Here @samp{quit} is input to @value{GDBN}; the rest is output from
35223 @value{GDBN}. The three lines beginning @samp{^Z^Z} (where @samp{^Z}
35224 denotes a @samp{control-z} character) are annotations; the rest is
35225 output from @value{GDBN}.
35227 @node Server Prefix
35228 @section The Server Prefix
35229 @cindex server prefix
35231 If you prefix a command with @samp{server } then it will not affect
35232 the command history, nor will it affect @value{GDBN}'s notion of which
35233 command to repeat if @key{RET} is pressed on a line by itself. This
35234 means that commands can be run behind a user's back by a front-end in
35235 a transparent manner.
35237 The @code{server } prefix does not affect the recording of values into
35238 the value history; to print a value without recording it into the
35239 value history, use the @code{output} command instead of the
35240 @code{print} command.
35242 Using this prefix also disables confirmation requests
35243 (@pxref{confirmation requests}).
35246 @section Annotation for @value{GDBN} Input
35248 @cindex annotations for prompts
35249 When @value{GDBN} prompts for input, it annotates this fact so it is possible
35250 to know when to send output, when the output from a given command is
35253 Different kinds of input each have a different @dfn{input type}. Each
35254 input type has three annotations: a @code{pre-} annotation, which
35255 denotes the beginning of any prompt which is being output, a plain
35256 annotation, which denotes the end of the prompt, and then a @code{post-}
35257 annotation which denotes the end of any echo which may (or may not) be
35258 associated with the input. For example, the @code{prompt} input type
35259 features the following annotations:
35267 The input types are
35270 @findex pre-prompt annotation
35271 @findex prompt annotation
35272 @findex post-prompt annotation
35274 When @value{GDBN} is prompting for a command (the main @value{GDBN} prompt).
35276 @findex pre-commands annotation
35277 @findex commands annotation
35278 @findex post-commands annotation
35280 When @value{GDBN} prompts for a set of commands, like in the @code{commands}
35281 command. The annotations are repeated for each command which is input.
35283 @findex pre-overload-choice annotation
35284 @findex overload-choice annotation
35285 @findex post-overload-choice annotation
35286 @item overload-choice
35287 When @value{GDBN} wants the user to select between various overloaded functions.
35289 @findex pre-query annotation
35290 @findex query annotation
35291 @findex post-query annotation
35293 When @value{GDBN} wants the user to confirm a potentially dangerous operation.
35295 @findex pre-prompt-for-continue annotation
35296 @findex prompt-for-continue annotation
35297 @findex post-prompt-for-continue annotation
35298 @item prompt-for-continue
35299 When @value{GDBN} is asking the user to press return to continue. Note: Don't
35300 expect this to work well; instead use @code{set height 0} to disable
35301 prompting. This is because the counting of lines is buggy in the
35302 presence of annotations.
35307 @cindex annotations for errors, warnings and interrupts
35309 @findex quit annotation
35314 This annotation occurs right before @value{GDBN} responds to an interrupt.
35316 @findex error annotation
35321 This annotation occurs right before @value{GDBN} responds to an error.
35323 Quit and error annotations indicate that any annotations which @value{GDBN} was
35324 in the middle of may end abruptly. For example, if a
35325 @code{value-history-begin} annotation is followed by a @code{error}, one
35326 cannot expect to receive the matching @code{value-history-end}. One
35327 cannot expect not to receive it either, however; an error annotation
35328 does not necessarily mean that @value{GDBN} is immediately returning all the way
35331 @findex error-begin annotation
35332 A quit or error annotation may be preceded by
35338 Any output between that and the quit or error annotation is the error
35341 Warning messages are not yet annotated.
35342 @c If we want to change that, need to fix warning(), type_error(),
35343 @c range_error(), and possibly other places.
35346 @section Invalidation Notices
35348 @cindex annotations for invalidation messages
35349 The following annotations say that certain pieces of state may have
35353 @findex frames-invalid annotation
35354 @item ^Z^Zframes-invalid
35356 The frames (for example, output from the @code{backtrace} command) may
35359 @findex breakpoints-invalid annotation
35360 @item ^Z^Zbreakpoints-invalid
35362 The breakpoints may have changed. For example, the user just added or
35363 deleted a breakpoint.
35366 @node Annotations for Running
35367 @section Running the Program
35368 @cindex annotations for running programs
35370 @findex starting annotation
35371 @findex stopping annotation
35372 When the program starts executing due to a @value{GDBN} command such as
35373 @code{step} or @code{continue},
35379 is output. When the program stops,
35385 is output. Before the @code{stopped} annotation, a variety of
35386 annotations describe how the program stopped.
35389 @findex exited annotation
35390 @item ^Z^Zexited @var{exit-status}
35391 The program exited, and @var{exit-status} is the exit status (zero for
35392 successful exit, otherwise nonzero).
35394 @findex signalled annotation
35395 @findex signal-name annotation
35396 @findex signal-name-end annotation
35397 @findex signal-string annotation
35398 @findex signal-string-end annotation
35399 @item ^Z^Zsignalled
35400 The program exited with a signal. After the @code{^Z^Zsignalled}, the
35401 annotation continues:
35407 ^Z^Zsignal-name-end
35411 ^Z^Zsignal-string-end
35416 where @var{name} is the name of the signal, such as @code{SIGILL} or
35417 @code{SIGSEGV}, and @var{string} is the explanation of the signal, such
35418 as @code{Illegal Instruction} or @code{Segmentation fault}. The arguments
35419 @var{intro-text}, @var{middle-text}, and @var{end-text} are for the
35420 user's benefit and have no particular format.
35422 @findex signal annotation
35424 The syntax of this annotation is just like @code{signalled}, but @value{GDBN} is
35425 just saying that the program received the signal, not that it was
35426 terminated with it.
35428 @findex breakpoint annotation
35429 @item ^Z^Zbreakpoint @var{number}
35430 The program hit breakpoint number @var{number}.
35432 @findex watchpoint annotation
35433 @item ^Z^Zwatchpoint @var{number}
35434 The program hit watchpoint number @var{number}.
35437 @node Source Annotations
35438 @section Displaying Source
35439 @cindex annotations for source display
35441 @findex source annotation
35442 The following annotation is used instead of displaying source code:
35445 ^Z^Zsource @var{filename}:@var{line}:@var{character}:@var{middle}:@var{addr}
35448 where @var{filename} is an absolute file name indicating which source
35449 file, @var{line} is the line number within that file (where 1 is the
35450 first line in the file), @var{character} is the character position
35451 within the file (where 0 is the first character in the file) (for most
35452 debug formats this will necessarily point to the beginning of a line),
35453 @var{middle} is @samp{middle} if @var{addr} is in the middle of the
35454 line, or @samp{beg} if @var{addr} is at the beginning of the line, and
35455 @var{addr} is the address in the target program associated with the
35456 source which is being displayed. The @var{addr} is in the form @samp{0x}
35457 followed by one or more lowercase hex digits (note that this does not
35458 depend on the language).
35460 @node JIT Interface
35461 @chapter JIT Compilation Interface
35462 @cindex just-in-time compilation
35463 @cindex JIT compilation interface
35465 This chapter documents @value{GDBN}'s @dfn{just-in-time} (JIT) compilation
35466 interface. A JIT compiler is a program or library that generates native
35467 executable code at runtime and executes it, usually in order to achieve good
35468 performance while maintaining platform independence.
35470 Programs that use JIT compilation are normally difficult to debug because
35471 portions of their code are generated at runtime, instead of being loaded from
35472 object files, which is where @value{GDBN} normally finds the program's symbols
35473 and debug information. In order to debug programs that use JIT compilation,
35474 @value{GDBN} has an interface that allows the program to register in-memory
35475 symbol files with @value{GDBN} at runtime.
35477 If you are using @value{GDBN} to debug a program that uses this interface, then
35478 it should work transparently so long as you have not stripped the binary. If
35479 you are developing a JIT compiler, then the interface is documented in the rest
35480 of this chapter. At this time, the only known client of this interface is the
35483 Broadly speaking, the JIT interface mirrors the dynamic loader interface. The
35484 JIT compiler communicates with @value{GDBN} by writing data into a global
35485 variable and calling a fuction at a well-known symbol. When @value{GDBN}
35486 attaches, it reads a linked list of symbol files from the global variable to
35487 find existing code, and puts a breakpoint in the function so that it can find
35488 out about additional code.
35491 * Declarations:: Relevant C struct declarations
35492 * Registering Code:: Steps to register code
35493 * Unregistering Code:: Steps to unregister code
35494 * Custom Debug Info:: Emit debug information in a custom format
35498 @section JIT Declarations
35500 These are the relevant struct declarations that a C program should include to
35501 implement the interface:
35511 struct jit_code_entry
35513 struct jit_code_entry *next_entry;
35514 struct jit_code_entry *prev_entry;
35515 const char *symfile_addr;
35516 uint64_t symfile_size;
35519 struct jit_descriptor
35522 /* This type should be jit_actions_t, but we use uint32_t
35523 to be explicit about the bitwidth. */
35524 uint32_t action_flag;
35525 struct jit_code_entry *relevant_entry;
35526 struct jit_code_entry *first_entry;
35529 /* GDB puts a breakpoint in this function. */
35530 void __attribute__((noinline)) __jit_debug_register_code() @{ @};
35532 /* Make sure to specify the version statically, because the
35533 debugger may check the version before we can set it. */
35534 struct jit_descriptor __jit_debug_descriptor = @{ 1, 0, 0, 0 @};
35537 If the JIT is multi-threaded, then it is important that the JIT synchronize any
35538 modifications to this global data properly, which can easily be done by putting
35539 a global mutex around modifications to these structures.
35541 @node Registering Code
35542 @section Registering Code
35544 To register code with @value{GDBN}, the JIT should follow this protocol:
35548 Generate an object file in memory with symbols and other desired debug
35549 information. The file must include the virtual addresses of the sections.
35552 Create a code entry for the file, which gives the start and size of the symbol
35556 Add it to the linked list in the JIT descriptor.
35559 Point the relevant_entry field of the descriptor at the entry.
35562 Set @code{action_flag} to @code{JIT_REGISTER} and call
35563 @code{__jit_debug_register_code}.
35566 When @value{GDBN} is attached and the breakpoint fires, @value{GDBN} uses the
35567 @code{relevant_entry} pointer so it doesn't have to walk the list looking for
35568 new code. However, the linked list must still be maintained in order to allow
35569 @value{GDBN} to attach to a running process and still find the symbol files.
35571 @node Unregistering Code
35572 @section Unregistering Code
35574 If code is freed, then the JIT should use the following protocol:
35578 Remove the code entry corresponding to the code from the linked list.
35581 Point the @code{relevant_entry} field of the descriptor at the code entry.
35584 Set @code{action_flag} to @code{JIT_UNREGISTER} and call
35585 @code{__jit_debug_register_code}.
35588 If the JIT frees or recompiles code without unregistering it, then @value{GDBN}
35589 and the JIT will leak the memory used for the associated symbol files.
35591 @node Custom Debug Info
35592 @section Custom Debug Info
35593 @cindex custom JIT debug info
35594 @cindex JIT debug info reader
35596 Generating debug information in platform-native file formats (like ELF
35597 or COFF) may be an overkill for JIT compilers; especially if all the
35598 debug info is used for is displaying a meaningful backtrace. The
35599 issue can be resolved by having the JIT writers decide on a debug info
35600 format and also provide a reader that parses the debug info generated
35601 by the JIT compiler. This section gives a brief overview on writing
35602 such a parser. More specific details can be found in the source file
35603 @file{gdb/jit-reader.in}, which is also installed as a header at
35604 @file{@var{includedir}/gdb/jit-reader.h} for easy inclusion.
35606 The reader is implemented as a shared object (so this functionality is
35607 not available on platforms which don't allow loading shared objects at
35608 runtime). Two @value{GDBN} commands, @code{jit-reader-load} and
35609 @code{jit-reader-unload} are provided, to be used to load and unload
35610 the readers from a preconfigured directory. Once loaded, the shared
35611 object is used the parse the debug information emitted by the JIT
35615 * Using JIT Debug Info Readers:: How to use supplied readers correctly
35616 * Writing JIT Debug Info Readers:: Creating a debug-info reader
35619 @node Using JIT Debug Info Readers
35620 @subsection Using JIT Debug Info Readers
35621 @kindex jit-reader-load
35622 @kindex jit-reader-unload
35624 Readers can be loaded and unloaded using the @code{jit-reader-load}
35625 and @code{jit-reader-unload} commands.
35628 @item jit-reader-load @var{reader}
35629 Load the JIT reader named @var{reader}, which is a shared
35630 object specified as either an absolute or a relative file name. In
35631 the latter case, @value{GDBN} will try to load the reader from a
35632 pre-configured directory, usually @file{@var{libdir}/gdb/} on a UNIX
35633 system (here @var{libdir} is the system library directory, often
35634 @file{/usr/local/lib}).
35636 Only one reader can be active at a time; trying to load a second
35637 reader when one is already loaded will result in @value{GDBN}
35638 reporting an error. A new JIT reader can be loaded by first unloading
35639 the current one using @code{jit-reader-unload} and then invoking
35640 @code{jit-reader-load}.
35642 @item jit-reader-unload
35643 Unload the currently loaded JIT reader.
35647 @node Writing JIT Debug Info Readers
35648 @subsection Writing JIT Debug Info Readers
35649 @cindex writing JIT debug info readers
35651 As mentioned, a reader is essentially a shared object conforming to a
35652 certain ABI. This ABI is described in @file{jit-reader.h}.
35654 @file{jit-reader.h} defines the structures, macros and functions
35655 required to write a reader. It is installed (along with
35656 @value{GDBN}), in @file{@var{includedir}/gdb} where @var{includedir} is
35657 the system include directory.
35659 Readers need to be released under a GPL compatible license. A reader
35660 can be declared as released under such a license by placing the macro
35661 @code{GDB_DECLARE_GPL_COMPATIBLE_READER} in a source file.
35663 The entry point for readers is the symbol @code{gdb_init_reader},
35664 which is expected to be a function with the prototype
35666 @findex gdb_init_reader
35668 extern struct gdb_reader_funcs *gdb_init_reader (void);
35671 @cindex @code{struct gdb_reader_funcs}
35673 @code{struct gdb_reader_funcs} contains a set of pointers to callback
35674 functions. These functions are executed to read the debug info
35675 generated by the JIT compiler (@code{read}), to unwind stack frames
35676 (@code{unwind}) and to create canonical frame IDs
35677 (@code{get_Frame_id}). It also has a callback that is called when the
35678 reader is being unloaded (@code{destroy}). The struct looks like this
35681 struct gdb_reader_funcs
35683 /* Must be set to GDB_READER_INTERFACE_VERSION. */
35684 int reader_version;
35686 /* For use by the reader. */
35689 gdb_read_debug_info *read;
35690 gdb_unwind_frame *unwind;
35691 gdb_get_frame_id *get_frame_id;
35692 gdb_destroy_reader *destroy;
35696 @cindex @code{struct gdb_symbol_callbacks}
35697 @cindex @code{struct gdb_unwind_callbacks}
35699 The callbacks are provided with another set of callbacks by
35700 @value{GDBN} to do their job. For @code{read}, these callbacks are
35701 passed in a @code{struct gdb_symbol_callbacks} and for @code{unwind}
35702 and @code{get_frame_id}, in a @code{struct gdb_unwind_callbacks}.
35703 @code{struct gdb_symbol_callbacks} has callbacks to create new object
35704 files and new symbol tables inside those object files. @code{struct
35705 gdb_unwind_callbacks} has callbacks to read registers off the current
35706 frame and to write out the values of the registers in the previous
35707 frame. Both have a callback (@code{target_read}) to read bytes off the
35708 target's address space.
35710 @node In-Process Agent
35711 @chapter In-Process Agent
35712 @cindex debugging agent
35713 The traditional debugging model is conceptually low-speed, but works fine,
35714 because most bugs can be reproduced in debugging-mode execution. However,
35715 as multi-core or many-core processors are becoming mainstream, and
35716 multi-threaded programs become more and more popular, there should be more
35717 and more bugs that only manifest themselves at normal-mode execution, for
35718 example, thread races, because debugger's interference with the program's
35719 timing may conceal the bugs. On the other hand, in some applications,
35720 it is not feasible for the debugger to interrupt the program's execution
35721 long enough for the developer to learn anything helpful about its behavior.
35722 If the program's correctness depends on its real-time behavior, delays
35723 introduced by a debugger might cause the program to fail, even when the
35724 code itself is correct. It is useful to be able to observe the program's
35725 behavior without interrupting it.
35727 Therefore, traditional debugging model is too intrusive to reproduce
35728 some bugs. In order to reduce the interference with the program, we can
35729 reduce the number of operations performed by debugger. The
35730 @dfn{In-Process Agent}, a shared library, is running within the same
35731 process with inferior, and is able to perform some debugging operations
35732 itself. As a result, debugger is only involved when necessary, and
35733 performance of debugging can be improved accordingly. Note that
35734 interference with program can be reduced but can't be removed completely,
35735 because the in-process agent will still stop or slow down the program.
35737 The in-process agent can interpret and execute Agent Expressions
35738 (@pxref{Agent Expressions}) during performing debugging operations. The
35739 agent expressions can be used for different purposes, such as collecting
35740 data in tracepoints, and condition evaluation in breakpoints.
35742 @anchor{Control Agent}
35743 You can control whether the in-process agent is used as an aid for
35744 debugging with the following commands:
35747 @kindex set agent on
35749 Causes the in-process agent to perform some operations on behalf of the
35750 debugger. Just which operations requested by the user will be done
35751 by the in-process agent depends on the its capabilities. For example,
35752 if you request to evaluate breakpoint conditions in the in-process agent,
35753 and the in-process agent has such capability as well, then breakpoint
35754 conditions will be evaluated in the in-process agent.
35756 @kindex set agent off
35757 @item set agent off
35758 Disables execution of debugging operations by the in-process agent. All
35759 of the operations will be performed by @value{GDBN}.
35763 Display the current setting of execution of debugging operations by
35764 the in-process agent.
35768 * In-Process Agent Protocol::
35771 @node In-Process Agent Protocol
35772 @section In-Process Agent Protocol
35773 @cindex in-process agent protocol
35775 The in-process agent is able to communicate with both @value{GDBN} and
35776 GDBserver (@pxref{In-Process Agent}). This section documents the protocol
35777 used for communications between @value{GDBN} or GDBserver and the IPA.
35778 In general, @value{GDBN} or GDBserver sends commands
35779 (@pxref{IPA Protocol Commands}) and data to in-process agent, and then
35780 in-process agent replies back with the return result of the command, or
35781 some other information. The data sent to in-process agent is composed
35782 of primitive data types, such as 4-byte or 8-byte type, and composite
35783 types, which are called objects (@pxref{IPA Protocol Objects}).
35786 * IPA Protocol Objects::
35787 * IPA Protocol Commands::
35790 @node IPA Protocol Objects
35791 @subsection IPA Protocol Objects
35792 @cindex ipa protocol objects
35794 The commands sent to and results received from agent may contain some
35795 complex data types called @dfn{objects}.
35797 The in-process agent is running on the same machine with @value{GDBN}
35798 or GDBserver, so it doesn't have to handle as much differences between
35799 two ends as remote protocol (@pxref{Remote Protocol}) tries to handle.
35800 However, there are still some differences of two ends in two processes:
35804 word size. On some 64-bit machines, @value{GDBN} or GDBserver can be
35805 compiled as a 64-bit executable, while in-process agent is a 32-bit one.
35807 ABI. Some machines may have multiple types of ABI, @value{GDBN} or
35808 GDBserver is compiled with one, and in-process agent is compiled with
35812 Here are the IPA Protocol Objects:
35816 agent expression object. It represents an agent expression
35817 (@pxref{Agent Expressions}).
35818 @anchor{agent expression object}
35820 tracepoint action object. It represents a tracepoint action
35821 (@pxref{Tracepoint Actions,,Tracepoint Action Lists}) to collect registers,
35822 memory, static trace data and to evaluate expression.
35823 @anchor{tracepoint action object}
35825 tracepoint object. It represents a tracepoint (@pxref{Tracepoints}).
35826 @anchor{tracepoint object}
35830 The following table describes important attributes of each IPA protocol
35833 @multitable @columnfractions .30 .20 .50
35834 @headitem Name @tab Size @tab Description
35835 @item @emph{agent expression object} @tab @tab
35836 @item length @tab 4 @tab length of bytes code
35837 @item byte code @tab @var{length} @tab contents of byte code
35838 @item @emph{tracepoint action for collecting memory} @tab @tab
35839 @item 'M' @tab 1 @tab type of tracepoint action
35840 @item addr @tab 8 @tab if @var{basereg} is @samp{-1}, @var{addr} is the
35841 address of the lowest byte to collect, otherwise @var{addr} is the offset
35842 of @var{basereg} for memory collecting.
35843 @item len @tab 8 @tab length of memory for collecting
35844 @item basereg @tab 4 @tab the register number containing the starting
35845 memory address for collecting.
35846 @item @emph{tracepoint action for collecting registers} @tab @tab
35847 @item 'R' @tab 1 @tab type of tracepoint action
35848 @item @emph{tracepoint action for collecting static trace data} @tab @tab
35849 @item 'L' @tab 1 @tab type of tracepoint action
35850 @item @emph{tracepoint action for expression evaluation} @tab @tab
35851 @item 'X' @tab 1 @tab type of tracepoint action
35852 @item agent expression @tab length of @tab @ref{agent expression object}
35853 @item @emph{tracepoint object} @tab @tab
35854 @item number @tab 4 @tab number of tracepoint
35855 @item address @tab 8 @tab address of tracepoint inserted on
35856 @item type @tab 4 @tab type of tracepoint
35857 @item enabled @tab 1 @tab enable or disable of tracepoint
35858 @item step_count @tab 8 @tab step
35859 @item pass_count @tab 8 @tab pass
35860 @item numactions @tab 4 @tab number of tracepoint actions
35861 @item hit count @tab 8 @tab hit count
35862 @item trace frame usage @tab 8 @tab trace frame usage
35863 @item compiled_cond @tab 8 @tab compiled condition
35864 @item orig_size @tab 8 @tab orig size
35865 @item condition @tab 4 if condition is NULL otherwise length of
35866 @ref{agent expression object}
35867 @tab zero if condition is NULL, otherwise is
35868 @ref{agent expression object}
35869 @item actions @tab variable
35870 @tab numactions number of @ref{tracepoint action object}
35873 @node IPA Protocol Commands
35874 @subsection IPA Protocol Commands
35875 @cindex ipa protocol commands
35877 The spaces in each command are delimiters to ease reading this commands
35878 specification. They don't exist in real commands.
35882 @item FastTrace:@var{tracepoint_object} @var{gdb_jump_pad_head}
35883 Installs a new fast tracepoint described by @var{tracepoint_object}
35884 (@pxref{tracepoint object}). The @var{gdb_jump_pad_head}, 8-byte long, is the
35885 head of @dfn{jumppad}, which is used to jump to data collection routine
35890 @item OK @var{target_address} @var{gdb_jump_pad_head} @var{fjump_size} @var{fjump}
35891 @var{target_address} is address of tracepoint in the inferior.
35892 The @var{gdb_jump_pad_head} is updated head of jumppad. Both of
35893 @var{target_address} and @var{gdb_jump_pad_head} are 8-byte long.
35894 The @var{fjump} contains a sequence of instructions jump to jumppad entry.
35895 The @var{fjump_size}, 4-byte long, is the size of @var{fjump}.
35902 Closes the in-process agent. This command is sent when @value{GDBN} or GDBserver
35903 is about to kill inferiors.
35911 @item probe_marker_at:@var{address}
35912 Asks in-process agent to probe the marker at @var{address}.
35919 @item unprobe_marker_at:@var{address}
35920 Asks in-process agent to unprobe the marker at @var{address}.
35924 @chapter Reporting Bugs in @value{GDBN}
35925 @cindex bugs in @value{GDBN}
35926 @cindex reporting bugs in @value{GDBN}
35928 Your bug reports play an essential role in making @value{GDBN} reliable.
35930 Reporting a bug may help you by bringing a solution to your problem, or it
35931 may not. But in any case the principal function of a bug report is to help
35932 the entire community by making the next version of @value{GDBN} work better. Bug
35933 reports are your contribution to the maintenance of @value{GDBN}.
35935 In order for a bug report to serve its purpose, you must include the
35936 information that enables us to fix the bug.
35939 * Bug Criteria:: Have you found a bug?
35940 * Bug Reporting:: How to report bugs
35944 @section Have You Found a Bug?
35945 @cindex bug criteria
35947 If you are not sure whether you have found a bug, here are some guidelines:
35950 @cindex fatal signal
35951 @cindex debugger crash
35952 @cindex crash of debugger
35954 If the debugger gets a fatal signal, for any input whatever, that is a
35955 @value{GDBN} bug. Reliable debuggers never crash.
35957 @cindex error on valid input
35959 If @value{GDBN} produces an error message for valid input, that is a
35960 bug. (Note that if you're cross debugging, the problem may also be
35961 somewhere in the connection to the target.)
35963 @cindex invalid input
35965 If @value{GDBN} does not produce an error message for invalid input,
35966 that is a bug. However, you should note that your idea of
35967 ``invalid input'' might be our idea of ``an extension'' or ``support
35968 for traditional practice''.
35971 If you are an experienced user of debugging tools, your suggestions
35972 for improvement of @value{GDBN} are welcome in any case.
35975 @node Bug Reporting
35976 @section How to Report Bugs
35977 @cindex bug reports
35978 @cindex @value{GDBN} bugs, reporting
35980 A number of companies and individuals offer support for @sc{gnu} products.
35981 If you obtained @value{GDBN} from a support organization, we recommend you
35982 contact that organization first.
35984 You can find contact information for many support companies and
35985 individuals in the file @file{etc/SERVICE} in the @sc{gnu} Emacs
35987 @c should add a web page ref...
35990 @ifset BUGURL_DEFAULT
35991 In any event, we also recommend that you submit bug reports for
35992 @value{GDBN}. The preferred method is to submit them directly using
35993 @uref{http://www.gnu.org/software/gdb/bugs/, @value{GDBN}'s Bugs web
35994 page}. Alternatively, the @email{bug-gdb@@gnu.org, e-mail gateway} can
35997 @strong{Do not send bug reports to @samp{info-gdb}, or to
35998 @samp{help-gdb}, or to any newsgroups.} Most users of @value{GDBN} do
35999 not want to receive bug reports. Those that do have arranged to receive
36002 The mailing list @samp{bug-gdb} has a newsgroup @samp{gnu.gdb.bug} which
36003 serves as a repeater. The mailing list and the newsgroup carry exactly
36004 the same messages. Often people think of posting bug reports to the
36005 newsgroup instead of mailing them. This appears to work, but it has one
36006 problem which can be crucial: a newsgroup posting often lacks a mail
36007 path back to the sender. Thus, if we need to ask for more information,
36008 we may be unable to reach you. For this reason, it is better to send
36009 bug reports to the mailing list.
36011 @ifclear BUGURL_DEFAULT
36012 In any event, we also recommend that you submit bug reports for
36013 @value{GDBN} to @value{BUGURL}.
36017 The fundamental principle of reporting bugs usefully is this:
36018 @strong{report all the facts}. If you are not sure whether to state a
36019 fact or leave it out, state it!
36021 Often people omit facts because they think they know what causes the
36022 problem and assume that some details do not matter. Thus, you might
36023 assume that the name of the variable you use in an example does not matter.
36024 Well, probably it does not, but one cannot be sure. Perhaps the bug is a
36025 stray memory reference which happens to fetch from the location where that
36026 name is stored in memory; perhaps, if the name were different, the contents
36027 of that location would fool the debugger into doing the right thing despite
36028 the bug. Play it safe and give a specific, complete example. That is the
36029 easiest thing for you to do, and the most helpful.
36031 Keep in mind that the purpose of a bug report is to enable us to fix the
36032 bug. It may be that the bug has been reported previously, but neither
36033 you nor we can know that unless your bug report is complete and
36036 Sometimes people give a few sketchy facts and ask, ``Does this ring a
36037 bell?'' Those bug reports are useless, and we urge everyone to
36038 @emph{refuse to respond to them} except to chide the sender to report
36041 To enable us to fix the bug, you should include all these things:
36045 The version of @value{GDBN}. @value{GDBN} announces it if you start
36046 with no arguments; you can also print it at any time using @code{show
36049 Without this, we will not know whether there is any point in looking for
36050 the bug in the current version of @value{GDBN}.
36053 The type of machine you are using, and the operating system name and
36057 The details of the @value{GDBN} build-time configuration.
36058 @value{GDBN} shows these details if you invoke it with the
36059 @option{--configuration} command-line option, or if you type
36060 @code{show configuration} at @value{GDBN}'s prompt.
36063 What compiler (and its version) was used to compile @value{GDBN}---e.g.@:
36064 ``@value{GCC}--2.8.1''.
36067 What compiler (and its version) was used to compile the program you are
36068 debugging---e.g.@: ``@value{GCC}--2.8.1'', or ``HP92453-01 A.10.32.03 HP
36069 C Compiler''. For @value{NGCC}, you can say @kbd{@value{GCC} --version}
36070 to get this information; for other compilers, see the documentation for
36074 The command arguments you gave the compiler to compile your example and
36075 observe the bug. For example, did you use @samp{-O}? To guarantee
36076 you will not omit something important, list them all. A copy of the
36077 Makefile (or the output from make) is sufficient.
36079 If we were to try to guess the arguments, we would probably guess wrong
36080 and then we might not encounter the bug.
36083 A complete input script, and all necessary source files, that will
36087 A description of what behavior you observe that you believe is
36088 incorrect. For example, ``It gets a fatal signal.''
36090 Of course, if the bug is that @value{GDBN} gets a fatal signal, then we
36091 will certainly notice it. But if the bug is incorrect output, we might
36092 not notice unless it is glaringly wrong. You might as well not give us
36093 a chance to make a mistake.
36095 Even if the problem you experience is a fatal signal, you should still
36096 say so explicitly. Suppose something strange is going on, such as, your
36097 copy of @value{GDBN} is out of synch, or you have encountered a bug in
36098 the C library on your system. (This has happened!) Your copy might
36099 crash and ours would not. If you told us to expect a crash, then when
36100 ours fails to crash, we would know that the bug was not happening for
36101 us. If you had not told us to expect a crash, then we would not be able
36102 to draw any conclusion from our observations.
36105 @cindex recording a session script
36106 To collect all this information, you can use a session recording program
36107 such as @command{script}, which is available on many Unix systems.
36108 Just run your @value{GDBN} session inside @command{script} and then
36109 include the @file{typescript} file with your bug report.
36111 Another way to record a @value{GDBN} session is to run @value{GDBN}
36112 inside Emacs and then save the entire buffer to a file.
36115 If you wish to suggest changes to the @value{GDBN} source, send us context
36116 diffs. If you even discuss something in the @value{GDBN} source, refer to
36117 it by context, not by line number.
36119 The line numbers in our development sources will not match those in your
36120 sources. Your line numbers would convey no useful information to us.
36124 Here are some things that are not necessary:
36128 A description of the envelope of the bug.
36130 Often people who encounter a bug spend a lot of time investigating
36131 which changes to the input file will make the bug go away and which
36132 changes will not affect it.
36134 This is often time consuming and not very useful, because the way we
36135 will find the bug is by running a single example under the debugger
36136 with breakpoints, not by pure deduction from a series of examples.
36137 We recommend that you save your time for something else.
36139 Of course, if you can find a simpler example to report @emph{instead}
36140 of the original one, that is a convenience for us. Errors in the
36141 output will be easier to spot, running under the debugger will take
36142 less time, and so on.
36144 However, simplification is not vital; if you do not want to do this,
36145 report the bug anyway and send us the entire test case you used.
36148 A patch for the bug.
36150 A patch for the bug does help us if it is a good one. But do not omit
36151 the necessary information, such as the test case, on the assumption that
36152 a patch is all we need. We might see problems with your patch and decide
36153 to fix the problem another way, or we might not understand it at all.
36155 Sometimes with a program as complicated as @value{GDBN} it is very hard to
36156 construct an example that will make the program follow a certain path
36157 through the code. If you do not send us the example, we will not be able
36158 to construct one, so we will not be able to verify that the bug is fixed.
36160 And if we cannot understand what bug you are trying to fix, or why your
36161 patch should be an improvement, we will not install it. A test case will
36162 help us to understand.
36165 A guess about what the bug is or what it depends on.
36167 Such guesses are usually wrong. Even we cannot guess right about such
36168 things without first using the debugger to find the facts.
36171 @c The readline documentation is distributed with the readline code
36172 @c and consists of the two following files:
36175 @c Use -I with makeinfo to point to the appropriate directory,
36176 @c environment var TEXINPUTS with TeX.
36177 @ifclear SYSTEM_READLINE
36178 @include rluser.texi
36179 @include hsuser.texi
36183 @appendix In Memoriam
36185 The @value{GDBN} project mourns the loss of the following long-time
36190 Fred was a long-standing contributor to @value{GDBN} (1991-2006), and
36191 to Free Software in general. Outside of @value{GDBN}, he was known in
36192 the Amiga world for his series of Fish Disks, and the GeekGadget project.
36194 @item Michael Snyder
36195 Michael was one of the Global Maintainers of the @value{GDBN} project,
36196 with contributions recorded as early as 1996, until 2011. In addition
36197 to his day to day participation, he was a large driving force behind
36198 adding Reverse Debugging to @value{GDBN}.
36201 Beyond their technical contributions to the project, they were also
36202 enjoyable members of the Free Software Community. We will miss them.
36204 @node Formatting Documentation
36205 @appendix Formatting Documentation
36207 @cindex @value{GDBN} reference card
36208 @cindex reference card
36209 The @value{GDBN} 4 release includes an already-formatted reference card, ready
36210 for printing with PostScript or Ghostscript, in the @file{gdb}
36211 subdirectory of the main source directory@footnote{In
36212 @file{gdb-@value{GDBVN}/gdb/refcard.ps} of the version @value{GDBVN}
36213 release.}. If you can use PostScript or Ghostscript with your printer,
36214 you can print the reference card immediately with @file{refcard.ps}.
36216 The release also includes the source for the reference card. You
36217 can format it, using @TeX{}, by typing:
36223 The @value{GDBN} reference card is designed to print in @dfn{landscape}
36224 mode on US ``letter'' size paper;
36225 that is, on a sheet 11 inches wide by 8.5 inches
36226 high. You will need to specify this form of printing as an option to
36227 your @sc{dvi} output program.
36229 @cindex documentation
36231 All the documentation for @value{GDBN} comes as part of the machine-readable
36232 distribution. The documentation is written in Texinfo format, which is
36233 a documentation system that uses a single source file to produce both
36234 on-line information and a printed manual. You can use one of the Info
36235 formatting commands to create the on-line version of the documentation
36236 and @TeX{} (or @code{texi2roff}) to typeset the printed version.
36238 @value{GDBN} includes an already formatted copy of the on-line Info
36239 version of this manual in the @file{gdb} subdirectory. The main Info
36240 file is @file{gdb-@value{GDBVN}/gdb/gdb.info}, and it refers to
36241 subordinate files matching @samp{gdb.info*} in the same directory. If
36242 necessary, you can print out these files, or read them with any editor;
36243 but they are easier to read using the @code{info} subsystem in @sc{gnu}
36244 Emacs or the standalone @code{info} program, available as part of the
36245 @sc{gnu} Texinfo distribution.
36247 If you want to format these Info files yourself, you need one of the
36248 Info formatting programs, such as @code{texinfo-format-buffer} or
36251 If you have @code{makeinfo} installed, and are in the top level
36252 @value{GDBN} source directory (@file{gdb-@value{GDBVN}}, in the case of
36253 version @value{GDBVN}), you can make the Info file by typing:
36260 If you want to typeset and print copies of this manual, you need @TeX{},
36261 a program to print its @sc{dvi} output files, and @file{texinfo.tex}, the
36262 Texinfo definitions file.
36264 @TeX{} is a typesetting program; it does not print files directly, but
36265 produces output files called @sc{dvi} files. To print a typeset
36266 document, you need a program to print @sc{dvi} files. If your system
36267 has @TeX{} installed, chances are it has such a program. The precise
36268 command to use depends on your system; @kbd{lpr -d} is common; another
36269 (for PostScript devices) is @kbd{dvips}. The @sc{dvi} print command may
36270 require a file name without any extension or a @samp{.dvi} extension.
36272 @TeX{} also requires a macro definitions file called
36273 @file{texinfo.tex}. This file tells @TeX{} how to typeset a document
36274 written in Texinfo format. On its own, @TeX{} cannot either read or
36275 typeset a Texinfo file. @file{texinfo.tex} is distributed with GDB
36276 and is located in the @file{gdb-@var{version-number}/texinfo}
36279 If you have @TeX{} and a @sc{dvi} printer program installed, you can
36280 typeset and print this manual. First switch to the @file{gdb}
36281 subdirectory of the main source directory (for example, to
36282 @file{gdb-@value{GDBVN}/gdb}) and type:
36288 Then give @file{gdb.dvi} to your @sc{dvi} printing program.
36290 @node Installing GDB
36291 @appendix Installing @value{GDBN}
36292 @cindex installation
36295 * Requirements:: Requirements for building @value{GDBN}
36296 * Running Configure:: Invoking the @value{GDBN} @file{configure} script
36297 * Separate Objdir:: Compiling @value{GDBN} in another directory
36298 * Config Names:: Specifying names for hosts and targets
36299 * Configure Options:: Summary of options for configure
36300 * System-wide configuration:: Having a system-wide init file
36304 @section Requirements for Building @value{GDBN}
36305 @cindex building @value{GDBN}, requirements for
36307 Building @value{GDBN} requires various tools and packages to be available.
36308 Other packages will be used only if they are found.
36310 @heading Tools/Packages Necessary for Building @value{GDBN}
36312 @item C@t{++}11 compiler
36313 @value{GDBN} is written in C@t{++}11. It should be buildable with any
36314 recent C@t{++}11 compiler, e.g.@: GCC.
36317 @value{GDBN}'s build system relies on features only found in the GNU
36318 make program. Other variants of @code{make} will not work.
36321 @heading Tools/Packages Optional for Building @value{GDBN}
36325 @value{GDBN} can use the Expat XML parsing library. This library may be
36326 included with your operating system distribution; if it is not, you
36327 can get the latest version from @url{http://expat.sourceforge.net}.
36328 The @file{configure} script will search for this library in several
36329 standard locations; if it is installed in an unusual path, you can
36330 use the @option{--with-libexpat-prefix} option to specify its location.
36336 Remote protocol memory maps (@pxref{Memory Map Format})
36338 Target descriptions (@pxref{Target Descriptions})
36340 Remote shared library lists (@xref{Library List Format},
36341 or alternatively @pxref{Library List Format for SVR4 Targets})
36343 MS-Windows shared libraries (@pxref{Shared Libraries})
36345 Traceframe info (@pxref{Traceframe Info Format})
36347 Branch trace (@pxref{Branch Trace Format},
36348 @pxref{Branch Trace Configuration Format})
36352 @value{GDBN} can be scripted using GNU Guile. @xref{Guile}. By
36353 default, @value{GDBN} will be compiled if the Guile libraries are
36354 installed and are found by @file{configure}. You can use the
36355 @code{--with-guile} option to request Guile, and pass either the Guile
36356 version number or the file name of the relevant @code{pkg-config}
36357 program to choose a particular version of Guile.
36360 @value{GDBN}'s features related to character sets (@pxref{Character
36361 Sets}) require a functioning @code{iconv} implementation. If you are
36362 on a GNU system, then this is provided by the GNU C Library. Some
36363 other systems also provide a working @code{iconv}.
36365 If @value{GDBN} is using the @code{iconv} program which is installed
36366 in a non-standard place, you will need to tell @value{GDBN} where to
36367 find it. This is done with @option{--with-iconv-bin} which specifies
36368 the directory that contains the @code{iconv} program. This program is
36369 run in order to make a list of the available character sets.
36371 On systems without @code{iconv}, you can install GNU Libiconv. If
36372 Libiconv is installed in a standard place, @value{GDBN} will
36373 automatically use it if it is needed. If you have previously
36374 installed Libiconv in a non-standard place, you can use the
36375 @option{--with-libiconv-prefix} option to @file{configure}.
36377 @value{GDBN}'s top-level @file{configure} and @file{Makefile} will
36378 arrange to build Libiconv if a directory named @file{libiconv} appears
36379 in the top-most source directory. If Libiconv is built this way, and
36380 if the operating system does not provide a suitable @code{iconv}
36381 implementation, then the just-built library will automatically be used
36382 by @value{GDBN}. One easy way to set this up is to download GNU
36383 Libiconv, unpack it inside the top-level directory of the @value{GDBN}
36384 source tree, and then rename the directory holding the Libiconv source
36385 code to @samp{libiconv}.
36388 @value{GDBN} can support debugging sections that are compressed with
36389 the LZMA library. @xref{MiniDebugInfo}. If this library is not
36390 included with your operating system, you can find it in the xz package
36391 at @url{http://tukaani.org/xz/}. If the LZMA library is available in
36392 the usual place, then the @file{configure} script will use it
36393 automatically. If it is installed in an unusual path, you can use the
36394 @option{--with-lzma-prefix} option to specify its location.
36398 @value{GDBN} can use the GNU MPFR multiple-precision floating-point
36399 library. This library may be included with your operating system
36400 distribution; if it is not, you can get the latest version from
36401 @url{http://www.mpfr.org}. The @file{configure} script will search
36402 for this library in several standard locations; if it is installed
36403 in an unusual path, you can use the @option{--with-libmpfr-prefix}
36404 option to specify its location.
36406 GNU MPFR is used to emulate target floating-point arithmetic during
36407 expression evaluation when the target uses different floating-point
36408 formats than the host. If GNU MPFR it is not available, @value{GDBN}
36409 will fall back to using host floating-point arithmetic.
36412 @value{GDBN} can be scripted using Python language. @xref{Python}.
36413 By default, @value{GDBN} will be compiled if the Python libraries are
36414 installed and are found by @file{configure}. You can use the
36415 @code{--with-python} option to request Python, and pass either the
36416 file name of the relevant @code{python} executable, or the name of the
36417 directory in which Python is installed, to choose a particular
36418 installation of Python.
36421 @cindex compressed debug sections
36422 @value{GDBN} will use the @samp{zlib} library, if available, to read
36423 compressed debug sections. Some linkers, such as GNU gold, are capable
36424 of producing binaries with compressed debug sections. If @value{GDBN}
36425 is compiled with @samp{zlib}, it will be able to read the debug
36426 information in such binaries.
36428 The @samp{zlib} library is likely included with your operating system
36429 distribution; if it is not, you can get the latest version from
36430 @url{http://zlib.net}.
36433 @node Running Configure
36434 @section Invoking the @value{GDBN} @file{configure} Script
36435 @cindex configuring @value{GDBN}
36436 @value{GDBN} comes with a @file{configure} script that automates the process
36437 of preparing @value{GDBN} for installation; you can then use @code{make} to
36438 build the @code{gdb} program.
36440 @c irrelevant in info file; it's as current as the code it lives with.
36441 @footnote{If you have a more recent version of @value{GDBN} than @value{GDBVN},
36442 look at the @file{README} file in the sources; we may have improved the
36443 installation procedures since publishing this manual.}
36446 The @value{GDBN} distribution includes all the source code you need for
36447 @value{GDBN} in a single directory, whose name is usually composed by
36448 appending the version number to @samp{gdb}.
36450 For example, the @value{GDBN} version @value{GDBVN} distribution is in the
36451 @file{gdb-@value{GDBVN}} directory. That directory contains:
36454 @item gdb-@value{GDBVN}/configure @r{(and supporting files)}
36455 script for configuring @value{GDBN} and all its supporting libraries
36457 @item gdb-@value{GDBVN}/gdb
36458 the source specific to @value{GDBN} itself
36460 @item gdb-@value{GDBVN}/bfd
36461 source for the Binary File Descriptor library
36463 @item gdb-@value{GDBVN}/include
36464 @sc{gnu} include files
36466 @item gdb-@value{GDBVN}/libiberty
36467 source for the @samp{-liberty} free software library
36469 @item gdb-@value{GDBVN}/opcodes
36470 source for the library of opcode tables and disassemblers
36472 @item gdb-@value{GDBVN}/readline
36473 source for the @sc{gnu} command-line interface
36476 There may be other subdirectories as well.
36478 The simplest way to configure and build @value{GDBN} is to run @file{configure}
36479 from the @file{gdb-@var{version-number}} source directory, which in
36480 this example is the @file{gdb-@value{GDBVN}} directory.
36482 First switch to the @file{gdb-@var{version-number}} source directory
36483 if you are not already in it; then run @file{configure}. Pass the
36484 identifier for the platform on which @value{GDBN} will run as an
36490 cd gdb-@value{GDBVN}
36495 Running @samp{configure} and then running @code{make} builds the
36496 included supporting libraries, then @code{gdb} itself. The configured
36497 source files, and the binaries, are left in the corresponding source
36501 @file{configure} is a Bourne-shell (@code{/bin/sh}) script; if your
36502 system does not recognize this automatically when you run a different
36503 shell, you may need to run @code{sh} on it explicitly:
36509 You should run the @file{configure} script from the top directory in the
36510 source tree, the @file{gdb-@var{version-number}} directory. If you run
36511 @file{configure} from one of the subdirectories, you will configure only
36512 that subdirectory. That is usually not what you want. In particular,
36513 if you run the first @file{configure} from the @file{gdb} subdirectory
36514 of the @file{gdb-@var{version-number}} directory, you will omit the
36515 configuration of @file{bfd}, @file{readline}, and other sibling
36516 directories of the @file{gdb} subdirectory. This leads to build errors
36517 about missing include files such as @file{bfd/bfd.h}.
36519 You can install @code{@value{GDBN}} anywhere. The best way to do this
36520 is to pass the @code{--prefix} option to @code{configure}, and then
36521 install it with @code{make install}.
36523 @node Separate Objdir
36524 @section Compiling @value{GDBN} in Another Directory
36526 If you want to run @value{GDBN} versions for several host or target machines,
36527 you need a different @code{gdb} compiled for each combination of
36528 host and target. @file{configure} is designed to make this easy by
36529 allowing you to generate each configuration in a separate subdirectory,
36530 rather than in the source directory. If your @code{make} program
36531 handles the @samp{VPATH} feature (@sc{gnu} @code{make} does), running
36532 @code{make} in each of these directories builds the @code{gdb}
36533 program specified there.
36535 To build @code{gdb} in a separate directory, run @file{configure}
36536 with the @samp{--srcdir} option to specify where to find the source.
36537 (You also need to specify a path to find @file{configure}
36538 itself from your working directory. If the path to @file{configure}
36539 would be the same as the argument to @samp{--srcdir}, you can leave out
36540 the @samp{--srcdir} option; it is assumed.)
36542 For example, with version @value{GDBVN}, you can build @value{GDBN} in a
36543 separate directory for a Sun 4 like this:
36547 cd gdb-@value{GDBVN}
36550 ../gdb-@value{GDBVN}/configure
36555 When @file{configure} builds a configuration using a remote source
36556 directory, it creates a tree for the binaries with the same structure
36557 (and using the same names) as the tree under the source directory. In
36558 the example, you'd find the Sun 4 library @file{libiberty.a} in the
36559 directory @file{gdb-sun4/libiberty}, and @value{GDBN} itself in
36560 @file{gdb-sun4/gdb}.
36562 Make sure that your path to the @file{configure} script has just one
36563 instance of @file{gdb} in it. If your path to @file{configure} looks
36564 like @file{../gdb-@value{GDBVN}/gdb/configure}, you are configuring only
36565 one subdirectory of @value{GDBN}, not the whole package. This leads to
36566 build errors about missing include files such as @file{bfd/bfd.h}.
36568 One popular reason to build several @value{GDBN} configurations in separate
36569 directories is to configure @value{GDBN} for cross-compiling (where
36570 @value{GDBN} runs on one machine---the @dfn{host}---while debugging
36571 programs that run on another machine---the @dfn{target}).
36572 You specify a cross-debugging target by
36573 giving the @samp{--target=@var{target}} option to @file{configure}.
36575 When you run @code{make} to build a program or library, you must run
36576 it in a configured directory---whatever directory you were in when you
36577 called @file{configure} (or one of its subdirectories).
36579 The @code{Makefile} that @file{configure} generates in each source
36580 directory also runs recursively. If you type @code{make} in a source
36581 directory such as @file{gdb-@value{GDBVN}} (or in a separate configured
36582 directory configured with @samp{--srcdir=@var{dirname}/gdb-@value{GDBVN}}), you
36583 will build all the required libraries, and then build GDB.
36585 When you have multiple hosts or targets configured in separate
36586 directories, you can run @code{make} on them in parallel (for example,
36587 if they are NFS-mounted on each of the hosts); they will not interfere
36591 @section Specifying Names for Hosts and Targets
36593 The specifications used for hosts and targets in the @file{configure}
36594 script are based on a three-part naming scheme, but some short predefined
36595 aliases are also supported. The full naming scheme encodes three pieces
36596 of information in the following pattern:
36599 @var{architecture}-@var{vendor}-@var{os}
36602 For example, you can use the alias @code{sun4} as a @var{host} argument,
36603 or as the value for @var{target} in a @code{--target=@var{target}}
36604 option. The equivalent full name is @samp{sparc-sun-sunos4}.
36606 The @file{configure} script accompanying @value{GDBN} does not provide
36607 any query facility to list all supported host and target names or
36608 aliases. @file{configure} calls the Bourne shell script
36609 @code{config.sub} to map abbreviations to full names; you can read the
36610 script, if you wish, or you can use it to test your guesses on
36611 abbreviations---for example:
36614 % sh config.sub i386-linux
36616 % sh config.sub alpha-linux
36617 alpha-unknown-linux-gnu
36618 % sh config.sub hp9k700
36620 % sh config.sub sun4
36621 sparc-sun-sunos4.1.1
36622 % sh config.sub sun3
36623 m68k-sun-sunos4.1.1
36624 % sh config.sub i986v
36625 Invalid configuration `i986v': machine `i986v' not recognized
36629 @code{config.sub} is also distributed in the @value{GDBN} source
36630 directory (@file{gdb-@value{GDBVN}}, for version @value{GDBVN}).
36632 @node Configure Options
36633 @section @file{configure} Options
36635 Here is a summary of the @file{configure} options and arguments that
36636 are most often useful for building @value{GDBN}. @file{configure}
36637 also has several other options not listed here. @inforef{Running
36638 configure scripts,,autoconf.info}, for a full
36639 explanation of @file{configure}.
36642 configure @r{[}--help@r{]}
36643 @r{[}--prefix=@var{dir}@r{]}
36644 @r{[}--exec-prefix=@var{dir}@r{]}
36645 @r{[}--srcdir=@var{dirname}@r{]}
36646 @r{[}--target=@var{target}@r{]}
36650 You may introduce options with a single @samp{-} rather than
36651 @samp{--} if you prefer; but you may abbreviate option names if you use
36656 Display a quick summary of how to invoke @file{configure}.
36658 @item --prefix=@var{dir}
36659 Configure the source to install programs and files under directory
36662 @item --exec-prefix=@var{dir}
36663 Configure the source to install programs under directory
36666 @c avoid splitting the warning from the explanation:
36668 @item --srcdir=@var{dirname}
36669 Use this option to make configurations in directories separate from the
36670 @value{GDBN} source directories. Among other things, you can use this to
36671 build (or maintain) several configurations simultaneously, in separate
36672 directories. @file{configure} writes configuration-specific files in
36673 the current directory, but arranges for them to use the source in the
36674 directory @var{dirname}. @file{configure} creates directories under
36675 the working directory in parallel to the source directories below
36678 @item --target=@var{target}
36679 Configure @value{GDBN} for cross-debugging programs running on the specified
36680 @var{target}. Without this option, @value{GDBN} is configured to debug
36681 programs that run on the same machine (@var{host}) as @value{GDBN} itself.
36683 There is no convenient way to generate a list of all available
36684 targets. Also see the @code{--enable-targets} option, below.
36687 There are many other options that are specific to @value{GDBN}. This
36688 lists just the most common ones; there are some very specialized
36689 options not described here.
36692 @item --enable-targets=@r{[}@var{target}@r{]}@dots{}
36693 @itemx --enable-targets=all
36694 Configure @value{GDBN} for cross-debugging programs running on the
36695 specified list of targets. The special value @samp{all} configures
36696 @value{GDBN} for debugging programs running on any target it supports.
36698 @item --with-gdb-datadir=@var{path}
36699 Set the @value{GDBN}-specific data directory. @value{GDBN} will look
36700 here for certain supporting files or scripts. This defaults to the
36701 @file{gdb} subdirectory of @samp{datadi} (which can be set using
36704 @item --with-relocated-sources=@var{dir}
36705 Sets up the default source path substitution rule so that directory
36706 names recorded in debug information will be automatically adjusted for
36707 any directory under @var{dir}. @var{dir} should be a subdirectory of
36708 @value{GDBN}'s configured prefix, the one mentioned in the
36709 @code{--prefix} or @code{--exec-prefix} options to configure. This
36710 option is useful if GDB is supposed to be moved to a different place
36713 @item --enable-64-bit-bfd
36714 Enable 64-bit support in BFD on 32-bit hosts.
36716 @item --disable-gdbmi
36717 Build @value{GDBN} without the GDB/MI machine interface
36721 Build @value{GDBN} with the text-mode full-screen user interface
36722 (TUI). Requires a curses library (ncurses and cursesX are also
36725 @item --with-curses
36726 Use the curses library instead of the termcap library, for text-mode
36727 terminal operations.
36729 @item --with-libunwind-ia64
36730 Use the libunwind library for unwinding function call stack on ia64
36731 target platforms. See http://www.nongnu.org/libunwind/index.html for
36734 @item --with-system-readline
36735 Use the readline library installed on the host, rather than the
36736 library supplied as part of @value{GDBN}.
36738 @item --with-system-zlib
36739 Use the zlib library installed on the host, rather than the library
36740 supplied as part of @value{GDBN}.
36743 Build @value{GDBN} with Expat, a library for XML parsing. (Done by
36744 default if libexpat is installed and found at configure time.) This
36745 library is used to read XML files supplied with @value{GDBN}. If it
36746 is unavailable, some features, such as remote protocol memory maps,
36747 target descriptions, and shared library lists, that are based on XML
36748 files, will not be available in @value{GDBN}. If your host does not
36749 have libexpat installed, you can get the latest version from
36750 `http://expat.sourceforge.net'.
36752 @item --with-libiconv-prefix@r{[}=@var{dir}@r{]}
36754 Build @value{GDBN} with GNU libiconv, a character set encoding
36755 conversion library. This is not done by default, as on GNU systems
36756 the @code{iconv} that is built in to the C library is sufficient. If
36757 your host does not have a working @code{iconv}, you can get the latest
36758 version of GNU iconv from `https://www.gnu.org/software/libiconv/'.
36760 @value{GDBN}'s build system also supports building GNU libiconv as
36761 part of the overall build. @xref{Requirements}.
36764 Build @value{GDBN} with LZMA, a compression library. (Done by default
36765 if liblzma is installed and found at configure time.) LZMA is used by
36766 @value{GDBN}'s "mini debuginfo" feature, which is only useful on
36767 platforms using the ELF object file format. If your host does not
36768 have liblzma installed, you can get the latest version from
36769 `https://tukaani.org/xz/'.
36772 Build @value{GDBN} with GNU MPFR, a library for multiple-precision
36773 floating-point computation with correct rounding. (Done by default if
36774 GNU MPFR is installed and found at configure time.) This library is
36775 used to emulate target floating-point arithmetic during expression
36776 evaluation when the target uses different floating-point formats than
36777 the host. If GNU MPFR is not available, @value{GDBN} will fall back
36778 to using host floating-point arithmetic. If your host does not have
36779 GNU MPFR installed, you can get the latest version from
36780 `http://www.mpfr.org'.
36782 @item --with-python@r{[}=@var{python}@r{]}
36783 Build @value{GDBN} with Python scripting support. (Done by default if
36784 libpython is present and found at configure time.) Python makes
36785 @value{GDBN} scripting much more powerful than the restricted CLI
36786 scripting language. If your host does not have Python installed, you
36787 can find it on `http://www.python.org/download/'. The oldest version
36788 of Python supported by GDB is 2.6. The optional argument @var{python}
36789 is used to find the Python headers and libraries. It can be either
36790 the name of a Python executable, or the name of the directory in which
36791 Python is installed.
36793 @item --with-guile[=GUILE]'
36794 Build @value{GDBN} with GNU Guile scripting support. (Done by default
36795 if libguile is present and found at configure time.) If your host
36796 does not have Guile installed, you can find it at
36797 `https://www.gnu.org/software/guile/'. The optional argument GUILE
36798 can be a version number, which will cause @code{configure} to try to
36799 use that version of Guile; or the file name of a @code{pkg-config}
36800 executable, which will be queried to find the information needed to
36801 compile and link against Guile.
36803 @item --without-included-regex
36804 Don't use the regex library included with @value{GDBN} (as part of the
36805 libiberty library). This is the default on hosts with version 2 of
36808 @item --with-sysroot=@var{dir}
36809 Use @var{dir} as the default system root directory for libraries whose
36810 file names begin with @file{/lib}' or @file{/usr/lib'}. (The value of
36811 @var{dir} can be modified at run time by using the @command{set
36812 sysroot} command.) If @var{dir} is under the @value{GDBN} configured
36813 prefix (set with @code{--prefix} or @code{--exec-prefix options}, the
36814 default system root will be automatically adjusted if and when
36815 @value{GDBN} is moved to a different location.
36817 @item --with-system-gdbinit=@var{file}
36818 Configure @value{GDBN} to automatically load a system-wide init file.
36819 @var{file} should be an absolute file name. If @var{file} is in a
36820 directory under the configured prefix, and @value{GDBN} is moved to
36821 another location after being built, the location of the system-wide
36822 init file will be adjusted accordingly.
36824 @item --enable-build-warnings
36825 When building the @value{GDBN} sources, ask the compiler to warn about
36826 any code which looks even vaguely suspicious. It passes many
36827 different warning flags, depending on the exact version of the
36828 compiler you are using.
36830 @item --enable-werror
36831 Treat compiler warnings as werrors. It adds the @code{-Werror} flag
36832 to the compiler, which will fail the compilation if the compiler
36833 outputs any warning messages.
36835 @item --enable-ubsan
36836 Enable the GCC undefined behavior sanitizer. This is disabled by
36837 default, but passing @code{--enable-ubsan=yes} or
36838 @code{--enable-ubsan=auto} to @code{configure} will enable it. The
36839 undefined behavior sanitizer checks for C@t{++} undefined behavior.
36840 It has a performance cost, so if you are looking at @value{GDBN}'s
36841 performance, you should disable it. The undefined behavior sanitizer
36842 was first introduced in GCC 4.9.
36845 @node System-wide configuration
36846 @section System-wide configuration and settings
36847 @cindex system-wide init file
36849 @value{GDBN} can be configured to have a system-wide init file;
36850 this file will be read and executed at startup (@pxref{Startup, , What
36851 @value{GDBN} does during startup}).
36853 Here is the corresponding configure option:
36856 @item --with-system-gdbinit=@var{file}
36857 Specify that the default location of the system-wide init file is
36861 If @value{GDBN} has been configured with the option @option{--prefix=$prefix},
36862 it may be subject to relocation. Two possible cases:
36866 If the default location of this init file contains @file{$prefix},
36867 it will be subject to relocation. Suppose that the configure options
36868 are @option{--prefix=$prefix --with-system-gdbinit=$prefix/etc/gdbinit};
36869 if @value{GDBN} is moved from @file{$prefix} to @file{$install}, the system
36870 init file is looked for as @file{$install/etc/gdbinit} instead of
36871 @file{$prefix/etc/gdbinit}.
36874 By contrast, if the default location does not contain the prefix,
36875 it will not be relocated. E.g.@: if @value{GDBN} has been configured with
36876 @option{--prefix=/usr/local --with-system-gdbinit=/usr/share/gdb/gdbinit},
36877 then @value{GDBN} will always look for @file{/usr/share/gdb/gdbinit},
36878 wherever @value{GDBN} is installed.
36881 If the configured location of the system-wide init file (as given by the
36882 @option{--with-system-gdbinit} option at configure time) is in the
36883 data-directory (as specified by @option{--with-gdb-datadir} at configure
36884 time) or in one of its subdirectories, then @value{GDBN} will look for the
36885 system-wide init file in the directory specified by the
36886 @option{--data-directory} command-line option.
36887 Note that the system-wide init file is only read once, during @value{GDBN}
36888 initialization. If the data-directory is changed after @value{GDBN} has
36889 started with the @code{set data-directory} command, the file will not be
36893 * System-wide Configuration Scripts:: Installed System-wide Configuration Scripts
36896 @node System-wide Configuration Scripts
36897 @subsection Installed System-wide Configuration Scripts
36898 @cindex system-wide configuration scripts
36900 The @file{system-gdbinit} directory, located inside the data-directory
36901 (as specified by @option{--with-gdb-datadir} at configure time) contains
36902 a number of scripts which can be used as system-wide init files. To
36903 automatically source those scripts at startup, @value{GDBN} should be
36904 configured with @option{--with-system-gdbinit}. Otherwise, any user
36905 should be able to source them by hand as needed.
36907 The following scripts are currently available:
36910 @item @file{elinos.py}
36912 @cindex ELinOS system-wide configuration script
36913 This script is useful when debugging a program on an ELinOS target.
36914 It takes advantage of the environment variables defined in a standard
36915 ELinOS environment in order to determine the location of the system
36916 shared libraries, and then sets the @samp{solib-absolute-prefix}
36917 and @samp{solib-search-path} variables appropriately.
36919 @item @file{wrs-linux.py}
36920 @pindex wrs-linux.py
36921 @cindex Wind River Linux system-wide configuration script
36922 This script is useful when debugging a program on a target running
36923 Wind River Linux. It expects the @env{ENV_PREFIX} to be set to
36924 the host-side sysroot used by the target system.
36928 @node Maintenance Commands
36929 @appendix Maintenance Commands
36930 @cindex maintenance commands
36931 @cindex internal commands
36933 In addition to commands intended for @value{GDBN} users, @value{GDBN}
36934 includes a number of commands intended for @value{GDBN} developers,
36935 that are not documented elsewhere in this manual. These commands are
36936 provided here for reference. (For commands that turn on debugging
36937 messages, see @ref{Debugging Output}.)
36940 @kindex maint agent
36941 @kindex maint agent-eval
36942 @item maint agent @r{[}-at @var{location}@r{,}@r{]} @var{expression}
36943 @itemx maint agent-eval @r{[}-at @var{location}@r{,}@r{]} @var{expression}
36944 Translate the given @var{expression} into remote agent bytecodes.
36945 This command is useful for debugging the Agent Expression mechanism
36946 (@pxref{Agent Expressions}). The @samp{agent} version produces an
36947 expression useful for data collection, such as by tracepoints, while
36948 @samp{maint agent-eval} produces an expression that evaluates directly
36949 to a result. For instance, a collection expression for @code{globa +
36950 globb} will include bytecodes to record four bytes of memory at each
36951 of the addresses of @code{globa} and @code{globb}, while discarding
36952 the result of the addition, while an evaluation expression will do the
36953 addition and return the sum.
36954 If @code{-at} is given, generate remote agent bytecode for @var{location}.
36955 If not, generate remote agent bytecode for current frame PC address.
36957 @kindex maint agent-printf
36958 @item maint agent-printf @var{format},@var{expr},...
36959 Translate the given format string and list of argument expressions
36960 into remote agent bytecodes and display them as a disassembled list.
36961 This command is useful for debugging the agent version of dynamic
36962 printf (@pxref{Dynamic Printf}).
36964 @kindex maint info breakpoints
36965 @item @anchor{maint info breakpoints}maint info breakpoints
36966 Using the same format as @samp{info breakpoints}, display both the
36967 breakpoints you've set explicitly, and those @value{GDBN} is using for
36968 internal purposes. Internal breakpoints are shown with negative
36969 breakpoint numbers. The type column identifies what kind of breakpoint
36974 Normal, explicitly set breakpoint.
36977 Normal, explicitly set watchpoint.
36980 Internal breakpoint, used to handle correctly stepping through
36981 @code{longjmp} calls.
36983 @item longjmp resume
36984 Internal breakpoint at the target of a @code{longjmp}.
36987 Temporary internal breakpoint used by the @value{GDBN} @code{until} command.
36990 Temporary internal breakpoint used by the @value{GDBN} @code{finish} command.
36993 Shared library events.
36997 @kindex maint info btrace
36998 @item maint info btrace
36999 Pint information about raw branch tracing data.
37001 @kindex maint btrace packet-history
37002 @item maint btrace packet-history
37003 Print the raw branch trace packets that are used to compute the
37004 execution history for the @samp{record btrace} command. Both the
37005 information and the format in which it is printed depend on the btrace
37010 For the BTS recording format, print a list of blocks of sequential
37011 code. For each block, the following information is printed:
37015 Newer blocks have higher numbers. The oldest block has number zero.
37016 @item Lowest @samp{PC}
37017 @item Highest @samp{PC}
37021 For the Intel Processor Trace recording format, print a list of
37022 Intel Processor Trace packets. For each packet, the following
37023 information is printed:
37026 @item Packet number
37027 Newer packets have higher numbers. The oldest packet has number zero.
37029 The packet's offset in the trace stream.
37030 @item Packet opcode and payload
37034 @kindex maint btrace clear-packet-history
37035 @item maint btrace clear-packet-history
37036 Discards the cached packet history printed by the @samp{maint btrace
37037 packet-history} command. The history will be computed again when
37040 @kindex maint btrace clear
37041 @item maint btrace clear
37042 Discard the branch trace data. The data will be fetched anew and the
37043 branch trace will be recomputed when needed.
37045 This implicitly truncates the branch trace to a single branch trace
37046 buffer. When updating branch trace incrementally, the branch trace
37047 available to @value{GDBN} may be bigger than a single branch trace
37050 @kindex maint set btrace pt skip-pad
37051 @item maint set btrace pt skip-pad
37052 @kindex maint show btrace pt skip-pad
37053 @item maint show btrace pt skip-pad
37054 Control whether @value{GDBN} will skip PAD packets when computing the
37057 @kindex set displaced-stepping
37058 @kindex show displaced-stepping
37059 @cindex displaced stepping support
37060 @cindex out-of-line single-stepping
37061 @item set displaced-stepping
37062 @itemx show displaced-stepping
37063 Control whether or not @value{GDBN} will do @dfn{displaced stepping}
37064 if the target supports it. Displaced stepping is a way to single-step
37065 over breakpoints without removing them from the inferior, by executing
37066 an out-of-line copy of the instruction that was originally at the
37067 breakpoint location. It is also known as out-of-line single-stepping.
37070 @item set displaced-stepping on
37071 If the target architecture supports it, @value{GDBN} will use
37072 displaced stepping to step over breakpoints.
37074 @item set displaced-stepping off
37075 @value{GDBN} will not use displaced stepping to step over breakpoints,
37076 even if such is supported by the target architecture.
37078 @cindex non-stop mode, and @samp{set displaced-stepping}
37079 @item set displaced-stepping auto
37080 This is the default mode. @value{GDBN} will use displaced stepping
37081 only if non-stop mode is active (@pxref{Non-Stop Mode}) and the target
37082 architecture supports displaced stepping.
37085 @kindex maint check-psymtabs
37086 @item maint check-psymtabs
37087 Check the consistency of currently expanded psymtabs versus symtabs.
37088 Use this to check, for example, whether a symbol is in one but not the other.
37090 @kindex maint check-symtabs
37091 @item maint check-symtabs
37092 Check the consistency of currently expanded symtabs.
37094 @kindex maint expand-symtabs
37095 @item maint expand-symtabs [@var{regexp}]
37096 Expand symbol tables.
37097 If @var{regexp} is specified, only expand symbol tables for file
37098 names matching @var{regexp}.
37100 @kindex maint set catch-demangler-crashes
37101 @kindex maint show catch-demangler-crashes
37102 @cindex demangler crashes
37103 @item maint set catch-demangler-crashes [on|off]
37104 @itemx maint show catch-demangler-crashes
37105 Control whether @value{GDBN} should attempt to catch crashes in the
37106 symbol name demangler. The default is to attempt to catch crashes.
37107 If enabled, the first time a crash is caught, a core file is created,
37108 the offending symbol is displayed and the user is presented with the
37109 option to terminate the current session.
37111 @kindex maint cplus first_component
37112 @item maint cplus first_component @var{name}
37113 Print the first C@t{++} class/namespace component of @var{name}.
37115 @kindex maint cplus namespace
37116 @item maint cplus namespace
37117 Print the list of possible C@t{++} namespaces.
37119 @kindex maint deprecate
37120 @kindex maint undeprecate
37121 @cindex deprecated commands
37122 @item maint deprecate @var{command} @r{[}@var{replacement}@r{]}
37123 @itemx maint undeprecate @var{command}
37124 Deprecate or undeprecate the named @var{command}. Deprecated commands
37125 cause @value{GDBN} to issue a warning when you use them. The optional
37126 argument @var{replacement} says which newer command should be used in
37127 favor of the deprecated one; if it is given, @value{GDBN} will mention
37128 the replacement as part of the warning.
37130 @kindex maint dump-me
37131 @item maint dump-me
37132 @cindex @code{SIGQUIT} signal, dump core of @value{GDBN}
37133 Cause a fatal signal in the debugger and force it to dump its core.
37134 This is supported only on systems which support aborting a program
37135 with the @code{SIGQUIT} signal.
37137 @kindex maint internal-error
37138 @kindex maint internal-warning
37139 @kindex maint demangler-warning
37140 @cindex demangler crashes
37141 @item maint internal-error @r{[}@var{message-text}@r{]}
37142 @itemx maint internal-warning @r{[}@var{message-text}@r{]}
37143 @itemx maint demangler-warning @r{[}@var{message-text}@r{]}
37145 Cause @value{GDBN} to call the internal function @code{internal_error},
37146 @code{internal_warning} or @code{demangler_warning} and hence behave
37147 as though an internal problem has been detected. In addition to
37148 reporting the internal problem, these functions give the user the
37149 opportunity to either quit @value{GDBN} or (for @code{internal_error}
37150 and @code{internal_warning}) create a core file of the current
37151 @value{GDBN} session.
37153 These commands take an optional parameter @var{message-text} that is
37154 used as the text of the error or warning message.
37156 Here's an example of using @code{internal-error}:
37159 (@value{GDBP}) @kbd{maint internal-error testing, 1, 2}
37160 @dots{}/maint.c:121: internal-error: testing, 1, 2
37161 A problem internal to GDB has been detected. Further
37162 debugging may prove unreliable.
37163 Quit this debugging session? (y or n) @kbd{n}
37164 Create a core file? (y or n) @kbd{n}
37168 @cindex @value{GDBN} internal error
37169 @cindex internal errors, control of @value{GDBN} behavior
37170 @cindex demangler crashes
37172 @kindex maint set internal-error
37173 @kindex maint show internal-error
37174 @kindex maint set internal-warning
37175 @kindex maint show internal-warning
37176 @kindex maint set demangler-warning
37177 @kindex maint show demangler-warning
37178 @item maint set internal-error @var{action} [ask|yes|no]
37179 @itemx maint show internal-error @var{action}
37180 @itemx maint set internal-warning @var{action} [ask|yes|no]
37181 @itemx maint show internal-warning @var{action}
37182 @itemx maint set demangler-warning @var{action} [ask|yes|no]
37183 @itemx maint show demangler-warning @var{action}
37184 When @value{GDBN} reports an internal problem (error or warning) it
37185 gives the user the opportunity to both quit @value{GDBN} and create a
37186 core file of the current @value{GDBN} session. These commands let you
37187 override the default behaviour for each particular @var{action},
37188 described in the table below.
37192 You can specify that @value{GDBN} should always (yes) or never (no)
37193 quit. The default is to ask the user what to do.
37196 You can specify that @value{GDBN} should always (yes) or never (no)
37197 create a core file. The default is to ask the user what to do. Note
37198 that there is no @code{corefile} option for @code{demangler-warning}:
37199 demangler warnings always create a core file and this cannot be
37203 @kindex maint packet
37204 @item maint packet @var{text}
37205 If @value{GDBN} is talking to an inferior via the serial protocol,
37206 then this command sends the string @var{text} to the inferior, and
37207 displays the response packet. @value{GDBN} supplies the initial
37208 @samp{$} character, the terminating @samp{#} character, and the
37211 @kindex maint print architecture
37212 @item maint print architecture @r{[}@var{file}@r{]}
37213 Print the entire architecture configuration. The optional argument
37214 @var{file} names the file where the output goes.
37216 @kindex maint print c-tdesc @r{[}@var{file}@r{]}
37217 @item maint print c-tdesc
37218 Print the target description (@pxref{Target Descriptions}) as
37219 a C source file. By default, the target description is for the current
37220 target, but if the optional argument @var{file} is provided, that file
37221 is used to produce the description. The @var{file} should be an XML
37222 document, of the form described in @ref{Target Description Format}.
37223 The created source file is built into @value{GDBN} when @value{GDBN} is
37224 built again. This command is used by developers after they add or
37225 modify XML target descriptions.
37227 @kindex maint check xml-descriptions
37228 @item maint check xml-descriptions @var{dir}
37229 Check that the target descriptions dynamically created by @value{GDBN}
37230 equal the descriptions created from XML files found in @var{dir}.
37232 @anchor{maint check libthread-db}
37233 @kindex maint check libthread-db
37234 @item maint check libthread-db
37235 Run integrity checks on the current inferior's thread debugging
37236 library. This exercises all @code{libthread_db} functionality used by
37237 @value{GDBN} on GNU/Linux systems, and by extension also exercises the
37238 @code{proc_service} functions provided by @value{GDBN} that
37239 @code{libthread_db} uses. Note that parts of the test may be skipped
37240 on some platforms when debugging core files.
37242 @kindex maint print dummy-frames
37243 @item maint print dummy-frames
37244 Prints the contents of @value{GDBN}'s internal dummy-frame stack.
37247 (@value{GDBP}) @kbd{b add}
37249 (@value{GDBP}) @kbd{print add(2,3)}
37250 Breakpoint 2, add (a=2, b=3) at @dots{}
37252 The program being debugged stopped while in a function called from GDB.
37254 (@value{GDBP}) @kbd{maint print dummy-frames}
37255 0xa8206d8: id=@{stack=0xbfffe734,code=0xbfffe73f,!special@}, ptid=process 9353
37259 Takes an optional file parameter.
37261 @kindex maint print registers
37262 @kindex maint print raw-registers
37263 @kindex maint print cooked-registers
37264 @kindex maint print register-groups
37265 @kindex maint print remote-registers
37266 @item maint print registers @r{[}@var{file}@r{]}
37267 @itemx maint print raw-registers @r{[}@var{file}@r{]}
37268 @itemx maint print cooked-registers @r{[}@var{file}@r{]}
37269 @itemx maint print register-groups @r{[}@var{file}@r{]}
37270 @itemx maint print remote-registers @r{[}@var{file}@r{]}
37271 Print @value{GDBN}'s internal register data structures.
37273 The command @code{maint print raw-registers} includes the contents of
37274 the raw register cache; the command @code{maint print
37275 cooked-registers} includes the (cooked) value of all registers,
37276 including registers which aren't available on the target nor visible
37277 to user; the command @code{maint print register-groups} includes the
37278 groups that each register is a member of; and the command @code{maint
37279 print remote-registers} includes the remote target's register numbers
37280 and offsets in the `G' packets.
37282 These commands take an optional parameter, a file name to which to
37283 write the information.
37285 @kindex maint print reggroups
37286 @item maint print reggroups @r{[}@var{file}@r{]}
37287 Print @value{GDBN}'s internal register group data structures. The
37288 optional argument @var{file} tells to what file to write the
37291 The register groups info looks like this:
37294 (@value{GDBP}) @kbd{maint print reggroups}
37307 This command forces @value{GDBN} to flush its internal register cache.
37309 @kindex maint print objfiles
37310 @cindex info for known object files
37311 @item maint print objfiles @r{[}@var{regexp}@r{]}
37312 Print a dump of all known object files.
37313 If @var{regexp} is specified, only print object files whose names
37314 match @var{regexp}. For each object file, this command prints its name,
37315 address in memory, and all of its psymtabs and symtabs.
37317 @kindex maint print user-registers
37318 @cindex user registers
37319 @item maint print user-registers
37320 List all currently available @dfn{user registers}. User registers
37321 typically provide alternate names for actual hardware registers. They
37322 include the four ``standard'' registers @code{$fp}, @code{$pc},
37323 @code{$sp}, and @code{$ps}. @xref{standard registers}. User
37324 registers can be used in expressions in the same way as the canonical
37325 register names, but only the latter are listed by the @code{info
37326 registers} and @code{maint print registers} commands.
37328 @kindex maint print section-scripts
37329 @cindex info for known .debug_gdb_scripts-loaded scripts
37330 @item maint print section-scripts [@var{regexp}]
37331 Print a dump of scripts specified in the @code{.debug_gdb_section} section.
37332 If @var{regexp} is specified, only print scripts loaded by object files
37333 matching @var{regexp}.
37334 For each script, this command prints its name as specified in the objfile,
37335 and the full path if known.
37336 @xref{dotdebug_gdb_scripts section}.
37338 @kindex maint print statistics
37339 @cindex bcache statistics
37340 @item maint print statistics
37341 This command prints, for each object file in the program, various data
37342 about that object file followed by the byte cache (@dfn{bcache})
37343 statistics for the object file. The objfile data includes the number
37344 of minimal, partial, full, and stabs symbols, the number of types
37345 defined by the objfile, the number of as yet unexpanded psym tables,
37346 the number of line tables and string tables, and the amount of memory
37347 used by the various tables. The bcache statistics include the counts,
37348 sizes, and counts of duplicates of all and unique objects, max,
37349 average, and median entry size, total memory used and its overhead and
37350 savings, and various measures of the hash table size and chain
37353 @kindex maint print target-stack
37354 @cindex target stack description
37355 @item maint print target-stack
37356 A @dfn{target} is an interface between the debugger and a particular
37357 kind of file or process. Targets can be stacked in @dfn{strata},
37358 so that more than one target can potentially respond to a request.
37359 In particular, memory accesses will walk down the stack of targets
37360 until they find a target that is interested in handling that particular
37363 This command prints a short description of each layer that was pushed on
37364 the @dfn{target stack}, starting from the top layer down to the bottom one.
37366 @kindex maint print type
37367 @cindex type chain of a data type
37368 @item maint print type @var{expr}
37369 Print the type chain for a type specified by @var{expr}. The argument
37370 can be either a type name or a symbol. If it is a symbol, the type of
37371 that symbol is described. The type chain produced by this command is
37372 a recursive definition of the data type as stored in @value{GDBN}'s
37373 data structures, including its flags and contained types.
37375 @kindex maint selftest
37377 @item maint selftest @r{[}@var{filter}@r{]}
37378 Run any self tests that were compiled in to @value{GDBN}. This will
37379 print a message showing how many tests were run, and how many failed.
37380 If a @var{filter} is passed, only the tests with @var{filter} in their
37383 @kindex maint info selftests
37385 @item maint info selftests
37386 List the selftests compiled in to @value{GDBN}.
37388 @kindex maint set dwarf always-disassemble
37389 @kindex maint show dwarf always-disassemble
37390 @item maint set dwarf always-disassemble
37391 @item maint show dwarf always-disassemble
37392 Control the behavior of @code{info address} when using DWARF debugging
37395 The default is @code{off}, which means that @value{GDBN} should try to
37396 describe a variable's location in an easily readable format. When
37397 @code{on}, @value{GDBN} will instead display the DWARF location
37398 expression in an assembly-like format. Note that some locations are
37399 too complex for @value{GDBN} to describe simply; in this case you will
37400 always see the disassembly form.
37402 Here is an example of the resulting disassembly:
37405 (gdb) info addr argc
37406 Symbol "argc" is a complex DWARF expression:
37410 For more information on these expressions, see
37411 @uref{http://www.dwarfstd.org/, the DWARF standard}.
37413 @kindex maint set dwarf max-cache-age
37414 @kindex maint show dwarf max-cache-age
37415 @item maint set dwarf max-cache-age
37416 @itemx maint show dwarf max-cache-age
37417 Control the DWARF compilation unit cache.
37419 @cindex DWARF compilation units cache
37420 In object files with inter-compilation-unit references, such as those
37421 produced by the GCC option @samp{-feliminate-dwarf2-dups}, the DWARF
37422 reader needs to frequently refer to previously read compilation units.
37423 This setting controls how long a compilation unit will remain in the
37424 cache if it is not referenced. A higher limit means that cached
37425 compilation units will be stored in memory longer, and more total
37426 memory will be used. Setting it to zero disables caching, which will
37427 slow down @value{GDBN} startup, but reduce memory consumption.
37429 @kindex maint set dwarf unwinders
37430 @kindex maint show dwarf unwinders
37431 @item maint set dwarf unwinders
37432 @itemx maint show dwarf unwinders
37433 Control use of the DWARF frame unwinders.
37435 @cindex DWARF frame unwinders
37436 Many targets that support DWARF debugging use @value{GDBN}'s DWARF
37437 frame unwinders to build the backtrace. Many of these targets will
37438 also have a second mechanism for building the backtrace for use in
37439 cases where DWARF information is not available, this second mechanism
37440 is often an analysis of a function's prologue.
37442 In order to extend testing coverage of the second level stack
37443 unwinding mechanisms it is helpful to be able to disable the DWARF
37444 stack unwinders, this can be done with this switch.
37446 In normal use of @value{GDBN} disabling the DWARF unwinders is not
37447 advisable, there are cases that are better handled through DWARF than
37448 prologue analysis, and the debug experience is likely to be better
37449 with the DWARF frame unwinders enabled.
37451 If DWARF frame unwinders are not supported for a particular target
37452 architecture, then enabling this flag does not cause them to be used.
37453 @kindex maint set profile
37454 @kindex maint show profile
37455 @cindex profiling GDB
37456 @item maint set profile
37457 @itemx maint show profile
37458 Control profiling of @value{GDBN}.
37460 Profiling will be disabled until you use the @samp{maint set profile}
37461 command to enable it. When you enable profiling, the system will begin
37462 collecting timing and execution count data; when you disable profiling or
37463 exit @value{GDBN}, the results will be written to a log file. Remember that
37464 if you use profiling, @value{GDBN} will overwrite the profiling log file
37465 (often called @file{gmon.out}). If you have a record of important profiling
37466 data in a @file{gmon.out} file, be sure to move it to a safe location.
37468 Configuring with @samp{--enable-profiling} arranges for @value{GDBN} to be
37469 compiled with the @samp{-pg} compiler option.
37471 @kindex maint set show-debug-regs
37472 @kindex maint show show-debug-regs
37473 @cindex hardware debug registers
37474 @item maint set show-debug-regs
37475 @itemx maint show show-debug-regs
37476 Control whether to show variables that mirror the hardware debug
37477 registers. Use @code{on} to enable, @code{off} to disable. If
37478 enabled, the debug registers values are shown when @value{GDBN} inserts or
37479 removes a hardware breakpoint or watchpoint, and when the inferior
37480 triggers a hardware-assisted breakpoint or watchpoint.
37482 @kindex maint set show-all-tib
37483 @kindex maint show show-all-tib
37484 @item maint set show-all-tib
37485 @itemx maint show show-all-tib
37486 Control whether to show all non zero areas within a 1k block starting
37487 at thread local base, when using the @samp{info w32 thread-information-block}
37490 @kindex maint set target-async
37491 @kindex maint show target-async
37492 @item maint set target-async
37493 @itemx maint show target-async
37494 This controls whether @value{GDBN} targets operate in synchronous or
37495 asynchronous mode (@pxref{Background Execution}). Normally the
37496 default is asynchronous, if it is available; but this can be changed
37497 to more easily debug problems occurring only in synchronous mode.
37499 @kindex maint set target-non-stop @var{mode} [on|off|auto]
37500 @kindex maint show target-non-stop
37501 @item maint set target-non-stop
37502 @itemx maint show target-non-stop
37504 This controls whether @value{GDBN} targets always operate in non-stop
37505 mode even if @code{set non-stop} is @code{off} (@pxref{Non-Stop
37506 Mode}). The default is @code{auto}, meaning non-stop mode is enabled
37507 if supported by the target.
37510 @item maint set target-non-stop auto
37511 This is the default mode. @value{GDBN} controls the target in
37512 non-stop mode if the target supports it.
37514 @item maint set target-non-stop on
37515 @value{GDBN} controls the target in non-stop mode even if the target
37516 does not indicate support.
37518 @item maint set target-non-stop off
37519 @value{GDBN} does not control the target in non-stop mode even if the
37520 target supports it.
37523 @kindex maint set per-command
37524 @kindex maint show per-command
37525 @item maint set per-command
37526 @itemx maint show per-command
37527 @cindex resources used by commands
37529 @value{GDBN} can display the resources used by each command.
37530 This is useful in debugging performance problems.
37533 @item maint set per-command space [on|off]
37534 @itemx maint show per-command space
37535 Enable or disable the printing of the memory used by GDB for each command.
37536 If enabled, @value{GDBN} will display how much memory each command
37537 took, following the command's own output.
37538 This can also be requested by invoking @value{GDBN} with the
37539 @option{--statistics} command-line switch (@pxref{Mode Options}).
37541 @item maint set per-command time [on|off]
37542 @itemx maint show per-command time
37543 Enable or disable the printing of the execution time of @value{GDBN}
37545 If enabled, @value{GDBN} will display how much time it
37546 took to execute each command, following the command's own output.
37547 Both CPU time and wallclock time are printed.
37548 Printing both is useful when trying to determine whether the cost is
37549 CPU or, e.g., disk/network latency.
37550 Note that the CPU time printed is for @value{GDBN} only, it does not include
37551 the execution time of the inferior because there's no mechanism currently
37552 to compute how much time was spent by @value{GDBN} and how much time was
37553 spent by the program been debugged.
37554 This can also be requested by invoking @value{GDBN} with the
37555 @option{--statistics} command-line switch (@pxref{Mode Options}).
37557 @item maint set per-command symtab [on|off]
37558 @itemx maint show per-command symtab
37559 Enable or disable the printing of basic symbol table statistics
37561 If enabled, @value{GDBN} will display the following information:
37565 number of symbol tables
37567 number of primary symbol tables
37569 number of blocks in the blockvector
37573 @kindex maint set check-libthread-db
37574 @kindex maint show check-libthread-db
37575 @item maint set check-libthread-db [on|off]
37576 @itemx maint show check-libthread-db
37577 Control whether @value{GDBN} should run integrity checks on inferior
37578 specific thread debugging libraries as they are loaded. The default
37579 is not to perform such checks. If any check fails @value{GDBN} will
37580 unload the library and continue searching for a suitable candidate as
37581 described in @ref{set libthread-db-search-path}. For more information
37582 about the tests, see @ref{maint check libthread-db}.
37584 @kindex maint space
37585 @cindex memory used by commands
37586 @item maint space @var{value}
37587 An alias for @code{maint set per-command space}.
37588 A non-zero value enables it, zero disables it.
37591 @cindex time of command execution
37592 @item maint time @var{value}
37593 An alias for @code{maint set per-command time}.
37594 A non-zero value enables it, zero disables it.
37596 @kindex maint translate-address
37597 @item maint translate-address @r{[}@var{section}@r{]} @var{addr}
37598 Find the symbol stored at the location specified by the address
37599 @var{addr} and an optional section name @var{section}. If found,
37600 @value{GDBN} prints the name of the closest symbol and an offset from
37601 the symbol's location to the specified address. This is similar to
37602 the @code{info address} command (@pxref{Symbols}), except that this
37603 command also allows to find symbols in other sections.
37605 If section was not specified, the section in which the symbol was found
37606 is also printed. For dynamically linked executables, the name of
37607 executable or shared library containing the symbol is printed as well.
37609 @kindex maint test-options
37610 @item maint test-options require-delimiter
37611 @itemx maint test-options unknown-is-error
37612 @itemx maint test-options unknown-is-operand
37613 These commands are used by the testsuite to validate the command
37614 options framework. The @code{require-delimiter} variant requires a
37615 double-dash delimiter to indicate end of options. The
37616 @code{unknown-is-error} and @code{unknown-is-operand} do not. The
37617 @code{unknown-is-error} variant throws an error on unknown option,
37618 while @code{unknown-is-operand} treats unknown options as the start of
37619 the command's operands. When run, the commands output the result of
37620 the processed options. When completed, the commands store the
37621 internal result of completion in a variable exposed by the @code{maint
37622 show test-options-completion-result} command.
37624 @kindex maint show test-options-completion-result
37625 @item maint show test-options-completion-result
37626 Shows the result of completing the @code{maint test-options}
37627 subcommands. This is used by the testsuite to validate completion
37628 support in the command options framework.
37630 @kindex maint test-settings
37631 @item maint test-settings set @var{kind}
37632 @itemx maint test-settings show @var{kind}
37633 These are representative commands for each @var{kind} of setting type
37634 @value{GDBN} supports. They are used by the testsuite for exercising
37635 the settings infrastructure.
37638 The following command is useful for non-interactive invocations of
37639 @value{GDBN}, such as in the test suite.
37642 @item set watchdog @var{nsec}
37643 @kindex set watchdog
37644 @cindex watchdog timer
37645 @cindex timeout for commands
37646 Set the maximum number of seconds @value{GDBN} will wait for the
37647 target operation to finish. If this time expires, @value{GDBN}
37648 reports and error and the command is aborted.
37650 @item show watchdog
37651 Show the current setting of the target wait timeout.
37654 @node Remote Protocol
37655 @appendix @value{GDBN} Remote Serial Protocol
37660 * Stop Reply Packets::
37661 * General Query Packets::
37662 * Architecture-Specific Protocol Details::
37663 * Tracepoint Packets::
37664 * Host I/O Packets::
37666 * Notification Packets::
37667 * Remote Non-Stop::
37668 * Packet Acknowledgment::
37670 * File-I/O Remote Protocol Extension::
37671 * Library List Format::
37672 * Library List Format for SVR4 Targets::
37673 * Memory Map Format::
37674 * Thread List Format::
37675 * Traceframe Info Format::
37676 * Branch Trace Format::
37677 * Branch Trace Configuration Format::
37683 There may be occasions when you need to know something about the
37684 protocol---for example, if there is only one serial port to your target
37685 machine, you might want your program to do something special if it
37686 recognizes a packet meant for @value{GDBN}.
37688 In the examples below, @samp{->} and @samp{<-} are used to indicate
37689 transmitted and received data, respectively.
37691 @cindex protocol, @value{GDBN} remote serial
37692 @cindex serial protocol, @value{GDBN} remote
37693 @cindex remote serial protocol
37694 All @value{GDBN} commands and responses (other than acknowledgments
37695 and notifications, see @ref{Notification Packets}) are sent as a
37696 @var{packet}. A @var{packet} is introduced with the character
37697 @samp{$}, the actual @var{packet-data}, and the terminating character
37698 @samp{#} followed by a two-digit @var{checksum}:
37701 @code{$}@var{packet-data}@code{#}@var{checksum}
37705 @cindex checksum, for @value{GDBN} remote
37707 The two-digit @var{checksum} is computed as the modulo 256 sum of all
37708 characters between the leading @samp{$} and the trailing @samp{#} (an
37709 eight bit unsigned checksum).
37711 Implementors should note that prior to @value{GDBN} 5.0 the protocol
37712 specification also included an optional two-digit @var{sequence-id}:
37715 @code{$}@var{sequence-id}@code{:}@var{packet-data}@code{#}@var{checksum}
37718 @cindex sequence-id, for @value{GDBN} remote
37720 That @var{sequence-id} was appended to the acknowledgment. @value{GDBN}
37721 has never output @var{sequence-id}s. Stubs that handle packets added
37722 since @value{GDBN} 5.0 must not accept @var{sequence-id}.
37724 When either the host or the target machine receives a packet, the first
37725 response expected is an acknowledgment: either @samp{+} (to indicate
37726 the package was received correctly) or @samp{-} (to request
37730 -> @code{$}@var{packet-data}@code{#}@var{checksum}
37735 The @samp{+}/@samp{-} acknowledgments can be disabled
37736 once a connection is established.
37737 @xref{Packet Acknowledgment}, for details.
37739 The host (@value{GDBN}) sends @var{command}s, and the target (the
37740 debugging stub incorporated in your program) sends a @var{response}. In
37741 the case of step and continue @var{command}s, the response is only sent
37742 when the operation has completed, and the target has again stopped all
37743 threads in all attached processes. This is the default all-stop mode
37744 behavior, but the remote protocol also supports @value{GDBN}'s non-stop
37745 execution mode; see @ref{Remote Non-Stop}, for details.
37747 @var{packet-data} consists of a sequence of characters with the
37748 exception of @samp{#} and @samp{$} (see @samp{X} packet for additional
37751 @cindex remote protocol, field separator
37752 Fields within the packet should be separated using @samp{,} @samp{;} or
37753 @samp{:}. Except where otherwise noted all numbers are represented in
37754 @sc{hex} with leading zeros suppressed.
37756 Implementors should note that prior to @value{GDBN} 5.0, the character
37757 @samp{:} could not appear as the third character in a packet (as it
37758 would potentially conflict with the @var{sequence-id}).
37760 @cindex remote protocol, binary data
37761 @anchor{Binary Data}
37762 Binary data in most packets is encoded either as two hexadecimal
37763 digits per byte of binary data. This allowed the traditional remote
37764 protocol to work over connections which were only seven-bit clean.
37765 Some packets designed more recently assume an eight-bit clean
37766 connection, and use a more efficient encoding to send and receive
37769 The binary data representation uses @code{7d} (@sc{ascii} @samp{@}})
37770 as an escape character. Any escaped byte is transmitted as the escape
37771 character followed by the original character XORed with @code{0x20}.
37772 For example, the byte @code{0x7d} would be transmitted as the two
37773 bytes @code{0x7d 0x5d}. The bytes @code{0x23} (@sc{ascii} @samp{#}),
37774 @code{0x24} (@sc{ascii} @samp{$}), and @code{0x7d} (@sc{ascii}
37775 @samp{@}}) must always be escaped. Responses sent by the stub
37776 must also escape @code{0x2a} (@sc{ascii} @samp{*}), so that it
37777 is not interpreted as the start of a run-length encoded sequence
37780 Response @var{data} can be run-length encoded to save space.
37781 Run-length encoding replaces runs of identical characters with one
37782 instance of the repeated character, followed by a @samp{*} and a
37783 repeat count. The repeat count is itself sent encoded, to avoid
37784 binary characters in @var{data}: a value of @var{n} is sent as
37785 @code{@var{n}+29}. For a repeat count greater or equal to 3, this
37786 produces a printable @sc{ascii} character, e.g.@: a space (@sc{ascii}
37787 code 32) for a repeat count of 3. (This is because run-length
37788 encoding starts to win for counts 3 or more.) Thus, for example,
37789 @samp{0* } is a run-length encoding of ``0000'': the space character
37790 after @samp{*} means repeat the leading @code{0} @w{@code{32 - 29 =
37793 The printable characters @samp{#} and @samp{$} or with a numeric value
37794 greater than 126 must not be used. Runs of six repeats (@samp{#}) or
37795 seven repeats (@samp{$}) can be expanded using a repeat count of only
37796 five (@samp{"}). For example, @samp{00000000} can be encoded as
37799 The error response returned for some packets includes a two character
37800 error number. That number is not well defined.
37802 @cindex empty response, for unsupported packets
37803 For any @var{command} not supported by the stub, an empty response
37804 (@samp{$#00}) should be returned. That way it is possible to extend the
37805 protocol. A newer @value{GDBN} can tell if a packet is supported based
37808 At a minimum, a stub is required to support the @samp{g} and @samp{G}
37809 commands for register access, and the @samp{m} and @samp{M} commands
37810 for memory access. Stubs that only control single-threaded targets
37811 can implement run control with the @samp{c} (continue), and @samp{s}
37812 (step) commands. Stubs that support multi-threading targets should
37813 support the @samp{vCont} command. All other commands are optional.
37818 The following table provides a complete list of all currently defined
37819 @var{command}s and their corresponding response @var{data}.
37820 @xref{File-I/O Remote Protocol Extension}, for details about the File
37821 I/O extension of the remote protocol.
37823 Each packet's description has a template showing the packet's overall
37824 syntax, followed by an explanation of the packet's meaning. We
37825 include spaces in some of the templates for clarity; these are not
37826 part of the packet's syntax. No @value{GDBN} packet uses spaces to
37827 separate its components. For example, a template like @samp{foo
37828 @var{bar} @var{baz}} describes a packet beginning with the three ASCII
37829 bytes @samp{foo}, followed by a @var{bar}, followed directly by a
37830 @var{baz}. @value{GDBN} does not transmit a space character between the
37831 @samp{foo} and the @var{bar}, or between the @var{bar} and the
37834 @cindex @var{thread-id}, in remote protocol
37835 @anchor{thread-id syntax}
37836 Several packets and replies include a @var{thread-id} field to identify
37837 a thread. Normally these are positive numbers with a target-specific
37838 interpretation, formatted as big-endian hex strings. A @var{thread-id}
37839 can also be a literal @samp{-1} to indicate all threads, or @samp{0} to
37842 In addition, the remote protocol supports a multiprocess feature in
37843 which the @var{thread-id} syntax is extended to optionally include both
37844 process and thread ID fields, as @samp{p@var{pid}.@var{tid}}.
37845 The @var{pid} (process) and @var{tid} (thread) components each have the
37846 format described above: a positive number with target-specific
37847 interpretation formatted as a big-endian hex string, literal @samp{-1}
37848 to indicate all processes or threads (respectively), or @samp{0} to
37849 indicate an arbitrary process or thread. Specifying just a process, as
37850 @samp{p@var{pid}}, is equivalent to @samp{p@var{pid}.-1}. It is an
37851 error to specify all processes but a specific thread, such as
37852 @samp{p-1.@var{tid}}. Note that the @samp{p} prefix is @emph{not} used
37853 for those packets and replies explicitly documented to include a process
37854 ID, rather than a @var{thread-id}.
37856 The multiprocess @var{thread-id} syntax extensions are only used if both
37857 @value{GDBN} and the stub report support for the @samp{multiprocess}
37858 feature using @samp{qSupported}. @xref{multiprocess extensions}, for
37861 Note that all packet forms beginning with an upper- or lower-case
37862 letter, other than those described here, are reserved for future use.
37864 Here are the packet descriptions.
37869 @cindex @samp{!} packet
37870 @anchor{extended mode}
37871 Enable extended mode. In extended mode, the remote server is made
37872 persistent. The @samp{R} packet is used to restart the program being
37878 The remote target both supports and has enabled extended mode.
37882 @cindex @samp{?} packet
37884 Indicate the reason the target halted. The reply is the same as for
37885 step and continue. This packet has a special interpretation when the
37886 target is in non-stop mode; see @ref{Remote Non-Stop}.
37889 @xref{Stop Reply Packets}, for the reply specifications.
37891 @item A @var{arglen},@var{argnum},@var{arg},@dots{}
37892 @cindex @samp{A} packet
37893 Initialized @code{argv[]} array passed into program. @var{arglen}
37894 specifies the number of bytes in the hex encoded byte stream
37895 @var{arg}. See @code{gdbserver} for more details.
37900 The arguments were set.
37906 @cindex @samp{b} packet
37907 (Don't use this packet; its behavior is not well-defined.)
37908 Change the serial line speed to @var{baud}.
37910 JTC: @emph{When does the transport layer state change? When it's
37911 received, or after the ACK is transmitted. In either case, there are
37912 problems if the command or the acknowledgment packet is dropped.}
37914 Stan: @emph{If people really wanted to add something like this, and get
37915 it working for the first time, they ought to modify ser-unix.c to send
37916 some kind of out-of-band message to a specially-setup stub and have the
37917 switch happen "in between" packets, so that from remote protocol's point
37918 of view, nothing actually happened.}
37920 @item B @var{addr},@var{mode}
37921 @cindex @samp{B} packet
37922 Set (@var{mode} is @samp{S}) or clear (@var{mode} is @samp{C}) a
37923 breakpoint at @var{addr}.
37925 Don't use this packet. Use the @samp{Z} and @samp{z} packets instead
37926 (@pxref{insert breakpoint or watchpoint packet}).
37928 @cindex @samp{bc} packet
37931 Backward continue. Execute the target system in reverse. No parameter.
37932 @xref{Reverse Execution}, for more information.
37935 @xref{Stop Reply Packets}, for the reply specifications.
37937 @cindex @samp{bs} packet
37940 Backward single step. Execute one instruction in reverse. No parameter.
37941 @xref{Reverse Execution}, for more information.
37944 @xref{Stop Reply Packets}, for the reply specifications.
37946 @item c @r{[}@var{addr}@r{]}
37947 @cindex @samp{c} packet
37948 Continue at @var{addr}, which is the address to resume. If @var{addr}
37949 is omitted, resume at current address.
37951 This packet is deprecated for multi-threading support. @xref{vCont
37955 @xref{Stop Reply Packets}, for the reply specifications.
37957 @item C @var{sig}@r{[};@var{addr}@r{]}
37958 @cindex @samp{C} packet
37959 Continue with signal @var{sig} (hex signal number). If
37960 @samp{;@var{addr}} is omitted, resume at same address.
37962 This packet is deprecated for multi-threading support. @xref{vCont
37966 @xref{Stop Reply Packets}, for the reply specifications.
37969 @cindex @samp{d} packet
37972 Don't use this packet; instead, define a general set packet
37973 (@pxref{General Query Packets}).
37977 @cindex @samp{D} packet
37978 The first form of the packet is used to detach @value{GDBN} from the
37979 remote system. It is sent to the remote target
37980 before @value{GDBN} disconnects via the @code{detach} command.
37982 The second form, including a process ID, is used when multiprocess
37983 protocol extensions are enabled (@pxref{multiprocess extensions}), to
37984 detach only a specific process. The @var{pid} is specified as a
37985 big-endian hex string.
37995 @item F @var{RC},@var{EE},@var{CF};@var{XX}
37996 @cindex @samp{F} packet
37997 A reply from @value{GDBN} to an @samp{F} packet sent by the target.
37998 This is part of the File-I/O protocol extension. @xref{File-I/O
37999 Remote Protocol Extension}, for the specification.
38002 @anchor{read registers packet}
38003 @cindex @samp{g} packet
38004 Read general registers.
38008 @item @var{XX@dots{}}
38009 Each byte of register data is described by two hex digits. The bytes
38010 with the register are transmitted in target byte order. The size of
38011 each register and their position within the @samp{g} packet are
38012 determined by the @value{GDBN} internal gdbarch functions
38013 @code{DEPRECATED_REGISTER_RAW_SIZE} and @code{gdbarch_register_name}.
38015 When reading registers from a trace frame (@pxref{Analyze Collected
38016 Data,,Using the Collected Data}), the stub may also return a string of
38017 literal @samp{x}'s in place of the register data digits, to indicate
38018 that the corresponding register has not been collected, thus its value
38019 is unavailable. For example, for an architecture with 4 registers of
38020 4 bytes each, the following reply indicates to @value{GDBN} that
38021 registers 0 and 2 have not been collected, while registers 1 and 3
38022 have been collected, and both have zero value:
38026 <- @code{xxxxxxxx00000000xxxxxxxx00000000}
38033 @item G @var{XX@dots{}}
38034 @cindex @samp{G} packet
38035 Write general registers. @xref{read registers packet}, for a
38036 description of the @var{XX@dots{}} data.
38046 @item H @var{op} @var{thread-id}
38047 @cindex @samp{H} packet
38048 Set thread for subsequent operations (@samp{m}, @samp{M}, @samp{g},
38049 @samp{G}, et.al.). Depending on the operation to be performed, @var{op}
38050 should be @samp{c} for step and continue operations (note that this
38051 is deprecated, supporting the @samp{vCont} command is a better
38052 option), and @samp{g} for other operations. The thread designator
38053 @var{thread-id} has the format and interpretation described in
38054 @ref{thread-id syntax}.
38065 @c 'H': How restrictive (or permissive) is the thread model. If a
38066 @c thread is selected and stopped, are other threads allowed
38067 @c to continue to execute? As I mentioned above, I think the
38068 @c semantics of each command when a thread is selected must be
38069 @c described. For example:
38071 @c 'g': If the stub supports threads and a specific thread is
38072 @c selected, returns the register block from that thread;
38073 @c otherwise returns current registers.
38075 @c 'G' If the stub supports threads and a specific thread is
38076 @c selected, sets the registers of the register block of
38077 @c that thread; otherwise sets current registers.
38079 @item i @r{[}@var{addr}@r{[},@var{nnn}@r{]]}
38080 @anchor{cycle step packet}
38081 @cindex @samp{i} packet
38082 Step the remote target by a single clock cycle. If @samp{,@var{nnn}} is
38083 present, cycle step @var{nnn} cycles. If @var{addr} is present, cycle
38084 step starting at that address.
38087 @cindex @samp{I} packet
38088 Signal, then cycle step. @xref{step with signal packet}. @xref{cycle
38092 @cindex @samp{k} packet
38095 The exact effect of this packet is not specified.
38097 For a bare-metal target, it may power cycle or reset the target
38098 system. For that reason, the @samp{k} packet has no reply.
38100 For a single-process target, it may kill that process if possible.
38102 A multiple-process target may choose to kill just one process, or all
38103 that are under @value{GDBN}'s control. For more precise control, use
38104 the vKill packet (@pxref{vKill packet}).
38106 If the target system immediately closes the connection in response to
38107 @samp{k}, @value{GDBN} does not consider the lack of packet
38108 acknowledgment to be an error, and assumes the kill was successful.
38110 If connected using @kbd{target extended-remote}, and the target does
38111 not close the connection in response to a kill request, @value{GDBN}
38112 probes the target state as if a new connection was opened
38113 (@pxref{? packet}).
38115 @item m @var{addr},@var{length}
38116 @cindex @samp{m} packet
38117 Read @var{length} addressable memory units starting at address @var{addr}
38118 (@pxref{addressable memory unit}). Note that @var{addr} may not be aligned to
38119 any particular boundary.
38121 The stub need not use any particular size or alignment when gathering
38122 data from memory for the response; even if @var{addr} is word-aligned
38123 and @var{length} is a multiple of the word size, the stub is free to
38124 use byte accesses, or not. For this reason, this packet may not be
38125 suitable for accessing memory-mapped I/O devices.
38126 @cindex alignment of remote memory accesses
38127 @cindex size of remote memory accesses
38128 @cindex memory, alignment and size of remote accesses
38132 @item @var{XX@dots{}}
38133 Memory contents; each byte is transmitted as a two-digit hexadecimal number.
38134 The reply may contain fewer addressable memory units than requested if the
38135 server was able to read only part of the region of memory.
38140 @item M @var{addr},@var{length}:@var{XX@dots{}}
38141 @cindex @samp{M} packet
38142 Write @var{length} addressable memory units starting at address @var{addr}
38143 (@pxref{addressable memory unit}). The data is given by @var{XX@dots{}}; each
38144 byte is transmitted as a two-digit hexadecimal number.
38151 for an error (this includes the case where only part of the data was
38156 @cindex @samp{p} packet
38157 Read the value of register @var{n}; @var{n} is in hex.
38158 @xref{read registers packet}, for a description of how the returned
38159 register value is encoded.
38163 @item @var{XX@dots{}}
38164 the register's value
38168 Indicating an unrecognized @var{query}.
38171 @item P @var{n@dots{}}=@var{r@dots{}}
38172 @anchor{write register packet}
38173 @cindex @samp{P} packet
38174 Write register @var{n@dots{}} with value @var{r@dots{}}. The register
38175 number @var{n} is in hexadecimal, and @var{r@dots{}} contains two hex
38176 digits for each byte in the register (target byte order).
38186 @item q @var{name} @var{params}@dots{}
38187 @itemx Q @var{name} @var{params}@dots{}
38188 @cindex @samp{q} packet
38189 @cindex @samp{Q} packet
38190 General query (@samp{q}) and set (@samp{Q}). These packets are
38191 described fully in @ref{General Query Packets}.
38194 @cindex @samp{r} packet
38195 Reset the entire system.
38197 Don't use this packet; use the @samp{R} packet instead.
38200 @cindex @samp{R} packet
38201 Restart the program being debugged. The @var{XX}, while needed, is ignored.
38202 This packet is only available in extended mode (@pxref{extended mode}).
38204 The @samp{R} packet has no reply.
38206 @item s @r{[}@var{addr}@r{]}
38207 @cindex @samp{s} packet
38208 Single step, resuming at @var{addr}. If
38209 @var{addr} is omitted, resume at same address.
38211 This packet is deprecated for multi-threading support. @xref{vCont
38215 @xref{Stop Reply Packets}, for the reply specifications.
38217 @item S @var{sig}@r{[};@var{addr}@r{]}
38218 @anchor{step with signal packet}
38219 @cindex @samp{S} packet
38220 Step with signal. This is analogous to the @samp{C} packet, but
38221 requests a single-step, rather than a normal resumption of execution.
38223 This packet is deprecated for multi-threading support. @xref{vCont
38227 @xref{Stop Reply Packets}, for the reply specifications.
38229 @item t @var{addr}:@var{PP},@var{MM}
38230 @cindex @samp{t} packet
38231 Search backwards starting at address @var{addr} for a match with pattern
38232 @var{PP} and mask @var{MM}, both of which are are 4 byte long.
38233 There must be at least 3 digits in @var{addr}.
38235 @item T @var{thread-id}
38236 @cindex @samp{T} packet
38237 Find out if the thread @var{thread-id} is alive. @xref{thread-id syntax}.
38242 thread is still alive
38248 Packets starting with @samp{v} are identified by a multi-letter name,
38249 up to the first @samp{;} or @samp{?} (or the end of the packet).
38251 @item vAttach;@var{pid}
38252 @cindex @samp{vAttach} packet
38253 Attach to a new process with the specified process ID @var{pid}.
38254 The process ID is a
38255 hexadecimal integer identifying the process. In all-stop mode, all
38256 threads in the attached process are stopped; in non-stop mode, it may be
38257 attached without being stopped if that is supported by the target.
38259 @c In non-stop mode, on a successful vAttach, the stub should set the
38260 @c current thread to a thread of the newly-attached process. After
38261 @c attaching, GDB queries for the attached process's thread ID with qC.
38262 @c Also note that, from a user perspective, whether or not the
38263 @c target is stopped on attach in non-stop mode depends on whether you
38264 @c use the foreground or background version of the attach command, not
38265 @c on what vAttach does; GDB does the right thing with respect to either
38266 @c stopping or restarting threads.
38268 This packet is only available in extended mode (@pxref{extended mode}).
38274 @item @r{Any stop packet}
38275 for success in all-stop mode (@pxref{Stop Reply Packets})
38277 for success in non-stop mode (@pxref{Remote Non-Stop})
38280 @item vCont@r{[};@var{action}@r{[}:@var{thread-id}@r{]]}@dots{}
38281 @cindex @samp{vCont} packet
38282 @anchor{vCont packet}
38283 Resume the inferior, specifying different actions for each thread.
38285 For each inferior thread, the leftmost action with a matching
38286 @var{thread-id} is applied. Threads that don't match any action
38287 remain in their current state. Thread IDs are specified using the
38288 syntax described in @ref{thread-id syntax}. If multiprocess
38289 extensions (@pxref{multiprocess extensions}) are supported, actions
38290 can be specified to match all threads in a process by using the
38291 @samp{p@var{pid}.-1} form of the @var{thread-id}. An action with no
38292 @var{thread-id} matches all threads. Specifying no actions is an
38295 Currently supported actions are:
38301 Continue with signal @var{sig}. The signal @var{sig} should be two hex digits.
38305 Step with signal @var{sig}. The signal @var{sig} should be two hex digits.
38308 @item r @var{start},@var{end}
38309 Step once, and then keep stepping as long as the thread stops at
38310 addresses between @var{start} (inclusive) and @var{end} (exclusive).
38311 The remote stub reports a stop reply when either the thread goes out
38312 of the range or is stopped due to an unrelated reason, such as hitting
38313 a breakpoint. @xref{range stepping}.
38315 If the range is empty (@var{start} == @var{end}), then the action
38316 becomes equivalent to the @samp{s} action. In other words,
38317 single-step once, and report the stop (even if the stepped instruction
38318 jumps to @var{start}).
38320 (A stop reply may be sent at any point even if the PC is still within
38321 the stepping range; for example, it is valid to implement this packet
38322 in a degenerate way as a single instruction step operation.)
38326 The optional argument @var{addr} normally associated with the
38327 @samp{c}, @samp{C}, @samp{s}, and @samp{S} packets is
38328 not supported in @samp{vCont}.
38330 The @samp{t} action is only relevant in non-stop mode
38331 (@pxref{Remote Non-Stop}) and may be ignored by the stub otherwise.
38332 A stop reply should be generated for any affected thread not already stopped.
38333 When a thread is stopped by means of a @samp{t} action,
38334 the corresponding stop reply should indicate that the thread has stopped with
38335 signal @samp{0}, regardless of whether the target uses some other signal
38336 as an implementation detail.
38338 The server must ignore @samp{c}, @samp{C}, @samp{s}, @samp{S}, and
38339 @samp{r} actions for threads that are already running. Conversely,
38340 the server must ignore @samp{t} actions for threads that are already
38343 @emph{Note:} In non-stop mode, a thread is considered running until
38344 @value{GDBN} acknowleges an asynchronous stop notification for it with
38345 the @samp{vStopped} packet (@pxref{Remote Non-Stop}).
38347 The stub must support @samp{vCont} if it reports support for
38348 multiprocess extensions (@pxref{multiprocess extensions}).
38351 @xref{Stop Reply Packets}, for the reply specifications.
38354 @cindex @samp{vCont?} packet
38355 Request a list of actions supported by the @samp{vCont} packet.
38359 @item vCont@r{[};@var{action}@dots{}@r{]}
38360 The @samp{vCont} packet is supported. Each @var{action} is a supported
38361 command in the @samp{vCont} packet.
38363 The @samp{vCont} packet is not supported.
38366 @anchor{vCtrlC packet}
38368 @cindex @samp{vCtrlC} packet
38369 Interrupt remote target as if a control-C was pressed on the remote
38370 terminal. This is the equivalent to reacting to the @code{^C}
38371 (@samp{\003}, the control-C character) character in all-stop mode
38372 while the target is running, except this works in non-stop mode.
38373 @xref{interrupting remote targets}, for more info on the all-stop
38384 @item vFile:@var{operation}:@var{parameter}@dots{}
38385 @cindex @samp{vFile} packet
38386 Perform a file operation on the target system. For details,
38387 see @ref{Host I/O Packets}.
38389 @item vFlashErase:@var{addr},@var{length}
38390 @cindex @samp{vFlashErase} packet
38391 Direct the stub to erase @var{length} bytes of flash starting at
38392 @var{addr}. The region may enclose any number of flash blocks, but
38393 its start and end must fall on block boundaries, as indicated by the
38394 flash block size appearing in the memory map (@pxref{Memory Map
38395 Format}). @value{GDBN} groups flash memory programming operations
38396 together, and sends a @samp{vFlashDone} request after each group; the
38397 stub is allowed to delay erase operation until the @samp{vFlashDone}
38398 packet is received.
38408 @item vFlashWrite:@var{addr}:@var{XX@dots{}}
38409 @cindex @samp{vFlashWrite} packet
38410 Direct the stub to write data to flash address @var{addr}. The data
38411 is passed in binary form using the same encoding as for the @samp{X}
38412 packet (@pxref{Binary Data}). The memory ranges specified by
38413 @samp{vFlashWrite} packets preceding a @samp{vFlashDone} packet must
38414 not overlap, and must appear in order of increasing addresses
38415 (although @samp{vFlashErase} packets for higher addresses may already
38416 have been received; the ordering is guaranteed only between
38417 @samp{vFlashWrite} packets). If a packet writes to an address that was
38418 neither erased by a preceding @samp{vFlashErase} packet nor by some other
38419 target-specific method, the results are unpredictable.
38427 for vFlashWrite addressing non-flash memory
38433 @cindex @samp{vFlashDone} packet
38434 Indicate to the stub that flash programming operation is finished.
38435 The stub is permitted to delay or batch the effects of a group of
38436 @samp{vFlashErase} and @samp{vFlashWrite} packets until a
38437 @samp{vFlashDone} packet is received. The contents of the affected
38438 regions of flash memory are unpredictable until the @samp{vFlashDone}
38439 request is completed.
38441 @item vKill;@var{pid}
38442 @cindex @samp{vKill} packet
38443 @anchor{vKill packet}
38444 Kill the process with the specified process ID @var{pid}, which is a
38445 hexadecimal integer identifying the process. This packet is used in
38446 preference to @samp{k} when multiprocess protocol extensions are
38447 supported; see @ref{multiprocess extensions}.
38457 @item vMustReplyEmpty
38458 @cindex @samp{vMustReplyEmpty} packet
38459 The correct reply to an unknown @samp{v} packet is to return the empty
38460 string, however, some older versions of @command{gdbserver} would
38461 incorrectly return @samp{OK} for unknown @samp{v} packets.
38463 The @samp{vMustReplyEmpty} is used as a feature test to check how
38464 @command{gdbserver} handles unknown packets, it is important that this
38465 packet be handled in the same way as other unknown @samp{v} packets.
38466 If this packet is handled differently to other unknown @samp{v}
38467 packets then it is possile that @value{GDBN} may run into problems in
38468 other areas, specifically around use of @samp{vFile:setfs:}.
38470 @item vRun;@var{filename}@r{[};@var{argument}@r{]}@dots{}
38471 @cindex @samp{vRun} packet
38472 Run the program @var{filename}, passing it each @var{argument} on its
38473 command line. The file and arguments are hex-encoded strings. If
38474 @var{filename} is an empty string, the stub may use a default program
38475 (e.g.@: the last program run). The program is created in the stopped
38478 @c FIXME: What about non-stop mode?
38480 This packet is only available in extended mode (@pxref{extended mode}).
38486 @item @r{Any stop packet}
38487 for success (@pxref{Stop Reply Packets})
38491 @cindex @samp{vStopped} packet
38492 @xref{Notification Packets}.
38494 @item X @var{addr},@var{length}:@var{XX@dots{}}
38496 @cindex @samp{X} packet
38497 Write data to memory, where the data is transmitted in binary.
38498 Memory is specified by its address @var{addr} and number of addressable memory
38499 units @var{length} (@pxref{addressable memory unit});
38500 @samp{@var{XX}@dots{}} is binary data (@pxref{Binary Data}).
38510 @item z @var{type},@var{addr},@var{kind}
38511 @itemx Z @var{type},@var{addr},@var{kind}
38512 @anchor{insert breakpoint or watchpoint packet}
38513 @cindex @samp{z} packet
38514 @cindex @samp{Z} packets
38515 Insert (@samp{Z}) or remove (@samp{z}) a @var{type} breakpoint or
38516 watchpoint starting at address @var{address} of kind @var{kind}.
38518 Each breakpoint and watchpoint packet @var{type} is documented
38521 @emph{Implementation notes: A remote target shall return an empty string
38522 for an unrecognized breakpoint or watchpoint packet @var{type}. A
38523 remote target shall support either both or neither of a given
38524 @samp{Z@var{type}@dots{}} and @samp{z@var{type}@dots{}} packet pair. To
38525 avoid potential problems with duplicate packets, the operations should
38526 be implemented in an idempotent way.}
38528 @item z0,@var{addr},@var{kind}
38529 @itemx Z0,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
38530 @cindex @samp{z0} packet
38531 @cindex @samp{Z0} packet
38532 Insert (@samp{Z0}) or remove (@samp{z0}) a software breakpoint at address
38533 @var{addr} of type @var{kind}.
38535 A software breakpoint is implemented by replacing the instruction at
38536 @var{addr} with a software breakpoint or trap instruction. The
38537 @var{kind} is target-specific and typically indicates the size of the
38538 breakpoint in bytes that should be inserted. E.g., the @sc{arm} and
38539 @sc{mips} can insert either a 2 or 4 byte breakpoint. Some
38540 architectures have additional meanings for @var{kind}
38541 (@pxref{Architecture-Specific Protocol Details}); if no
38542 architecture-specific value is being used, it should be @samp{0}.
38543 @var{kind} is hex-encoded. @var{cond_list} is an optional list of
38544 conditional expressions in bytecode form that should be evaluated on
38545 the target's side. These are the conditions that should be taken into
38546 consideration when deciding if the breakpoint trigger should be
38547 reported back to @value{GDBN}.
38549 See also the @samp{swbreak} stop reason (@pxref{swbreak stop reason})
38550 for how to best report a software breakpoint event to @value{GDBN}.
38552 The @var{cond_list} parameter is comprised of a series of expressions,
38553 concatenated without separators. Each expression has the following form:
38557 @item X @var{len},@var{expr}
38558 @var{len} is the length of the bytecode expression and @var{expr} is the
38559 actual conditional expression in bytecode form.
38563 The optional @var{cmd_list} parameter introduces commands that may be
38564 run on the target, rather than being reported back to @value{GDBN}.
38565 The parameter starts with a numeric flag @var{persist}; if the flag is
38566 nonzero, then the breakpoint may remain active and the commands
38567 continue to be run even when @value{GDBN} disconnects from the target.
38568 Following this flag is a series of expressions concatenated with no
38569 separators. Each expression has the following form:
38573 @item X @var{len},@var{expr}
38574 @var{len} is the length of the bytecode expression and @var{expr} is the
38575 actual commands expression in bytecode form.
38579 @emph{Implementation note: It is possible for a target to copy or move
38580 code that contains software breakpoints (e.g., when implementing
38581 overlays). The behavior of this packet, in the presence of such a
38582 target, is not defined.}
38594 @item z1,@var{addr},@var{kind}
38595 @itemx Z1,@var{addr},@var{kind}@r{[};@var{cond_list}@dots{}@r{]}@r{[};cmds:@var{persist},@var{cmd_list}@dots{}@r{]}
38596 @cindex @samp{z1} packet
38597 @cindex @samp{Z1} packet
38598 Insert (@samp{Z1}) or remove (@samp{z1}) a hardware breakpoint at
38599 address @var{addr}.
38601 A hardware breakpoint is implemented using a mechanism that is not
38602 dependent on being able to modify the target's memory. The
38603 @var{kind}, @var{cond_list}, and @var{cmd_list} arguments have the
38604 same meaning as in @samp{Z0} packets.
38606 @emph{Implementation note: A hardware breakpoint is not affected by code
38619 @item z2,@var{addr},@var{kind}
38620 @itemx Z2,@var{addr},@var{kind}
38621 @cindex @samp{z2} packet
38622 @cindex @samp{Z2} packet
38623 Insert (@samp{Z2}) or remove (@samp{z2}) a write watchpoint at @var{addr}.
38624 The number of bytes to watch is specified by @var{kind}.
38636 @item z3,@var{addr},@var{kind}
38637 @itemx Z3,@var{addr},@var{kind}
38638 @cindex @samp{z3} packet
38639 @cindex @samp{Z3} packet
38640 Insert (@samp{Z3}) or remove (@samp{z3}) a read watchpoint at @var{addr}.
38641 The number of bytes to watch is specified by @var{kind}.
38653 @item z4,@var{addr},@var{kind}
38654 @itemx Z4,@var{addr},@var{kind}
38655 @cindex @samp{z4} packet
38656 @cindex @samp{Z4} packet
38657 Insert (@samp{Z4}) or remove (@samp{z4}) an access watchpoint at @var{addr}.
38658 The number of bytes to watch is specified by @var{kind}.
38672 @node Stop Reply Packets
38673 @section Stop Reply Packets
38674 @cindex stop reply packets
38676 The @samp{C}, @samp{c}, @samp{S}, @samp{s}, @samp{vCont},
38677 @samp{vAttach}, @samp{vRun}, @samp{vStopped}, and @samp{?} packets can
38678 receive any of the below as a reply. Except for @samp{?}
38679 and @samp{vStopped}, that reply is only returned
38680 when the target halts. In the below the exact meaning of @dfn{signal
38681 number} is defined by the header @file{include/gdb/signals.h} in the
38682 @value{GDBN} source code.
38684 In non-stop mode, the server will simply reply @samp{OK} to commands
38685 such as @samp{vCont}; any stop will be the subject of a future
38686 notification. @xref{Remote Non-Stop}.
38688 As in the description of request packets, we include spaces in the
38689 reply templates for clarity; these are not part of the reply packet's
38690 syntax. No @value{GDBN} stop reply packet uses spaces to separate its
38696 The program received signal number @var{AA} (a two-digit hexadecimal
38697 number). This is equivalent to a @samp{T} response with no
38698 @var{n}:@var{r} pairs.
38700 @item T @var{AA} @var{n1}:@var{r1};@var{n2}:@var{r2};@dots{}
38701 @cindex @samp{T} packet reply
38702 The program received signal number @var{AA} (a two-digit hexadecimal
38703 number). This is equivalent to an @samp{S} response, except that the
38704 @samp{@var{n}:@var{r}} pairs can carry values of important registers
38705 and other information directly in the stop reply packet, reducing
38706 round-trip latency. Single-step and breakpoint traps are reported
38707 this way. Each @samp{@var{n}:@var{r}} pair is interpreted as follows:
38711 If @var{n} is a hexadecimal number, it is a register number, and the
38712 corresponding @var{r} gives that register's value. The data @var{r} is a
38713 series of bytes in target byte order, with each byte given by a
38714 two-digit hex number.
38717 If @var{n} is @samp{thread}, then @var{r} is the @var{thread-id} of
38718 the stopped thread, as specified in @ref{thread-id syntax}.
38721 If @var{n} is @samp{core}, then @var{r} is the hexadecimal number of
38722 the core on which the stop event was detected.
38725 If @var{n} is a recognized @dfn{stop reason}, it describes a more
38726 specific event that stopped the target. The currently defined stop
38727 reasons are listed below. The @var{aa} should be @samp{05}, the trap
38728 signal. At most one stop reason should be present.
38731 Otherwise, @value{GDBN} should ignore this @samp{@var{n}:@var{r}} pair
38732 and go on to the next; this allows us to extend the protocol in the
38736 The currently defined stop reasons are:
38742 The packet indicates a watchpoint hit, and @var{r} is the data address, in
38745 @item syscall_entry
38746 @itemx syscall_return
38747 The packet indicates a syscall entry or return, and @var{r} is the
38748 syscall number, in hex.
38750 @cindex shared library events, remote reply
38752 The packet indicates that the loaded libraries have changed.
38753 @value{GDBN} should use @samp{qXfer:libraries:read} to fetch a new
38754 list of loaded libraries. The @var{r} part is ignored.
38756 @cindex replay log events, remote reply
38758 The packet indicates that the target cannot continue replaying
38759 logged execution events, because it has reached the end (or the
38760 beginning when executing backward) of the log. The value of @var{r}
38761 will be either @samp{begin} or @samp{end}. @xref{Reverse Execution},
38762 for more information.
38765 @anchor{swbreak stop reason}
38766 The packet indicates a software breakpoint instruction was executed,
38767 irrespective of whether it was @value{GDBN} that planted the
38768 breakpoint or the breakpoint is hardcoded in the program. The @var{r}
38769 part must be left empty.
38771 On some architectures, such as x86, at the architecture level, when a
38772 breakpoint instruction executes the program counter points at the
38773 breakpoint address plus an offset. On such targets, the stub is
38774 responsible for adjusting the PC to point back at the breakpoint
38777 This packet should not be sent by default; older @value{GDBN} versions
38778 did not support it. @value{GDBN} requests it, by supplying an
38779 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38780 remote stub must also supply the appropriate @samp{qSupported} feature
38781 indicating support.
38783 This packet is required for correct non-stop mode operation.
38786 The packet indicates the target stopped for a hardware breakpoint.
38787 The @var{r} part must be left empty.
38789 The same remarks about @samp{qSupported} and non-stop mode above
38792 @cindex fork events, remote reply
38794 The packet indicates that @code{fork} was called, and @var{r}
38795 is the thread ID of the new child process. Refer to
38796 @ref{thread-id syntax} for the format of the @var{thread-id}
38797 field. This packet is only applicable to targets that support
38800 This packet should not be sent by default; older @value{GDBN} versions
38801 did not support it. @value{GDBN} requests it, by supplying an
38802 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38803 remote stub must also supply the appropriate @samp{qSupported} feature
38804 indicating support.
38806 @cindex vfork events, remote reply
38808 The packet indicates that @code{vfork} was called, and @var{r}
38809 is the thread ID of the new child process. Refer to
38810 @ref{thread-id syntax} for the format of the @var{thread-id}
38811 field. This packet is only applicable to targets that support
38814 This packet should not be sent by default; older @value{GDBN} versions
38815 did not support it. @value{GDBN} requests it, by supplying an
38816 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38817 remote stub must also supply the appropriate @samp{qSupported} feature
38818 indicating support.
38820 @cindex vforkdone events, remote reply
38822 The packet indicates that a child process created by a vfork
38823 has either called @code{exec} or terminated, so that the
38824 address spaces of the parent and child process are no longer
38825 shared. The @var{r} part is ignored. This packet is only
38826 applicable to targets that support vforkdone events.
38828 This packet should not be sent by default; older @value{GDBN} versions
38829 did not support it. @value{GDBN} requests it, by supplying an
38830 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38831 remote stub must also supply the appropriate @samp{qSupported} feature
38832 indicating support.
38834 @cindex exec events, remote reply
38836 The packet indicates that @code{execve} was called, and @var{r}
38837 is the absolute pathname of the file that was executed, in hex.
38838 This packet is only applicable to targets that support exec events.
38840 This packet should not be sent by default; older @value{GDBN} versions
38841 did not support it. @value{GDBN} requests it, by supplying an
38842 appropriate @samp{qSupported} feature (@pxref{qSupported}). The
38843 remote stub must also supply the appropriate @samp{qSupported} feature
38844 indicating support.
38846 @cindex thread create event, remote reply
38847 @anchor{thread create event}
38849 The packet indicates that the thread was just created. The new thread
38850 is stopped until @value{GDBN} sets it running with a resumption packet
38851 (@pxref{vCont packet}). This packet should not be sent by default;
38852 @value{GDBN} requests it with the @ref{QThreadEvents} packet. See
38853 also the @samp{w} (@pxref{thread exit event}) remote reply below. The
38854 @var{r} part is ignored.
38859 @itemx W @var{AA} ; process:@var{pid}
38860 The process exited, and @var{AA} is the exit status. This is only
38861 applicable to certain targets.
38863 The second form of the response, including the process ID of the
38864 exited process, can be used only when @value{GDBN} has reported
38865 support for multiprocess protocol extensions; see @ref{multiprocess
38866 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
38870 @itemx X @var{AA} ; process:@var{pid}
38871 The process terminated with signal @var{AA}.
38873 The second form of the response, including the process ID of the
38874 terminated process, can be used only when @value{GDBN} has reported
38875 support for multiprocess protocol extensions; see @ref{multiprocess
38876 extensions}. Both @var{AA} and @var{pid} are formatted as big-endian
38879 @anchor{thread exit event}
38880 @cindex thread exit event, remote reply
38881 @item w @var{AA} ; @var{tid}
38883 The thread exited, and @var{AA} is the exit status. This response
38884 should not be sent by default; @value{GDBN} requests it with the
38885 @ref{QThreadEvents} packet. See also @ref{thread create event} above.
38886 @var{AA} is formatted as a big-endian hex string.
38889 There are no resumed threads left in the target. In other words, even
38890 though the process is alive, the last resumed thread has exited. For
38891 example, say the target process has two threads: thread 1 and thread
38892 2. The client leaves thread 1 stopped, and resumes thread 2, which
38893 subsequently exits. At this point, even though the process is still
38894 alive, and thus no @samp{W} stop reply is sent, no thread is actually
38895 executing either. The @samp{N} stop reply thus informs the client
38896 that it can stop waiting for stop replies. This packet should not be
38897 sent by default; older @value{GDBN} versions did not support it.
38898 @value{GDBN} requests it, by supplying an appropriate
38899 @samp{qSupported} feature (@pxref{qSupported}). The remote stub must
38900 also supply the appropriate @samp{qSupported} feature indicating
38903 @item O @var{XX}@dots{}
38904 @samp{@var{XX}@dots{}} is hex encoding of @sc{ascii} data, to be
38905 written as the program's console output. This can happen at any time
38906 while the program is running and the debugger should continue to wait
38907 for @samp{W}, @samp{T}, etc. This reply is not permitted in non-stop mode.
38909 @item F @var{call-id},@var{parameter}@dots{}
38910 @var{call-id} is the identifier which says which host system call should
38911 be called. This is just the name of the function. Translation into the
38912 correct system call is only applicable as it's defined in @value{GDBN}.
38913 @xref{File-I/O Remote Protocol Extension}, for a list of implemented
38916 @samp{@var{parameter}@dots{}} is a list of parameters as defined for
38917 this very system call.
38919 The target replies with this packet when it expects @value{GDBN} to
38920 call a host system call on behalf of the target. @value{GDBN} replies
38921 with an appropriate @samp{F} packet and keeps up waiting for the next
38922 reply packet from the target. The latest @samp{C}, @samp{c}, @samp{S}
38923 or @samp{s} action is expected to be continued. @xref{File-I/O Remote
38924 Protocol Extension}, for more details.
38928 @node General Query Packets
38929 @section General Query Packets
38930 @cindex remote query requests
38932 Packets starting with @samp{q} are @dfn{general query packets};
38933 packets starting with @samp{Q} are @dfn{general set packets}. General
38934 query and set packets are a semi-unified form for retrieving and
38935 sending information to and from the stub.
38937 The initial letter of a query or set packet is followed by a name
38938 indicating what sort of thing the packet applies to. For example,
38939 @value{GDBN} may use a @samp{qSymbol} packet to exchange symbol
38940 definitions with the stub. These packet names follow some
38945 The name must not contain commas, colons or semicolons.
38947 Most @value{GDBN} query and set packets have a leading upper case
38950 The names of custom vendor packets should use a company prefix, in
38951 lower case, followed by a period. For example, packets designed at
38952 the Acme Corporation might begin with @samp{qacme.foo} (for querying
38953 foos) or @samp{Qacme.bar} (for setting bars).
38956 The name of a query or set packet should be separated from any
38957 parameters by a @samp{:}; the parameters themselves should be
38958 separated by @samp{,} or @samp{;}. Stubs must be careful to match the
38959 full packet name, and check for a separator or the end of the packet,
38960 in case two packet names share a common prefix. New packets should not begin
38961 with @samp{qC}, @samp{qP}, or @samp{qL}@footnote{The @samp{qP} and @samp{qL}
38962 packets predate these conventions, and have arguments without any terminator
38963 for the packet name; we suspect they are in widespread use in places that
38964 are difficult to upgrade. The @samp{qC} packet has no arguments, but some
38965 existing stubs (e.g.@: RedBoot) are known to not check for the end of the
38968 Like the descriptions of the other packets, each description here
38969 has a template showing the packet's overall syntax, followed by an
38970 explanation of the packet's meaning. We include spaces in some of the
38971 templates for clarity; these are not part of the packet's syntax. No
38972 @value{GDBN} packet uses spaces to separate its components.
38974 Here are the currently defined query and set packets:
38980 Turn on or off the agent as a helper to perform some debugging operations
38981 delegated from @value{GDBN} (@pxref{Control Agent}).
38983 @item QAllow:@var{op}:@var{val}@dots{}
38984 @cindex @samp{QAllow} packet
38985 Specify which operations @value{GDBN} expects to request of the
38986 target, as a semicolon-separated list of operation name and value
38987 pairs. Possible values for @var{op} include @samp{WriteReg},
38988 @samp{WriteMem}, @samp{InsertBreak}, @samp{InsertTrace},
38989 @samp{InsertFastTrace}, and @samp{Stop}. @var{val} is either 0,
38990 indicating that @value{GDBN} will not request the operation, or 1,
38991 indicating that it may. (The target can then use this to set up its
38992 own internals optimally, for instance if the debugger never expects to
38993 insert breakpoints, it may not need to install its own trap handler.)
38996 @cindex current thread, remote request
38997 @cindex @samp{qC} packet
38998 Return the current thread ID.
39002 @item QC @var{thread-id}
39003 Where @var{thread-id} is a thread ID as documented in
39004 @ref{thread-id syntax}.
39005 @item @r{(anything else)}
39006 Any other reply implies the old thread ID.
39009 @item qCRC:@var{addr},@var{length}
39010 @cindex CRC of memory block, remote request
39011 @cindex @samp{qCRC} packet
39012 @anchor{qCRC packet}
39013 Compute the CRC checksum of a block of memory using CRC-32 defined in
39014 IEEE 802.3. The CRC is computed byte at a time, taking the most
39015 significant bit of each byte first. The initial pattern code
39016 @code{0xffffffff} is used to ensure leading zeros affect the CRC.
39018 @emph{Note:} This is the same CRC used in validating separate debug
39019 files (@pxref{Separate Debug Files, , Debugging Information in Separate
39020 Files}). However the algorithm is slightly different. When validating
39021 separate debug files, the CRC is computed taking the @emph{least}
39022 significant bit of each byte first, and the final result is inverted to
39023 detect trailing zeros.
39028 An error (such as memory fault)
39029 @item C @var{crc32}
39030 The specified memory region's checksum is @var{crc32}.
39033 @item QDisableRandomization:@var{value}
39034 @cindex disable address space randomization, remote request
39035 @cindex @samp{QDisableRandomization} packet
39036 Some target operating systems will randomize the virtual address space
39037 of the inferior process as a security feature, but provide a feature
39038 to disable such randomization, e.g.@: to allow for a more deterministic
39039 debugging experience. On such systems, this packet with a @var{value}
39040 of 1 directs the target to disable address space randomization for
39041 processes subsequently started via @samp{vRun} packets, while a packet
39042 with a @var{value} of 0 tells the target to enable address space
39045 This packet is only available in extended mode (@pxref{extended mode}).
39050 The request succeeded.
39053 An error occurred. The error number @var{nn} is given as hex digits.
39056 An empty reply indicates that @samp{QDisableRandomization} is not supported
39060 This packet is not probed by default; the remote stub must request it,
39061 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39062 This should only be done on targets that actually support disabling
39063 address space randomization.
39065 @item QStartupWithShell:@var{value}
39066 @cindex startup with shell, remote request
39067 @cindex @samp{QStartupWithShell} packet
39068 On UNIX-like targets, it is possible to start the inferior using a
39069 shell program. This is the default behavior on both @value{GDBN} and
39070 @command{gdbserver} (@pxref{set startup-with-shell}). This packet is
39071 used to inform @command{gdbserver} whether it should start the
39072 inferior using a shell or not.
39074 If @var{value} is @samp{0}, @command{gdbserver} will not use a shell
39075 to start the inferior. If @var{value} is @samp{1},
39076 @command{gdbserver} will use a shell to start the inferior. All other
39077 values are considered an error.
39079 This packet is only available in extended mode (@pxref{extended
39085 The request succeeded.
39088 An error occurred. The error number @var{nn} is given as hex digits.
39091 This packet is not probed by default; the remote stub must request it,
39092 by supplying an appropriate @samp{qSupported} response
39093 (@pxref{qSupported}). This should only be done on targets that
39094 actually support starting the inferior using a shell.
39096 Use of this packet is controlled by the @code{set startup-with-shell}
39097 command; @pxref{set startup-with-shell}.
39099 @item QEnvironmentHexEncoded:@var{hex-value}
39100 @anchor{QEnvironmentHexEncoded}
39101 @cindex set environment variable, remote request
39102 @cindex @samp{QEnvironmentHexEncoded} packet
39103 On UNIX-like targets, it is possible to set environment variables that
39104 will be passed to the inferior during the startup process. This
39105 packet is used to inform @command{gdbserver} of an environment
39106 variable that has been defined by the user on @value{GDBN} (@pxref{set
39109 The packet is composed by @var{hex-value}, an hex encoded
39110 representation of the @var{name=value} format representing an
39111 environment variable. The name of the environment variable is
39112 represented by @var{name}, and the value to be assigned to the
39113 environment variable is represented by @var{value}. If the variable
39114 has no value (i.e., the value is @code{null}), then @var{value} will
39117 This packet is only available in extended mode (@pxref{extended
39123 The request succeeded.
39126 This packet is not probed by default; the remote stub must request it,
39127 by supplying an appropriate @samp{qSupported} response
39128 (@pxref{qSupported}). This should only be done on targets that
39129 actually support passing environment variables to the starting
39132 This packet is related to the @code{set environment} command;
39133 @pxref{set environment}.
39135 @item QEnvironmentUnset:@var{hex-value}
39136 @anchor{QEnvironmentUnset}
39137 @cindex unset environment variable, remote request
39138 @cindex @samp{QEnvironmentUnset} packet
39139 On UNIX-like targets, it is possible to unset environment variables
39140 before starting the inferior in the remote target. This packet is
39141 used to inform @command{gdbserver} of an environment variable that has
39142 been unset by the user on @value{GDBN} (@pxref{unset environment}).
39144 The packet is composed by @var{hex-value}, an hex encoded
39145 representation of the name of the environment variable to be unset.
39147 This packet is only available in extended mode (@pxref{extended
39153 The request succeeded.
39156 This packet is not probed by default; the remote stub must request it,
39157 by supplying an appropriate @samp{qSupported} response
39158 (@pxref{qSupported}). This should only be done on targets that
39159 actually support passing environment variables to the starting
39162 This packet is related to the @code{unset environment} command;
39163 @pxref{unset environment}.
39165 @item QEnvironmentReset
39166 @anchor{QEnvironmentReset}
39167 @cindex reset environment, remote request
39168 @cindex @samp{QEnvironmentReset} packet
39169 On UNIX-like targets, this packet is used to reset the state of
39170 environment variables in the remote target before starting the
39171 inferior. In this context, reset means unsetting all environment
39172 variables that were previously set by the user (i.e., were not
39173 initially present in the environment). It is sent to
39174 @command{gdbserver} before the @samp{QEnvironmentHexEncoded}
39175 (@pxref{QEnvironmentHexEncoded}) and the @samp{QEnvironmentUnset}
39176 (@pxref{QEnvironmentUnset}) packets.
39178 This packet is only available in extended mode (@pxref{extended
39184 The request succeeded.
39187 This packet is not probed by default; the remote stub must request it,
39188 by supplying an appropriate @samp{qSupported} response
39189 (@pxref{qSupported}). This should only be done on targets that
39190 actually support passing environment variables to the starting
39193 @item QSetWorkingDir:@r{[}@var{directory}@r{]}
39194 @anchor{QSetWorkingDir packet}
39195 @cindex set working directory, remote request
39196 @cindex @samp{QSetWorkingDir} packet
39197 This packet is used to inform the remote server of the intended
39198 current working directory for programs that are going to be executed.
39200 The packet is composed by @var{directory}, an hex encoded
39201 representation of the directory that the remote inferior will use as
39202 its current working directory. If @var{directory} is an empty string,
39203 the remote server should reset the inferior's current working
39204 directory to its original, empty value.
39206 This packet is only available in extended mode (@pxref{extended
39212 The request succeeded.
39216 @itemx qsThreadInfo
39217 @cindex list active threads, remote request
39218 @cindex @samp{qfThreadInfo} packet
39219 @cindex @samp{qsThreadInfo} packet
39220 Obtain a list of all active thread IDs from the target (OS). Since there
39221 may be too many active threads to fit into one reply packet, this query
39222 works iteratively: it may require more than one query/reply sequence to
39223 obtain the entire list of threads. The first query of the sequence will
39224 be the @samp{qfThreadInfo} query; subsequent queries in the
39225 sequence will be the @samp{qsThreadInfo} query.
39227 NOTE: This packet replaces the @samp{qL} query (see below).
39231 @item m @var{thread-id}
39233 @item m @var{thread-id},@var{thread-id}@dots{}
39234 a comma-separated list of thread IDs
39236 (lower case letter @samp{L}) denotes end of list.
39239 In response to each query, the target will reply with a list of one or
39240 more thread IDs, separated by commas.
39241 @value{GDBN} will respond to each reply with a request for more thread
39242 ids (using the @samp{qs} form of the query), until the target responds
39243 with @samp{l} (lower-case ell, for @dfn{last}).
39244 Refer to @ref{thread-id syntax}, for the format of the @var{thread-id}
39247 @emph{Note: @value{GDBN} will send the @code{qfThreadInfo} query during the
39248 initial connection with the remote target, and the very first thread ID
39249 mentioned in the reply will be stopped by @value{GDBN} in a subsequent
39250 message. Therefore, the stub should ensure that the first thread ID in
39251 the @code{qfThreadInfo} reply is suitable for being stopped by @value{GDBN}.}
39253 @item qGetTLSAddr:@var{thread-id},@var{offset},@var{lm}
39254 @cindex get thread-local storage address, remote request
39255 @cindex @samp{qGetTLSAddr} packet
39256 Fetch the address associated with thread local storage specified
39257 by @var{thread-id}, @var{offset}, and @var{lm}.
39259 @var{thread-id} is the thread ID associated with the
39260 thread for which to fetch the TLS address. @xref{thread-id syntax}.
39262 @var{offset} is the (big endian, hex encoded) offset associated with the
39263 thread local variable. (This offset is obtained from the debug
39264 information associated with the variable.)
39266 @var{lm} is the (big endian, hex encoded) OS/ABI-specific encoding of the
39267 load module associated with the thread local storage. For example,
39268 a @sc{gnu}/Linux system will pass the link map address of the shared
39269 object associated with the thread local storage under consideration.
39270 Other operating environments may choose to represent the load module
39271 differently, so the precise meaning of this parameter will vary.
39275 @item @var{XX}@dots{}
39276 Hex encoded (big endian) bytes representing the address of the thread
39277 local storage requested.
39280 An error occurred. The error number @var{nn} is given as hex digits.
39283 An empty reply indicates that @samp{qGetTLSAddr} is not supported by the stub.
39286 @item qGetTIBAddr:@var{thread-id}
39287 @cindex get thread information block address
39288 @cindex @samp{qGetTIBAddr} packet
39289 Fetch address of the Windows OS specific Thread Information Block.
39291 @var{thread-id} is the thread ID associated with the thread.
39295 @item @var{XX}@dots{}
39296 Hex encoded (big endian) bytes representing the linear address of the
39297 thread information block.
39300 An error occured. This means that either the thread was not found, or the
39301 address could not be retrieved.
39304 An empty reply indicates that @samp{qGetTIBAddr} is not supported by the stub.
39307 @item qL @var{startflag} @var{threadcount} @var{nextthread}
39308 Obtain thread information from RTOS. Where: @var{startflag} (one hex
39309 digit) is one to indicate the first query and zero to indicate a
39310 subsequent query; @var{threadcount} (two hex digits) is the maximum
39311 number of threads the response packet can contain; and @var{nextthread}
39312 (eight hex digits), for subsequent queries (@var{startflag} is zero), is
39313 returned in the response as @var{argthread}.
39315 Don't use this packet; use the @samp{qfThreadInfo} query instead (see above).
39319 @item qM @var{count} @var{done} @var{argthread} @var{thread}@dots{}
39320 Where: @var{count} (two hex digits) is the number of threads being
39321 returned; @var{done} (one hex digit) is zero to indicate more threads
39322 and one indicates no further threads; @var{argthreadid} (eight hex
39323 digits) is @var{nextthread} from the request packet; @var{thread}@dots{}
39324 is a sequence of thread IDs, @var{threadid} (eight hex
39325 digits), from the target. See @code{remote.c:parse_threadlist_response()}.
39329 @cindex section offsets, remote request
39330 @cindex @samp{qOffsets} packet
39331 Get section offsets that the target used when relocating the downloaded
39336 @item Text=@var{xxx};Data=@var{yyy}@r{[};Bss=@var{zzz}@r{]}
39337 Relocate the @code{Text} section by @var{xxx} from its original address.
39338 Relocate the @code{Data} section by @var{yyy} from its original address.
39339 If the object file format provides segment information (e.g.@: @sc{elf}
39340 @samp{PT_LOAD} program headers), @value{GDBN} will relocate entire
39341 segments by the supplied offsets.
39343 @emph{Note: while a @code{Bss} offset may be included in the response,
39344 @value{GDBN} ignores this and instead applies the @code{Data} offset
39345 to the @code{Bss} section.}
39347 @item TextSeg=@var{xxx}@r{[};DataSeg=@var{yyy}@r{]}
39348 Relocate the first segment of the object file, which conventionally
39349 contains program code, to a starting address of @var{xxx}. If
39350 @samp{DataSeg} is specified, relocate the second segment, which
39351 conventionally contains modifiable data, to a starting address of
39352 @var{yyy}. @value{GDBN} will report an error if the object file
39353 does not contain segment information, or does not contain at least
39354 as many segments as mentioned in the reply. Extra segments are
39355 kept at fixed offsets relative to the last relocated segment.
39358 @item qP @var{mode} @var{thread-id}
39359 @cindex thread information, remote request
39360 @cindex @samp{qP} packet
39361 Returns information on @var{thread-id}. Where: @var{mode} is a hex
39362 encoded 32 bit mode; @var{thread-id} is a thread ID
39363 (@pxref{thread-id syntax}).
39365 Don't use this packet; use the @samp{qThreadExtraInfo} query instead
39368 Reply: see @code{remote.c:remote_unpack_thread_info_response()}.
39372 @cindex non-stop mode, remote request
39373 @cindex @samp{QNonStop} packet
39375 Enter non-stop (@samp{QNonStop:1}) or all-stop (@samp{QNonStop:0}) mode.
39376 @xref{Remote Non-Stop}, for more information.
39381 The request succeeded.
39384 An error occurred. The error number @var{nn} is given as hex digits.
39387 An empty reply indicates that @samp{QNonStop} is not supported by
39391 This packet is not probed by default; the remote stub must request it,
39392 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39393 Use of this packet is controlled by the @code{set non-stop} command;
39394 @pxref{Non-Stop Mode}.
39396 @item QCatchSyscalls:1 @r{[};@var{sysno}@r{]}@dots{}
39397 @itemx QCatchSyscalls:0
39398 @cindex catch syscalls from inferior, remote request
39399 @cindex @samp{QCatchSyscalls} packet
39400 @anchor{QCatchSyscalls}
39401 Enable (@samp{QCatchSyscalls:1}) or disable (@samp{QCatchSyscalls:0})
39402 catching syscalls from the inferior process.
39404 For @samp{QCatchSyscalls:1}, each listed syscall @var{sysno} (encoded
39405 in hex) should be reported to @value{GDBN}. If no syscall @var{sysno}
39406 is listed, every system call should be reported.
39408 Note that if a syscall not in the list is reported, @value{GDBN} will
39409 still filter the event according to its own list from all corresponding
39410 @code{catch syscall} commands. However, it is more efficient to only
39411 report the requested syscalls.
39413 Multiple @samp{QCatchSyscalls:1} packets do not combine; any earlier
39414 @samp{QCatchSyscalls:1} list is completely replaced by the new list.
39416 If the inferior process execs, the state of @samp{QCatchSyscalls} is
39417 kept for the new process too. On targets where exec may affect syscall
39418 numbers, for example with exec between 32 and 64-bit processes, the
39419 client should send a new packet with the new syscall list.
39424 The request succeeded.
39427 An error occurred. @var{nn} are hex digits.
39430 An empty reply indicates that @samp{QCatchSyscalls} is not supported by
39434 Use of this packet is controlled by the @code{set remote catch-syscalls}
39435 command (@pxref{Remote Configuration, set remote catch-syscalls}).
39436 This packet is not probed by default; the remote stub must request it,
39437 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39439 @item QPassSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
39440 @cindex pass signals to inferior, remote request
39441 @cindex @samp{QPassSignals} packet
39442 @anchor{QPassSignals}
39443 Each listed @var{signal} should be passed directly to the inferior process.
39444 Signals are numbered identically to continue packets and stop replies
39445 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
39446 strictly greater than the previous item. These signals do not need to stop
39447 the inferior, or be reported to @value{GDBN}. All other signals should be
39448 reported to @value{GDBN}. Multiple @samp{QPassSignals} packets do not
39449 combine; any earlier @samp{QPassSignals} list is completely replaced by the
39450 new list. This packet improves performance when using @samp{handle
39451 @var{signal} nostop noprint pass}.
39456 The request succeeded.
39459 An error occurred. The error number @var{nn} is given as hex digits.
39462 An empty reply indicates that @samp{QPassSignals} is not supported by
39466 Use of this packet is controlled by the @code{set remote pass-signals}
39467 command (@pxref{Remote Configuration, set remote pass-signals}).
39468 This packet is not probed by default; the remote stub must request it,
39469 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39471 @item QProgramSignals: @var{signal} @r{[};@var{signal}@r{]}@dots{}
39472 @cindex signals the inferior may see, remote request
39473 @cindex @samp{QProgramSignals} packet
39474 @anchor{QProgramSignals}
39475 Each listed @var{signal} may be delivered to the inferior process.
39476 Others should be silently discarded.
39478 In some cases, the remote stub may need to decide whether to deliver a
39479 signal to the program or not without @value{GDBN} involvement. One
39480 example of that is while detaching --- the program's threads may have
39481 stopped for signals that haven't yet had a chance of being reported to
39482 @value{GDBN}, and so the remote stub can use the signal list specified
39483 by this packet to know whether to deliver or ignore those pending
39486 This does not influence whether to deliver a signal as requested by a
39487 resumption packet (@pxref{vCont packet}).
39489 Signals are numbered identically to continue packets and stop replies
39490 (@pxref{Stop Reply Packets}). Each @var{signal} list item should be
39491 strictly greater than the previous item. Multiple
39492 @samp{QProgramSignals} packets do not combine; any earlier
39493 @samp{QProgramSignals} list is completely replaced by the new list.
39498 The request succeeded.
39501 An error occurred. The error number @var{nn} is given as hex digits.
39504 An empty reply indicates that @samp{QProgramSignals} is not supported
39508 Use of this packet is controlled by the @code{set remote program-signals}
39509 command (@pxref{Remote Configuration, set remote program-signals}).
39510 This packet is not probed by default; the remote stub must request it,
39511 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
39513 @anchor{QThreadEvents}
39514 @item QThreadEvents:1
39515 @itemx QThreadEvents:0
39516 @cindex thread create/exit events, remote request
39517 @cindex @samp{QThreadEvents} packet
39519 Enable (@samp{QThreadEvents:1}) or disable (@samp{QThreadEvents:0})
39520 reporting of thread create and exit events. @xref{thread create
39521 event}, for the reply specifications. For example, this is used in
39522 non-stop mode when @value{GDBN} stops a set of threads and
39523 synchronously waits for the their corresponding stop replies. Without
39524 exit events, if one of the threads exits, @value{GDBN} would hang
39525 forever not knowing that it should no longer expect a stop for that
39526 same thread. @value{GDBN} does not enable this feature unless the
39527 stub reports that it supports it by including @samp{QThreadEvents+} in
39528 its @samp{qSupported} reply.
39533 The request succeeded.
39536 An error occurred. The error number @var{nn} is given as hex digits.
39539 An empty reply indicates that @samp{QThreadEvents} is not supported by
39543 Use of this packet is controlled by the @code{set remote thread-events}
39544 command (@pxref{Remote Configuration, set remote thread-events}).
39546 @item qRcmd,@var{command}
39547 @cindex execute remote command, remote request
39548 @cindex @samp{qRcmd} packet
39549 @var{command} (hex encoded) is passed to the local interpreter for
39550 execution. Invalid commands should be reported using the output
39551 string. Before the final result packet, the target may also respond
39552 with a number of intermediate @samp{O@var{output}} console output
39553 packets. @emph{Implementors should note that providing access to a
39554 stubs's interpreter may have security implications}.
39559 A command response with no output.
39561 A command response with the hex encoded output string @var{OUTPUT}.
39563 Indicate a badly formed request.
39565 An empty reply indicates that @samp{qRcmd} is not recognized.
39568 (Note that the @code{qRcmd} packet's name is separated from the
39569 command by a @samp{,}, not a @samp{:}, contrary to the naming
39570 conventions above. Please don't use this packet as a model for new
39573 @item qSearch:memory:@var{address};@var{length};@var{search-pattern}
39574 @cindex searching memory, in remote debugging
39576 @cindex @samp{qSearch:memory} packet
39578 @cindex @samp{qSearch memory} packet
39579 @anchor{qSearch memory}
39580 Search @var{length} bytes at @var{address} for @var{search-pattern}.
39581 Both @var{address} and @var{length} are encoded in hex;
39582 @var{search-pattern} is a sequence of bytes, also hex encoded.
39587 The pattern was not found.
39589 The pattern was found at @var{address}.
39591 A badly formed request or an error was encountered while searching memory.
39593 An empty reply indicates that @samp{qSearch:memory} is not recognized.
39596 @item QStartNoAckMode
39597 @cindex @samp{QStartNoAckMode} packet
39598 @anchor{QStartNoAckMode}
39599 Request that the remote stub disable the normal @samp{+}/@samp{-}
39600 protocol acknowledgments (@pxref{Packet Acknowledgment}).
39605 The stub has switched to no-acknowledgment mode.
39606 @value{GDBN} acknowledges this reponse,
39607 but neither the stub nor @value{GDBN} shall send or expect further
39608 @samp{+}/@samp{-} acknowledgments in the current connection.
39610 An empty reply indicates that the stub does not support no-acknowledgment mode.
39613 @item qSupported @r{[}:@var{gdbfeature} @r{[};@var{gdbfeature}@r{]}@dots{} @r{]}
39614 @cindex supported packets, remote query
39615 @cindex features of the remote protocol
39616 @cindex @samp{qSupported} packet
39617 @anchor{qSupported}
39618 Tell the remote stub about features supported by @value{GDBN}, and
39619 query the stub for features it supports. This packet allows
39620 @value{GDBN} and the remote stub to take advantage of each others'
39621 features. @samp{qSupported} also consolidates multiple feature probes
39622 at startup, to improve @value{GDBN} performance---a single larger
39623 packet performs better than multiple smaller probe packets on
39624 high-latency links. Some features may enable behavior which must not
39625 be on by default, e.g.@: because it would confuse older clients or
39626 stubs. Other features may describe packets which could be
39627 automatically probed for, but are not. These features must be
39628 reported before @value{GDBN} will use them. This ``default
39629 unsupported'' behavior is not appropriate for all packets, but it
39630 helps to keep the initial connection time under control with new
39631 versions of @value{GDBN} which support increasing numbers of packets.
39635 @item @var{stubfeature} @r{[};@var{stubfeature}@r{]}@dots{}
39636 The stub supports or does not support each returned @var{stubfeature},
39637 depending on the form of each @var{stubfeature} (see below for the
39640 An empty reply indicates that @samp{qSupported} is not recognized,
39641 or that no features needed to be reported to @value{GDBN}.
39644 The allowed forms for each feature (either a @var{gdbfeature} in the
39645 @samp{qSupported} packet, or a @var{stubfeature} in the response)
39649 @item @var{name}=@var{value}
39650 The remote protocol feature @var{name} is supported, and associated
39651 with the specified @var{value}. The format of @var{value} depends
39652 on the feature, but it must not include a semicolon.
39654 The remote protocol feature @var{name} is supported, and does not
39655 need an associated value.
39657 The remote protocol feature @var{name} is not supported.
39659 The remote protocol feature @var{name} may be supported, and
39660 @value{GDBN} should auto-detect support in some other way when it is
39661 needed. This form will not be used for @var{gdbfeature} notifications,
39662 but may be used for @var{stubfeature} responses.
39665 Whenever the stub receives a @samp{qSupported} request, the
39666 supplied set of @value{GDBN} features should override any previous
39667 request. This allows @value{GDBN} to put the stub in a known
39668 state, even if the stub had previously been communicating with
39669 a different version of @value{GDBN}.
39671 The following values of @var{gdbfeature} (for the packet sent by @value{GDBN})
39676 This feature indicates whether @value{GDBN} supports multiprocess
39677 extensions to the remote protocol. @value{GDBN} does not use such
39678 extensions unless the stub also reports that it supports them by
39679 including @samp{multiprocess+} in its @samp{qSupported} reply.
39680 @xref{multiprocess extensions}, for details.
39683 This feature indicates that @value{GDBN} supports the XML target
39684 description. If the stub sees @samp{xmlRegisters=} with target
39685 specific strings separated by a comma, it will report register
39689 This feature indicates whether @value{GDBN} supports the
39690 @samp{qRelocInsn} packet (@pxref{Tracepoint Packets,,Relocate
39691 instruction reply packet}).
39694 This feature indicates whether @value{GDBN} supports the swbreak stop
39695 reason in stop replies. @xref{swbreak stop reason}, for details.
39698 This feature indicates whether @value{GDBN} supports the hwbreak stop
39699 reason in stop replies. @xref{swbreak stop reason}, for details.
39702 This feature indicates whether @value{GDBN} supports fork event
39703 extensions to the remote protocol. @value{GDBN} does not use such
39704 extensions unless the stub also reports that it supports them by
39705 including @samp{fork-events+} in its @samp{qSupported} reply.
39708 This feature indicates whether @value{GDBN} supports vfork event
39709 extensions to the remote protocol. @value{GDBN} does not use such
39710 extensions unless the stub also reports that it supports them by
39711 including @samp{vfork-events+} in its @samp{qSupported} reply.
39714 This feature indicates whether @value{GDBN} supports exec event
39715 extensions to the remote protocol. @value{GDBN} does not use such
39716 extensions unless the stub also reports that it supports them by
39717 including @samp{exec-events+} in its @samp{qSupported} reply.
39719 @item vContSupported
39720 This feature indicates whether @value{GDBN} wants to know the
39721 supported actions in the reply to @samp{vCont?} packet.
39724 Stubs should ignore any unknown values for
39725 @var{gdbfeature}. Any @value{GDBN} which sends a @samp{qSupported}
39726 packet supports receiving packets of unlimited length (earlier
39727 versions of @value{GDBN} may reject overly long responses). Additional values
39728 for @var{gdbfeature} may be defined in the future to let the stub take
39729 advantage of new features in @value{GDBN}, e.g.@: incompatible
39730 improvements in the remote protocol---the @samp{multiprocess} feature is
39731 an example of such a feature. The stub's reply should be independent
39732 of the @var{gdbfeature} entries sent by @value{GDBN}; first @value{GDBN}
39733 describes all the features it supports, and then the stub replies with
39734 all the features it supports.
39736 Similarly, @value{GDBN} will silently ignore unrecognized stub feature
39737 responses, as long as each response uses one of the standard forms.
39739 Some features are flags. A stub which supports a flag feature
39740 should respond with a @samp{+} form response. Other features
39741 require values, and the stub should respond with an @samp{=}
39744 Each feature has a default value, which @value{GDBN} will use if
39745 @samp{qSupported} is not available or if the feature is not mentioned
39746 in the @samp{qSupported} response. The default values are fixed; a
39747 stub is free to omit any feature responses that match the defaults.
39749 Not all features can be probed, but for those which can, the probing
39750 mechanism is useful: in some cases, a stub's internal
39751 architecture may not allow the protocol layer to know some information
39752 about the underlying target in advance. This is especially common in
39753 stubs which may be configured for multiple targets.
39755 These are the currently defined stub features and their properties:
39757 @multitable @columnfractions 0.35 0.2 0.12 0.2
39758 @c NOTE: The first row should be @headitem, but we do not yet require
39759 @c a new enough version of Texinfo (4.7) to use @headitem.
39761 @tab Value Required
39765 @item @samp{PacketSize}
39770 @item @samp{qXfer:auxv:read}
39775 @item @samp{qXfer:btrace:read}
39780 @item @samp{qXfer:btrace-conf:read}
39785 @item @samp{qXfer:exec-file:read}
39790 @item @samp{qXfer:features:read}
39795 @item @samp{qXfer:libraries:read}
39800 @item @samp{qXfer:libraries-svr4:read}
39805 @item @samp{augmented-libraries-svr4-read}
39810 @item @samp{qXfer:memory-map:read}
39815 @item @samp{qXfer:sdata:read}
39820 @item @samp{qXfer:spu:read}
39825 @item @samp{qXfer:spu:write}
39830 @item @samp{qXfer:siginfo:read}
39835 @item @samp{qXfer:siginfo:write}
39840 @item @samp{qXfer:threads:read}
39845 @item @samp{qXfer:traceframe-info:read}
39850 @item @samp{qXfer:uib:read}
39855 @item @samp{qXfer:fdpic:read}
39860 @item @samp{Qbtrace:off}
39865 @item @samp{Qbtrace:bts}
39870 @item @samp{Qbtrace:pt}
39875 @item @samp{Qbtrace-conf:bts:size}
39880 @item @samp{Qbtrace-conf:pt:size}
39885 @item @samp{QNonStop}
39890 @item @samp{QCatchSyscalls}
39895 @item @samp{QPassSignals}
39900 @item @samp{QStartNoAckMode}
39905 @item @samp{multiprocess}
39910 @item @samp{ConditionalBreakpoints}
39915 @item @samp{ConditionalTracepoints}
39920 @item @samp{ReverseContinue}
39925 @item @samp{ReverseStep}
39930 @item @samp{TracepointSource}
39935 @item @samp{QAgent}
39940 @item @samp{QAllow}
39945 @item @samp{QDisableRandomization}
39950 @item @samp{EnableDisableTracepoints}
39955 @item @samp{QTBuffer:size}
39960 @item @samp{tracenz}
39965 @item @samp{BreakpointCommands}
39970 @item @samp{swbreak}
39975 @item @samp{hwbreak}
39980 @item @samp{fork-events}
39985 @item @samp{vfork-events}
39990 @item @samp{exec-events}
39995 @item @samp{QThreadEvents}
40000 @item @samp{no-resumed}
40007 These are the currently defined stub features, in more detail:
40010 @cindex packet size, remote protocol
40011 @item PacketSize=@var{bytes}
40012 The remote stub can accept packets up to at least @var{bytes} in
40013 length. @value{GDBN} will send packets up to this size for bulk
40014 transfers, and will never send larger packets. This is a limit on the
40015 data characters in the packet, including the frame and checksum.
40016 There is no trailing NUL byte in a remote protocol packet; if the stub
40017 stores packets in a NUL-terminated format, it should allow an extra
40018 byte in its buffer for the NUL. If this stub feature is not supported,
40019 @value{GDBN} guesses based on the size of the @samp{g} packet response.
40021 @item qXfer:auxv:read
40022 The remote stub understands the @samp{qXfer:auxv:read} packet
40023 (@pxref{qXfer auxiliary vector read}).
40025 @item qXfer:btrace:read
40026 The remote stub understands the @samp{qXfer:btrace:read}
40027 packet (@pxref{qXfer btrace read}).
40029 @item qXfer:btrace-conf:read
40030 The remote stub understands the @samp{qXfer:btrace-conf:read}
40031 packet (@pxref{qXfer btrace-conf read}).
40033 @item qXfer:exec-file:read
40034 The remote stub understands the @samp{qXfer:exec-file:read} packet
40035 (@pxref{qXfer executable filename read}).
40037 @item qXfer:features:read
40038 The remote stub understands the @samp{qXfer:features:read} packet
40039 (@pxref{qXfer target description read}).
40041 @item qXfer:libraries:read
40042 The remote stub understands the @samp{qXfer:libraries:read} packet
40043 (@pxref{qXfer library list read}).
40045 @item qXfer:libraries-svr4:read
40046 The remote stub understands the @samp{qXfer:libraries-svr4:read} packet
40047 (@pxref{qXfer svr4 library list read}).
40049 @item augmented-libraries-svr4-read
40050 The remote stub understands the augmented form of the
40051 @samp{qXfer:libraries-svr4:read} packet
40052 (@pxref{qXfer svr4 library list read}).
40054 @item qXfer:memory-map:read
40055 The remote stub understands the @samp{qXfer:memory-map:read} packet
40056 (@pxref{qXfer memory map read}).
40058 @item qXfer:sdata:read
40059 The remote stub understands the @samp{qXfer:sdata:read} packet
40060 (@pxref{qXfer sdata read}).
40062 @item qXfer:spu:read
40063 The remote stub understands the @samp{qXfer:spu:read} packet
40064 (@pxref{qXfer spu read}).
40066 @item qXfer:spu:write
40067 The remote stub understands the @samp{qXfer:spu:write} packet
40068 (@pxref{qXfer spu write}).
40070 @item qXfer:siginfo:read
40071 The remote stub understands the @samp{qXfer:siginfo:read} packet
40072 (@pxref{qXfer siginfo read}).
40074 @item qXfer:siginfo:write
40075 The remote stub understands the @samp{qXfer:siginfo:write} packet
40076 (@pxref{qXfer siginfo write}).
40078 @item qXfer:threads:read
40079 The remote stub understands the @samp{qXfer:threads:read} packet
40080 (@pxref{qXfer threads read}).
40082 @item qXfer:traceframe-info:read
40083 The remote stub understands the @samp{qXfer:traceframe-info:read}
40084 packet (@pxref{qXfer traceframe info read}).
40086 @item qXfer:uib:read
40087 The remote stub understands the @samp{qXfer:uib:read}
40088 packet (@pxref{qXfer unwind info block}).
40090 @item qXfer:fdpic:read
40091 The remote stub understands the @samp{qXfer:fdpic:read}
40092 packet (@pxref{qXfer fdpic loadmap read}).
40095 The remote stub understands the @samp{QNonStop} packet
40096 (@pxref{QNonStop}).
40098 @item QCatchSyscalls
40099 The remote stub understands the @samp{QCatchSyscalls} packet
40100 (@pxref{QCatchSyscalls}).
40103 The remote stub understands the @samp{QPassSignals} packet
40104 (@pxref{QPassSignals}).
40106 @item QStartNoAckMode
40107 The remote stub understands the @samp{QStartNoAckMode} packet and
40108 prefers to operate in no-acknowledgment mode. @xref{Packet Acknowledgment}.
40111 @anchor{multiprocess extensions}
40112 @cindex multiprocess extensions, in remote protocol
40113 The remote stub understands the multiprocess extensions to the remote
40114 protocol syntax. The multiprocess extensions affect the syntax of
40115 thread IDs in both packets and replies (@pxref{thread-id syntax}), and
40116 add process IDs to the @samp{D} packet and @samp{W} and @samp{X}
40117 replies. Note that reporting this feature indicates support for the
40118 syntactic extensions only, not that the stub necessarily supports
40119 debugging of more than one process at a time. The stub must not use
40120 multiprocess extensions in packet replies unless @value{GDBN} has also
40121 indicated it supports them in its @samp{qSupported} request.
40123 @item qXfer:osdata:read
40124 The remote stub understands the @samp{qXfer:osdata:read} packet
40125 ((@pxref{qXfer osdata read}).
40127 @item ConditionalBreakpoints
40128 The target accepts and implements evaluation of conditional expressions
40129 defined for breakpoints. The target will only report breakpoint triggers
40130 when such conditions are true (@pxref{Conditions, ,Break Conditions}).
40132 @item ConditionalTracepoints
40133 The remote stub accepts and implements conditional expressions defined
40134 for tracepoints (@pxref{Tracepoint Conditions}).
40136 @item ReverseContinue
40137 The remote stub accepts and implements the reverse continue packet
40141 The remote stub accepts and implements the reverse step packet
40144 @item TracepointSource
40145 The remote stub understands the @samp{QTDPsrc} packet that supplies
40146 the source form of tracepoint definitions.
40149 The remote stub understands the @samp{QAgent} packet.
40152 The remote stub understands the @samp{QAllow} packet.
40154 @item QDisableRandomization
40155 The remote stub understands the @samp{QDisableRandomization} packet.
40157 @item StaticTracepoint
40158 @cindex static tracepoints, in remote protocol
40159 The remote stub supports static tracepoints.
40161 @item InstallInTrace
40162 @anchor{install tracepoint in tracing}
40163 The remote stub supports installing tracepoint in tracing.
40165 @item EnableDisableTracepoints
40166 The remote stub supports the @samp{QTEnable} (@pxref{QTEnable}) and
40167 @samp{QTDisable} (@pxref{QTDisable}) packets that allow tracepoints
40168 to be enabled and disabled while a trace experiment is running.
40170 @item QTBuffer:size
40171 The remote stub supports the @samp{QTBuffer:size} (@pxref{QTBuffer-size})
40172 packet that allows to change the size of the trace buffer.
40175 @cindex string tracing, in remote protocol
40176 The remote stub supports the @samp{tracenz} bytecode for collecting strings.
40177 See @ref{Bytecode Descriptions} for details about the bytecode.
40179 @item BreakpointCommands
40180 @cindex breakpoint commands, in remote protocol
40181 The remote stub supports running a breakpoint's command list itself,
40182 rather than reporting the hit to @value{GDBN}.
40185 The remote stub understands the @samp{Qbtrace:off} packet.
40188 The remote stub understands the @samp{Qbtrace:bts} packet.
40191 The remote stub understands the @samp{Qbtrace:pt} packet.
40193 @item Qbtrace-conf:bts:size
40194 The remote stub understands the @samp{Qbtrace-conf:bts:size} packet.
40196 @item Qbtrace-conf:pt:size
40197 The remote stub understands the @samp{Qbtrace-conf:pt:size} packet.
40200 The remote stub reports the @samp{swbreak} stop reason for memory
40204 The remote stub reports the @samp{hwbreak} stop reason for hardware
40208 The remote stub reports the @samp{fork} stop reason for fork events.
40211 The remote stub reports the @samp{vfork} stop reason for vfork events
40212 and vforkdone events.
40215 The remote stub reports the @samp{exec} stop reason for exec events.
40217 @item vContSupported
40218 The remote stub reports the supported actions in the reply to
40219 @samp{vCont?} packet.
40221 @item QThreadEvents
40222 The remote stub understands the @samp{QThreadEvents} packet.
40225 The remote stub reports the @samp{N} stop reply.
40230 @cindex symbol lookup, remote request
40231 @cindex @samp{qSymbol} packet
40232 Notify the target that @value{GDBN} is prepared to serve symbol lookup
40233 requests. Accept requests from the target for the values of symbols.
40238 The target does not need to look up any (more) symbols.
40239 @item qSymbol:@var{sym_name}
40240 The target requests the value of symbol @var{sym_name} (hex encoded).
40241 @value{GDBN} may provide the value by using the
40242 @samp{qSymbol:@var{sym_value}:@var{sym_name}} message, described
40246 @item qSymbol:@var{sym_value}:@var{sym_name}
40247 Set the value of @var{sym_name} to @var{sym_value}.
40249 @var{sym_name} (hex encoded) is the name of a symbol whose value the
40250 target has previously requested.
40252 @var{sym_value} (hex) is the value for symbol @var{sym_name}. If
40253 @value{GDBN} cannot supply a value for @var{sym_name}, then this field
40259 The target does not need to look up any (more) symbols.
40260 @item qSymbol:@var{sym_name}
40261 The target requests the value of a new symbol @var{sym_name} (hex
40262 encoded). @value{GDBN} will continue to supply the values of symbols
40263 (if available), until the target ceases to request them.
40268 @itemx QTDisconnected
40275 @itemx qTMinFTPILen
40277 @xref{Tracepoint Packets}.
40279 @item qThreadExtraInfo,@var{thread-id}
40280 @cindex thread attributes info, remote request
40281 @cindex @samp{qThreadExtraInfo} packet
40282 Obtain from the target OS a printable string description of thread
40283 attributes for the thread @var{thread-id}; see @ref{thread-id syntax},
40284 for the forms of @var{thread-id}. This
40285 string may contain anything that the target OS thinks is interesting
40286 for @value{GDBN} to tell the user about the thread. The string is
40287 displayed in @value{GDBN}'s @code{info threads} display. Some
40288 examples of possible thread extra info strings are @samp{Runnable}, or
40289 @samp{Blocked on Mutex}.
40293 @item @var{XX}@dots{}
40294 Where @samp{@var{XX}@dots{}} is a hex encoding of @sc{ascii} data,
40295 comprising the printable string containing the extra information about
40296 the thread's attributes.
40299 (Note that the @code{qThreadExtraInfo} packet's name is separated from
40300 the command by a @samp{,}, not a @samp{:}, contrary to the naming
40301 conventions above. Please don't use this packet as a model for new
40320 @xref{Tracepoint Packets}.
40322 @item qXfer:@var{object}:read:@var{annex}:@var{offset},@var{length}
40323 @cindex read special object, remote request
40324 @cindex @samp{qXfer} packet
40325 @anchor{qXfer read}
40326 Read uninterpreted bytes from the target's special data area
40327 identified by the keyword @var{object}. Request @var{length} bytes
40328 starting at @var{offset} bytes into the data. The content and
40329 encoding of @var{annex} is specific to @var{object}; it can supply
40330 additional details about what data to access.
40335 Data @var{data} (@pxref{Binary Data}) has been read from the
40336 target. There may be more data at a higher address (although
40337 it is permitted to return @samp{m} even for the last valid
40338 block of data, as long as at least one byte of data was read).
40339 It is possible for @var{data} to have fewer bytes than the @var{length} in the
40343 Data @var{data} (@pxref{Binary Data}) has been read from the target.
40344 There is no more data to be read. It is possible for @var{data} to
40345 have fewer bytes than the @var{length} in the request.
40348 The @var{offset} in the request is at the end of the data.
40349 There is no more data to be read.
40352 The request was malformed, or @var{annex} was invalid.
40355 The offset was invalid, or there was an error encountered reading the data.
40356 The @var{nn} part is a hex-encoded @code{errno} value.
40359 An empty reply indicates the @var{object} string was not recognized by
40360 the stub, or that the object does not support reading.
40363 Here are the specific requests of this form defined so far. All the
40364 @samp{qXfer:@var{object}:read:@dots{}} requests use the same reply
40365 formats, listed above.
40368 @item qXfer:auxv:read::@var{offset},@var{length}
40369 @anchor{qXfer auxiliary vector read}
40370 Access the target's @dfn{auxiliary vector}. @xref{OS Information,
40371 auxiliary vector}. Note @var{annex} must be empty.
40373 This packet is not probed by default; the remote stub must request it,
40374 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40376 @item qXfer:btrace:read:@var{annex}:@var{offset},@var{length}
40377 @anchor{qXfer btrace read}
40379 Return a description of the current branch trace.
40380 @xref{Branch Trace Format}. The annex part of the generic @samp{qXfer}
40381 packet may have one of the following values:
40385 Returns all available branch trace.
40388 Returns all available branch trace if the branch trace changed since
40389 the last read request.
40392 Returns the new branch trace since the last read request. Adds a new
40393 block to the end of the trace that begins at zero and ends at the source
40394 location of the first branch in the trace buffer. This extra block is
40395 used to stitch traces together.
40397 If the trace buffer overflowed, returns an error indicating the overflow.
40400 This packet is not probed by default; the remote stub must request it
40401 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40403 @item qXfer:btrace-conf:read::@var{offset},@var{length}
40404 @anchor{qXfer btrace-conf read}
40406 Return a description of the current branch trace configuration.
40407 @xref{Branch Trace Configuration Format}.
40409 This packet is not probed by default; the remote stub must request it
40410 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40412 @item qXfer:exec-file:read:@var{annex}:@var{offset},@var{length}
40413 @anchor{qXfer executable filename read}
40414 Return the full absolute name of the file that was executed to create
40415 a process running on the remote system. The annex specifies the
40416 numeric process ID of the process to query, encoded as a hexadecimal
40417 number. If the annex part is empty the remote stub should return the
40418 filename corresponding to the currently executing process.
40420 This packet is not probed by default; the remote stub must request it,
40421 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40423 @item qXfer:features:read:@var{annex}:@var{offset},@var{length}
40424 @anchor{qXfer target description read}
40425 Access the @dfn{target description}. @xref{Target Descriptions}. The
40426 annex specifies which XML document to access. The main description is
40427 always loaded from the @samp{target.xml} annex.
40429 This packet is not probed by default; the remote stub must request it,
40430 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40432 @item qXfer:libraries:read:@var{annex}:@var{offset},@var{length}
40433 @anchor{qXfer library list read}
40434 Access the target's list of loaded libraries. @xref{Library List Format}.
40435 The annex part of the generic @samp{qXfer} packet must be empty
40436 (@pxref{qXfer read}).
40438 Targets which maintain a list of libraries in the program's memory do
40439 not need to implement this packet; it is designed for platforms where
40440 the operating system manages the list of loaded libraries.
40442 This packet is not probed by default; the remote stub must request it,
40443 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40445 @item qXfer:libraries-svr4:read:@var{annex}:@var{offset},@var{length}
40446 @anchor{qXfer svr4 library list read}
40447 Access the target's list of loaded libraries when the target is an SVR4
40448 platform. @xref{Library List Format for SVR4 Targets}. The annex part
40449 of the generic @samp{qXfer} packet must be empty unless the remote
40450 stub indicated it supports the augmented form of this packet
40451 by supplying an appropriate @samp{qSupported} response
40452 (@pxref{qXfer read}, @ref{qSupported}).
40454 This packet is optional for better performance on SVR4 targets.
40455 @value{GDBN} uses memory read packets to read the SVR4 library list otherwise.
40457 This packet is not probed by default; the remote stub must request it,
40458 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40460 If the remote stub indicates it supports the augmented form of this
40461 packet then the annex part of the generic @samp{qXfer} packet may
40462 contain a semicolon-separated list of @samp{@var{name}=@var{value}}
40463 arguments. The currently supported arguments are:
40466 @item start=@var{address}
40467 A hexadecimal number specifying the address of the @samp{struct
40468 link_map} to start reading the library list from. If unset or zero
40469 then the first @samp{struct link_map} in the library list will be
40470 chosen as the starting point.
40472 @item prev=@var{address}
40473 A hexadecimal number specifying the address of the @samp{struct
40474 link_map} immediately preceding the @samp{struct link_map}
40475 specified by the @samp{start} argument. If unset or zero then
40476 the remote stub will expect that no @samp{struct link_map}
40477 exists prior to the starting point.
40481 Arguments that are not understood by the remote stub will be silently
40484 @item qXfer:memory-map:read::@var{offset},@var{length}
40485 @anchor{qXfer memory map read}
40486 Access the target's @dfn{memory-map}. @xref{Memory Map Format}. The
40487 annex part of the generic @samp{qXfer} packet must be empty
40488 (@pxref{qXfer read}).
40490 This packet is not probed by default; the remote stub must request it,
40491 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40493 @item qXfer:sdata:read::@var{offset},@var{length}
40494 @anchor{qXfer sdata read}
40496 Read contents of the extra collected static tracepoint marker
40497 information. The annex part of the generic @samp{qXfer} packet must
40498 be empty (@pxref{qXfer read}). @xref{Tracepoint Actions,,Tracepoint
40501 This packet is not probed by default; the remote stub must request it,
40502 by supplying an appropriate @samp{qSupported} response
40503 (@pxref{qSupported}).
40505 @item qXfer:siginfo:read::@var{offset},@var{length}
40506 @anchor{qXfer siginfo read}
40507 Read contents of the extra signal information on the target
40508 system. The annex part of the generic @samp{qXfer} packet must be
40509 empty (@pxref{qXfer read}).
40511 This packet is not probed by default; the remote stub must request it,
40512 by supplying an appropriate @samp{qSupported} response
40513 (@pxref{qSupported}).
40515 @item qXfer:spu:read:@var{annex}:@var{offset},@var{length}
40516 @anchor{qXfer spu read}
40517 Read contents of an @code{spufs} file on the target system. The
40518 annex specifies which file to read; it must be of the form
40519 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
40520 in the target process, and @var{name} identifes the @code{spufs} file
40521 in that context to be accessed.
40523 This packet is not probed by default; the remote stub must request it,
40524 by supplying an appropriate @samp{qSupported} response
40525 (@pxref{qSupported}).
40527 @item qXfer:threads:read::@var{offset},@var{length}
40528 @anchor{qXfer threads read}
40529 Access the list of threads on target. @xref{Thread List Format}. The
40530 annex part of the generic @samp{qXfer} packet must be empty
40531 (@pxref{qXfer read}).
40533 This packet is not probed by default; the remote stub must request it,
40534 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40536 @item qXfer:traceframe-info:read::@var{offset},@var{length}
40537 @anchor{qXfer traceframe info read}
40539 Return a description of the current traceframe's contents.
40540 @xref{Traceframe Info Format}. The annex part of the generic
40541 @samp{qXfer} packet must be empty (@pxref{qXfer read}).
40543 This packet is not probed by default; the remote stub must request it,
40544 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40546 @item qXfer:uib:read:@var{pc}:@var{offset},@var{length}
40547 @anchor{qXfer unwind info block}
40549 Return the unwind information block for @var{pc}. This packet is used
40550 on OpenVMS/ia64 to ask the kernel unwind information.
40552 This packet is not probed by default.
40554 @item qXfer:fdpic:read:@var{annex}:@var{offset},@var{length}
40555 @anchor{qXfer fdpic loadmap read}
40556 Read contents of @code{loadmap}s on the target system. The
40557 annex, either @samp{exec} or @samp{interp}, specifies which @code{loadmap},
40558 executable @code{loadmap} or interpreter @code{loadmap} to read.
40560 This packet is not probed by default; the remote stub must request it,
40561 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40563 @item qXfer:osdata:read::@var{offset},@var{length}
40564 @anchor{qXfer osdata read}
40565 Access the target's @dfn{operating system information}.
40566 @xref{Operating System Information}.
40570 @item qXfer:@var{object}:write:@var{annex}:@var{offset}:@var{data}@dots{}
40571 @cindex write data into object, remote request
40572 @anchor{qXfer write}
40573 Write uninterpreted bytes into the target's special data area
40574 identified by the keyword @var{object}, starting at @var{offset} bytes
40575 into the data. The binary-encoded data (@pxref{Binary Data}) to be
40576 written is given by @var{data}@dots{}. The content and encoding of @var{annex}
40577 is specific to @var{object}; it can supply additional details about what data
40583 @var{nn} (hex encoded) is the number of bytes written.
40584 This may be fewer bytes than supplied in the request.
40587 The request was malformed, or @var{annex} was invalid.
40590 The offset was invalid, or there was an error encountered writing the data.
40591 The @var{nn} part is a hex-encoded @code{errno} value.
40594 An empty reply indicates the @var{object} string was not
40595 recognized by the stub, or that the object does not support writing.
40598 Here are the specific requests of this form defined so far. All the
40599 @samp{qXfer:@var{object}:write:@dots{}} requests use the same reply
40600 formats, listed above.
40603 @item qXfer:siginfo:write::@var{offset}:@var{data}@dots{}
40604 @anchor{qXfer siginfo write}
40605 Write @var{data} to the extra signal information on the target system.
40606 The annex part of the generic @samp{qXfer} packet must be
40607 empty (@pxref{qXfer write}).
40609 This packet is not probed by default; the remote stub must request it,
40610 by supplying an appropriate @samp{qSupported} response
40611 (@pxref{qSupported}).
40613 @item qXfer:spu:write:@var{annex}:@var{offset}:@var{data}@dots{}
40614 @anchor{qXfer spu write}
40615 Write @var{data} to an @code{spufs} file on the target system. The
40616 annex specifies which file to write; it must be of the form
40617 @file{@var{id}/@var{name}}, where @var{id} specifies an SPU context ID
40618 in the target process, and @var{name} identifes the @code{spufs} file
40619 in that context to be accessed.
40621 This packet is not probed by default; the remote stub must request it,
40622 by supplying an appropriate @samp{qSupported} response (@pxref{qSupported}).
40625 @item qXfer:@var{object}:@var{operation}:@dots{}
40626 Requests of this form may be added in the future. When a stub does
40627 not recognize the @var{object} keyword, or its support for
40628 @var{object} does not recognize the @var{operation} keyword, the stub
40629 must respond with an empty packet.
40631 @item qAttached:@var{pid}
40632 @cindex query attached, remote request
40633 @cindex @samp{qAttached} packet
40634 Return an indication of whether the remote server attached to an
40635 existing process or created a new process. When the multiprocess
40636 protocol extensions are supported (@pxref{multiprocess extensions}),
40637 @var{pid} is an integer in hexadecimal format identifying the target
40638 process. Otherwise, @value{GDBN} will omit the @var{pid} field and
40639 the query packet will be simplified as @samp{qAttached}.
40641 This query is used, for example, to know whether the remote process
40642 should be detached or killed when a @value{GDBN} session is ended with
40643 the @code{quit} command.
40648 The remote server attached to an existing process.
40650 The remote server created a new process.
40652 A badly formed request or an error was encountered.
40656 Enable branch tracing for the current thread using Branch Trace Store.
40661 Branch tracing has been enabled.
40663 A badly formed request or an error was encountered.
40667 Enable branch tracing for the current thread using Intel Processor Trace.
40672 Branch tracing has been enabled.
40674 A badly formed request or an error was encountered.
40678 Disable branch tracing for the current thread.
40683 Branch tracing has been disabled.
40685 A badly formed request or an error was encountered.
40688 @item Qbtrace-conf:bts:size=@var{value}
40689 Set the requested ring buffer size for new threads that use the
40690 btrace recording method in bts format.
40695 The ring buffer size has been set.
40697 A badly formed request or an error was encountered.
40700 @item Qbtrace-conf:pt:size=@var{value}
40701 Set the requested ring buffer size for new threads that use the
40702 btrace recording method in pt format.
40707 The ring buffer size has been set.
40709 A badly formed request or an error was encountered.
40714 @node Architecture-Specific Protocol Details
40715 @section Architecture-Specific Protocol Details
40717 This section describes how the remote protocol is applied to specific
40718 target architectures. Also see @ref{Standard Target Features}, for
40719 details of XML target descriptions for each architecture.
40722 * ARM-Specific Protocol Details::
40723 * MIPS-Specific Protocol Details::
40726 @node ARM-Specific Protocol Details
40727 @subsection @acronym{ARM}-specific Protocol Details
40730 * ARM Breakpoint Kinds::
40733 @node ARM Breakpoint Kinds
40734 @subsubsection @acronym{ARM} Breakpoint Kinds
40735 @cindex breakpoint kinds, @acronym{ARM}
40737 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
40742 16-bit Thumb mode breakpoint.
40745 32-bit Thumb mode (Thumb-2) breakpoint.
40748 32-bit @acronym{ARM} mode breakpoint.
40752 @node MIPS-Specific Protocol Details
40753 @subsection @acronym{MIPS}-specific Protocol Details
40756 * MIPS Register packet Format::
40757 * MIPS Breakpoint Kinds::
40760 @node MIPS Register packet Format
40761 @subsubsection @acronym{MIPS} Register Packet Format
40762 @cindex register packet format, @acronym{MIPS}
40764 The following @code{g}/@code{G} packets have previously been defined.
40765 In the below, some thirty-two bit registers are transferred as
40766 sixty-four bits. Those registers should be zero/sign extended (which?)
40767 to fill the space allocated. Register bytes are transferred in target
40768 byte order. The two nibbles within a register byte are transferred
40769 most-significant -- least-significant.
40774 All registers are transferred as thirty-two bit quantities in the order:
40775 32 general-purpose; sr; lo; hi; bad; cause; pc; 32 floating-point
40776 registers; fsr; fir; fp.
40779 All registers are transferred as sixty-four bit quantities (including
40780 thirty-two bit registers such as @code{sr}). The ordering is the same
40785 @node MIPS Breakpoint Kinds
40786 @subsubsection @acronym{MIPS} Breakpoint Kinds
40787 @cindex breakpoint kinds, @acronym{MIPS}
40789 These breakpoint kinds are defined for the @samp{Z0} and @samp{Z1} packets.
40794 16-bit @acronym{MIPS16} mode breakpoint.
40797 16-bit @acronym{microMIPS} mode breakpoint.
40800 32-bit standard @acronym{MIPS} mode breakpoint.
40803 32-bit @acronym{microMIPS} mode breakpoint.
40807 @node Tracepoint Packets
40808 @section Tracepoint Packets
40809 @cindex tracepoint packets
40810 @cindex packets, tracepoint
40812 Here we describe the packets @value{GDBN} uses to implement
40813 tracepoints (@pxref{Tracepoints}).
40817 @item QTDP:@var{n}:@var{addr}:@var{ena}:@var{step}:@var{pass}[:F@var{flen}][:X@var{len},@var{bytes}]@r{[}-@r{]}
40818 @cindex @samp{QTDP} packet
40819 Create a new tracepoint, number @var{n}, at @var{addr}. If @var{ena}
40820 is @samp{E}, then the tracepoint is enabled; if it is @samp{D}, then
40821 the tracepoint is disabled. The @var{step} gives the tracepoint's step
40822 count, and @var{pass} gives its pass count. If an @samp{F} is present,
40823 then the tracepoint is to be a fast tracepoint, and the @var{flen} is
40824 the number of bytes that the target should copy elsewhere to make room
40825 for the tracepoint. If an @samp{X} is present, it introduces a
40826 tracepoint condition, which consists of a hexadecimal length, followed
40827 by a comma and hex-encoded bytes, in a manner similar to action
40828 encodings as described below. If the trailing @samp{-} is present,
40829 further @samp{QTDP} packets will follow to specify this tracepoint's
40835 The packet was understood and carried out.
40837 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
40839 The packet was not recognized.
40842 @item QTDP:-@var{n}:@var{addr}:@r{[}S@r{]}@var{action}@dots{}@r{[}-@r{]}
40843 Define actions to be taken when a tracepoint is hit. The @var{n} and
40844 @var{addr} must be the same as in the initial @samp{QTDP} packet for
40845 this tracepoint. This packet may only be sent immediately after
40846 another @samp{QTDP} packet that ended with a @samp{-}. If the
40847 trailing @samp{-} is present, further @samp{QTDP} packets will follow,
40848 specifying more actions for this tracepoint.
40850 In the series of action packets for a given tracepoint, at most one
40851 can have an @samp{S} before its first @var{action}. If such a packet
40852 is sent, it and the following packets define ``while-stepping''
40853 actions. Any prior packets define ordinary actions --- that is, those
40854 taken when the tracepoint is first hit. If no action packet has an
40855 @samp{S}, then all the packets in the series specify ordinary
40856 tracepoint actions.
40858 The @samp{@var{action}@dots{}} portion of the packet is a series of
40859 actions, concatenated without separators. Each action has one of the
40865 Collect the registers whose bits are set in @var{mask},
40866 a hexadecimal number whose @var{i}'th bit is set if register number
40867 @var{i} should be collected. (The least significant bit is numbered
40868 zero.) Note that @var{mask} may be any number of digits long; it may
40869 not fit in a 32-bit word.
40871 @item M @var{basereg},@var{offset},@var{len}
40872 Collect @var{len} bytes of memory starting at the address in register
40873 number @var{basereg}, plus @var{offset}. If @var{basereg} is
40874 @samp{-1}, then the range has a fixed address: @var{offset} is the
40875 address of the lowest byte to collect. The @var{basereg},
40876 @var{offset}, and @var{len} parameters are all unsigned hexadecimal
40877 values (the @samp{-1} value for @var{basereg} is a special case).
40879 @item X @var{len},@var{expr}
40880 Evaluate @var{expr}, whose length is @var{len}, and collect memory as
40881 it directs. The agent expression @var{expr} is as described in
40882 @ref{Agent Expressions}. Each byte of the expression is encoded as a
40883 two-digit hex number in the packet; @var{len} is the number of bytes
40884 in the expression (and thus one-half the number of hex digits in the
40889 Any number of actions may be packed together in a single @samp{QTDP}
40890 packet, as long as the packet does not exceed the maximum packet
40891 length (400 bytes, for many stubs). There may be only one @samp{R}
40892 action per tracepoint, and it must precede any @samp{M} or @samp{X}
40893 actions. Any registers referred to by @samp{M} and @samp{X} actions
40894 must be collected by a preceding @samp{R} action. (The
40895 ``while-stepping'' actions are treated as if they were attached to a
40896 separate tracepoint, as far as these restrictions are concerned.)
40901 The packet was understood and carried out.
40903 @xref{Tracepoint Packets,,Relocate instruction reply packet}.
40905 The packet was not recognized.
40908 @item QTDPsrc:@var{n}:@var{addr}:@var{type}:@var{start}:@var{slen}:@var{bytes}
40909 @cindex @samp{QTDPsrc} packet
40910 Specify a source string of tracepoint @var{n} at address @var{addr}.
40911 This is useful to get accurate reproduction of the tracepoints
40912 originally downloaded at the beginning of the trace run. The @var{type}
40913 is the name of the tracepoint part, such as @samp{cond} for the
40914 tracepoint's conditional expression (see below for a list of types), while
40915 @var{bytes} is the string, encoded in hexadecimal.
40917 @var{start} is the offset of the @var{bytes} within the overall source
40918 string, while @var{slen} is the total length of the source string.
40919 This is intended for handling source strings that are longer than will
40920 fit in a single packet.
40921 @c Add detailed example when this info is moved into a dedicated
40922 @c tracepoint descriptions section.
40924 The available string types are @samp{at} for the location,
40925 @samp{cond} for the conditional, and @samp{cmd} for an action command.
40926 @value{GDBN} sends a separate packet for each command in the action
40927 list, in the same order in which the commands are stored in the list.
40929 The target does not need to do anything with source strings except
40930 report them back as part of the replies to the @samp{qTfP}/@samp{qTsP}
40933 Although this packet is optional, and @value{GDBN} will only send it
40934 if the target replies with @samp{TracepointSource} @xref{General
40935 Query Packets}, it makes both disconnected tracing and trace files
40936 much easier to use. Otherwise the user must be careful that the
40937 tracepoints in effect while looking at trace frames are identical to
40938 the ones in effect during the trace run; even a small discrepancy
40939 could cause @samp{tdump} not to work, or a particular trace frame not
40942 @item QTDV:@var{n}:@var{value}:@var{builtin}:@var{name}
40943 @cindex define trace state variable, remote request
40944 @cindex @samp{QTDV} packet
40945 Create a new trace state variable, number @var{n}, with an initial
40946 value of @var{value}, which is a 64-bit signed integer. Both @var{n}
40947 and @var{value} are encoded as hexadecimal values. @value{GDBN} has
40948 the option of not using this packet for initial values of zero; the
40949 target should simply create the trace state variables as they are
40950 mentioned in expressions. The value @var{builtin} should be 1 (one)
40951 if the trace state variable is builtin and 0 (zero) if it is not builtin.
40952 @value{GDBN} only sets @var{builtin} to 1 if a previous @samp{qTfV} or
40953 @samp{qTsV} packet had it set. The contents of @var{name} is the
40954 hex-encoded name (without the leading @samp{$}) of the trace state
40957 @item QTFrame:@var{n}
40958 @cindex @samp{QTFrame} packet
40959 Select the @var{n}'th tracepoint frame from the buffer, and use the
40960 register and memory contents recorded there to answer subsequent
40961 request packets from @value{GDBN}.
40963 A successful reply from the stub indicates that the stub has found the
40964 requested frame. The response is a series of parts, concatenated
40965 without separators, describing the frame we selected. Each part has
40966 one of the following forms:
40970 The selected frame is number @var{n} in the trace frame buffer;
40971 @var{f} is a hexadecimal number. If @var{f} is @samp{-1}, then there
40972 was no frame matching the criteria in the request packet.
40975 The selected trace frame records a hit of tracepoint number @var{t};
40976 @var{t} is a hexadecimal number.
40980 @item QTFrame:pc:@var{addr}
40981 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
40982 currently selected frame whose PC is @var{addr};
40983 @var{addr} is a hexadecimal number.
40985 @item QTFrame:tdp:@var{t}
40986 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
40987 currently selected frame that is a hit of tracepoint @var{t}; @var{t}
40988 is a hexadecimal number.
40990 @item QTFrame:range:@var{start}:@var{end}
40991 Like @samp{QTFrame:@var{n}}, but select the first tracepoint frame after the
40992 currently selected frame whose PC is between @var{start} (inclusive)
40993 and @var{end} (inclusive); @var{start} and @var{end} are hexadecimal
40996 @item QTFrame:outside:@var{start}:@var{end}
40997 Like @samp{QTFrame:range:@var{start}:@var{end}}, but select the first
40998 frame @emph{outside} the given range of addresses (exclusive).
41001 @cindex @samp{qTMinFTPILen} packet
41002 This packet requests the minimum length of instruction at which a fast
41003 tracepoint (@pxref{Set Tracepoints}) may be placed. For instance, on
41004 the 32-bit x86 architecture, it is possible to use a 4-byte jump, but
41005 it depends on the target system being able to create trampolines in
41006 the first 64K of memory, which might or might not be possible for that
41007 system. So the reply to this packet will be 4 if it is able to
41014 The minimum instruction length is currently unknown.
41016 The minimum instruction length is @var{length}, where @var{length}
41017 is a hexadecimal number greater or equal to 1. A reply
41018 of 1 means that a fast tracepoint may be placed on any instruction
41019 regardless of size.
41021 An error has occurred.
41023 An empty reply indicates that the request is not supported by the stub.
41027 @cindex @samp{QTStart} packet
41028 Begin the tracepoint experiment. Begin collecting data from
41029 tracepoint hits in the trace frame buffer. This packet supports the
41030 @samp{qRelocInsn} reply (@pxref{Tracepoint Packets,,Relocate
41031 instruction reply packet}).
41034 @cindex @samp{QTStop} packet
41035 End the tracepoint experiment. Stop collecting trace frames.
41037 @item QTEnable:@var{n}:@var{addr}
41039 @cindex @samp{QTEnable} packet
41040 Enable tracepoint @var{n} at address @var{addr} in a started tracepoint
41041 experiment. If the tracepoint was previously disabled, then collection
41042 of data from it will resume.
41044 @item QTDisable:@var{n}:@var{addr}
41046 @cindex @samp{QTDisable} packet
41047 Disable tracepoint @var{n} at address @var{addr} in a started tracepoint
41048 experiment. No more data will be collected from the tracepoint unless
41049 @samp{QTEnable:@var{n}:@var{addr}} is subsequently issued.
41052 @cindex @samp{QTinit} packet
41053 Clear the table of tracepoints, and empty the trace frame buffer.
41055 @item QTro:@var{start1},@var{end1}:@var{start2},@var{end2}:@dots{}
41056 @cindex @samp{QTro} packet
41057 Establish the given ranges of memory as ``transparent''. The stub
41058 will answer requests for these ranges from memory's current contents,
41059 if they were not collected as part of the tracepoint hit.
41061 @value{GDBN} uses this to mark read-only regions of memory, like those
41062 containing program code. Since these areas never change, they should
41063 still have the same contents they did when the tracepoint was hit, so
41064 there's no reason for the stub to refuse to provide their contents.
41066 @item QTDisconnected:@var{value}
41067 @cindex @samp{QTDisconnected} packet
41068 Set the choice to what to do with the tracing run when @value{GDBN}
41069 disconnects from the target. A @var{value} of 1 directs the target to
41070 continue the tracing run, while 0 tells the target to stop tracing if
41071 @value{GDBN} is no longer in the picture.
41074 @cindex @samp{qTStatus} packet
41075 Ask the stub if there is a trace experiment running right now.
41077 The reply has the form:
41081 @item T@var{running}@r{[};@var{field}@r{]}@dots{}
41082 @var{running} is a single digit @code{1} if the trace is presently
41083 running, or @code{0} if not. It is followed by semicolon-separated
41084 optional fields that an agent may use to report additional status.
41088 If the trace is not running, the agent may report any of several
41089 explanations as one of the optional fields:
41094 No trace has been run yet.
41096 @item tstop[:@var{text}]:0
41097 The trace was stopped by a user-originated stop command. The optional
41098 @var{text} field is a user-supplied string supplied as part of the
41099 stop command (for instance, an explanation of why the trace was
41100 stopped manually). It is hex-encoded.
41103 The trace stopped because the trace buffer filled up.
41105 @item tdisconnected:0
41106 The trace stopped because @value{GDBN} disconnected from the target.
41108 @item tpasscount:@var{tpnum}
41109 The trace stopped because tracepoint @var{tpnum} exceeded its pass count.
41111 @item terror:@var{text}:@var{tpnum}
41112 The trace stopped because tracepoint @var{tpnum} had an error. The
41113 string @var{text} is available to describe the nature of the error
41114 (for instance, a divide by zero in the condition expression); it
41118 The trace stopped for some other reason.
41122 Additional optional fields supply statistical and other information.
41123 Although not required, they are extremely useful for users monitoring
41124 the progress of a trace run. If a trace has stopped, and these
41125 numbers are reported, they must reflect the state of the just-stopped
41130 @item tframes:@var{n}
41131 The number of trace frames in the buffer.
41133 @item tcreated:@var{n}
41134 The total number of trace frames created during the run. This may
41135 be larger than the trace frame count, if the buffer is circular.
41137 @item tsize:@var{n}
41138 The total size of the trace buffer, in bytes.
41140 @item tfree:@var{n}
41141 The number of bytes still unused in the buffer.
41143 @item circular:@var{n}
41144 The value of the circular trace buffer flag. @code{1} means that the
41145 trace buffer is circular and old trace frames will be discarded if
41146 necessary to make room, @code{0} means that the trace buffer is linear
41149 @item disconn:@var{n}
41150 The value of the disconnected tracing flag. @code{1} means that
41151 tracing will continue after @value{GDBN} disconnects, @code{0} means
41152 that the trace run will stop.
41156 @item qTP:@var{tp}:@var{addr}
41157 @cindex tracepoint status, remote request
41158 @cindex @samp{qTP} packet
41159 Ask the stub for the current state of tracepoint number @var{tp} at
41160 address @var{addr}.
41164 @item V@var{hits}:@var{usage}
41165 The tracepoint has been hit @var{hits} times so far during the trace
41166 run, and accounts for @var{usage} in the trace buffer. Note that
41167 @code{while-stepping} steps are not counted as separate hits, but the
41168 steps' space consumption is added into the usage number.
41172 @item qTV:@var{var}
41173 @cindex trace state variable value, remote request
41174 @cindex @samp{qTV} packet
41175 Ask the stub for the value of the trace state variable number @var{var}.
41180 The value of the variable is @var{value}. This will be the current
41181 value of the variable if the user is examining a running target, or a
41182 saved value if the variable was collected in the trace frame that the
41183 user is looking at. Note that multiple requests may result in
41184 different reply values, such as when requesting values while the
41185 program is running.
41188 The value of the variable is unknown. This would occur, for example,
41189 if the user is examining a trace frame in which the requested variable
41194 @cindex @samp{qTfP} packet
41196 @cindex @samp{qTsP} packet
41197 These packets request data about tracepoints that are being used by
41198 the target. @value{GDBN} sends @code{qTfP} to get the first piece
41199 of data, and multiple @code{qTsP} to get additional pieces. Replies
41200 to these packets generally take the form of the @code{QTDP} packets
41201 that define tracepoints. (FIXME add detailed syntax)
41204 @cindex @samp{qTfV} packet
41206 @cindex @samp{qTsV} packet
41207 These packets request data about trace state variables that are on the
41208 target. @value{GDBN} sends @code{qTfV} to get the first vari of data,
41209 and multiple @code{qTsV} to get additional variables. Replies to
41210 these packets follow the syntax of the @code{QTDV} packets that define
41211 trace state variables.
41217 @cindex @samp{qTfSTM} packet
41218 @cindex @samp{qTsSTM} packet
41219 These packets request data about static tracepoint markers that exist
41220 in the target program. @value{GDBN} sends @code{qTfSTM} to get the
41221 first piece of data, and multiple @code{qTsSTM} to get additional
41222 pieces. Replies to these packets take the following form:
41226 @item m @var{address}:@var{id}:@var{extra}
41228 @item m @var{address}:@var{id}:@var{extra},@var{address}:@var{id}:@var{extra}@dots{}
41229 a comma-separated list of markers
41231 (lower case letter @samp{L}) denotes end of list.
41233 An error occurred. The error number @var{nn} is given as hex digits.
41235 An empty reply indicates that the request is not supported by the
41239 The @var{address} is encoded in hex;
41240 @var{id} and @var{extra} are strings encoded in hex.
41242 In response to each query, the target will reply with a list of one or
41243 more markers, separated by commas. @value{GDBN} will respond to each
41244 reply with a request for more markers (using the @samp{qs} form of the
41245 query), until the target responds with @samp{l} (lower-case ell, for
41248 @item qTSTMat:@var{address}
41250 @cindex @samp{qTSTMat} packet
41251 This packets requests data about static tracepoint markers in the
41252 target program at @var{address}. Replies to this packet follow the
41253 syntax of the @samp{qTfSTM} and @code{qTsSTM} packets that list static
41254 tracepoint markers.
41256 @item QTSave:@var{filename}
41257 @cindex @samp{QTSave} packet
41258 This packet directs the target to save trace data to the file name
41259 @var{filename} in the target's filesystem. The @var{filename} is encoded
41260 as a hex string; the interpretation of the file name (relative vs
41261 absolute, wild cards, etc) is up to the target.
41263 @item qTBuffer:@var{offset},@var{len}
41264 @cindex @samp{qTBuffer} packet
41265 Return up to @var{len} bytes of the current contents of trace buffer,
41266 starting at @var{offset}. The trace buffer is treated as if it were
41267 a contiguous collection of traceframes, as per the trace file format.
41268 The reply consists as many hex-encoded bytes as the target can deliver
41269 in a packet; it is not an error to return fewer than were asked for.
41270 A reply consisting of just @code{l} indicates that no bytes are
41273 @item QTBuffer:circular:@var{value}
41274 This packet directs the target to use a circular trace buffer if
41275 @var{value} is 1, or a linear buffer if the value is 0.
41277 @item QTBuffer:size:@var{size}
41278 @anchor{QTBuffer-size}
41279 @cindex @samp{QTBuffer size} packet
41280 This packet directs the target to make the trace buffer be of size
41281 @var{size} if possible. A value of @code{-1} tells the target to
41282 use whatever size it prefers.
41284 @item QTNotes:@r{[}@var{type}:@var{text}@r{]}@r{[};@var{type}:@var{text}@r{]}@dots{}
41285 @cindex @samp{QTNotes} packet
41286 This packet adds optional textual notes to the trace run. Allowable
41287 types include @code{user}, @code{notes}, and @code{tstop}, the
41288 @var{text} fields are arbitrary strings, hex-encoded.
41292 @subsection Relocate instruction reply packet
41293 When installing fast tracepoints in memory, the target may need to
41294 relocate the instruction currently at the tracepoint address to a
41295 different address in memory. For most instructions, a simple copy is
41296 enough, but, for example, call instructions that implicitly push the
41297 return address on the stack, and relative branches or other
41298 PC-relative instructions require offset adjustment, so that the effect
41299 of executing the instruction at a different address is the same as if
41300 it had executed in the original location.
41302 In response to several of the tracepoint packets, the target may also
41303 respond with a number of intermediate @samp{qRelocInsn} request
41304 packets before the final result packet, to have @value{GDBN} handle
41305 this relocation operation. If a packet supports this mechanism, its
41306 documentation will explicitly say so. See for example the above
41307 descriptions for the @samp{QTStart} and @samp{QTDP} packets. The
41308 format of the request is:
41311 @item qRelocInsn:@var{from};@var{to}
41313 This requests @value{GDBN} to copy instruction at address @var{from}
41314 to address @var{to}, possibly adjusted so that executing the
41315 instruction at @var{to} has the same effect as executing it at
41316 @var{from}. @value{GDBN} writes the adjusted instruction to target
41317 memory starting at @var{to}.
41322 @item qRelocInsn:@var{adjusted_size}
41323 Informs the stub the relocation is complete. The @var{adjusted_size} is
41324 the length in bytes of resulting relocated instruction sequence.
41326 A badly formed request was detected, or an error was encountered while
41327 relocating the instruction.
41330 @node Host I/O Packets
41331 @section Host I/O Packets
41332 @cindex Host I/O, remote protocol
41333 @cindex file transfer, remote protocol
41335 The @dfn{Host I/O} packets allow @value{GDBN} to perform I/O
41336 operations on the far side of a remote link. For example, Host I/O is
41337 used to upload and download files to a remote target with its own
41338 filesystem. Host I/O uses the same constant values and data structure
41339 layout as the target-initiated File-I/O protocol. However, the
41340 Host I/O packets are structured differently. The target-initiated
41341 protocol relies on target memory to store parameters and buffers.
41342 Host I/O requests are initiated by @value{GDBN}, and the
41343 target's memory is not involved. @xref{File-I/O Remote Protocol
41344 Extension}, for more details on the target-initiated protocol.
41346 The Host I/O request packets all encode a single operation along with
41347 its arguments. They have this format:
41351 @item vFile:@var{operation}: @var{parameter}@dots{}
41352 @var{operation} is the name of the particular request; the target
41353 should compare the entire packet name up to the second colon when checking
41354 for a supported operation. The format of @var{parameter} depends on
41355 the operation. Numbers are always passed in hexadecimal. Negative
41356 numbers have an explicit minus sign (i.e.@: two's complement is not
41357 used). Strings (e.g.@: filenames) are encoded as a series of
41358 hexadecimal bytes. The last argument to a system call may be a
41359 buffer of escaped binary data (@pxref{Binary Data}).
41363 The valid responses to Host I/O packets are:
41367 @item F @var{result} [, @var{errno}] [; @var{attachment}]
41368 @var{result} is the integer value returned by this operation, usually
41369 non-negative for success and -1 for errors. If an error has occured,
41370 @var{errno} will be included in the result specifying a
41371 value defined by the File-I/O protocol (@pxref{Errno Values}). For
41372 operations which return data, @var{attachment} supplies the data as a
41373 binary buffer. Binary buffers in response packets are escaped in the
41374 normal way (@pxref{Binary Data}). See the individual packet
41375 documentation for the interpretation of @var{result} and
41379 An empty response indicates that this operation is not recognized.
41383 These are the supported Host I/O operations:
41386 @item vFile:open: @var{filename}, @var{flags}, @var{mode}
41387 Open a file at @var{filename} and return a file descriptor for it, or
41388 return -1 if an error occurs. The @var{filename} is a string,
41389 @var{flags} is an integer indicating a mask of open flags
41390 (@pxref{Open Flags}), and @var{mode} is an integer indicating a mask
41391 of mode bits to use if the file is created (@pxref{mode_t Values}).
41392 @xref{open}, for details of the open flags and mode values.
41394 @item vFile:close: @var{fd}
41395 Close the open file corresponding to @var{fd} and return 0, or
41396 -1 if an error occurs.
41398 @item vFile:pread: @var{fd}, @var{count}, @var{offset}
41399 Read data from the open file corresponding to @var{fd}. Up to
41400 @var{count} bytes will be read from the file, starting at @var{offset}
41401 relative to the start of the file. The target may read fewer bytes;
41402 common reasons include packet size limits and an end-of-file
41403 condition. The number of bytes read is returned. Zero should only be
41404 returned for a successful read at the end of the file, or if
41405 @var{count} was zero.
41407 The data read should be returned as a binary attachment on success.
41408 If zero bytes were read, the response should include an empty binary
41409 attachment (i.e.@: a trailing semicolon). The return value is the
41410 number of target bytes read; the binary attachment may be longer if
41411 some characters were escaped.
41413 @item vFile:pwrite: @var{fd}, @var{offset}, @var{data}
41414 Write @var{data} (a binary buffer) to the open file corresponding
41415 to @var{fd}. Start the write at @var{offset} from the start of the
41416 file. Unlike many @code{write} system calls, there is no
41417 separate @var{count} argument; the length of @var{data} in the
41418 packet is used. @samp{vFile:write} returns the number of bytes written,
41419 which may be shorter than the length of @var{data}, or -1 if an
41422 @item vFile:fstat: @var{fd}
41423 Get information about the open file corresponding to @var{fd}.
41424 On success the information is returned as a binary attachment
41425 and the return value is the size of this attachment in bytes.
41426 If an error occurs the return value is -1. The format of the
41427 returned binary attachment is as described in @ref{struct stat}.
41429 @item vFile:unlink: @var{filename}
41430 Delete the file at @var{filename} on the target. Return 0,
41431 or -1 if an error occurs. The @var{filename} is a string.
41433 @item vFile:readlink: @var{filename}
41434 Read value of symbolic link @var{filename} on the target. Return
41435 the number of bytes read, or -1 if an error occurs.
41437 The data read should be returned as a binary attachment on success.
41438 If zero bytes were read, the response should include an empty binary
41439 attachment (i.e.@: a trailing semicolon). The return value is the
41440 number of target bytes read; the binary attachment may be longer if
41441 some characters were escaped.
41443 @item vFile:setfs: @var{pid}
41444 Select the filesystem on which @code{vFile} operations with
41445 @var{filename} arguments will operate. This is required for
41446 @value{GDBN} to be able to access files on remote targets where
41447 the remote stub does not share a common filesystem with the
41450 If @var{pid} is nonzero, select the filesystem as seen by process
41451 @var{pid}. If @var{pid} is zero, select the filesystem as seen by
41452 the remote stub. Return 0 on success, or -1 if an error occurs.
41453 If @code{vFile:setfs:} indicates success, the selected filesystem
41454 remains selected until the next successful @code{vFile:setfs:}
41460 @section Interrupts
41461 @cindex interrupts (remote protocol)
41462 @anchor{interrupting remote targets}
41464 In all-stop mode, when a program on the remote target is running,
41465 @value{GDBN} may attempt to interrupt it by sending a @samp{Ctrl-C},
41466 @code{BREAK} or a @code{BREAK} followed by @code{g}, control of which
41467 is specified via @value{GDBN}'s @samp{interrupt-sequence}.
41469 The precise meaning of @code{BREAK} is defined by the transport
41470 mechanism and may, in fact, be undefined. @value{GDBN} does not
41471 currently define a @code{BREAK} mechanism for any of the network
41472 interfaces except for TCP, in which case @value{GDBN} sends the
41473 @code{telnet} BREAK sequence.
41475 @samp{Ctrl-C}, on the other hand, is defined and implemented for all
41476 transport mechanisms. It is represented by sending the single byte
41477 @code{0x03} without any of the usual packet overhead described in
41478 the Overview section (@pxref{Overview}). When a @code{0x03} byte is
41479 transmitted as part of a packet, it is considered to be packet data
41480 and does @emph{not} represent an interrupt. E.g., an @samp{X} packet
41481 (@pxref{X packet}), used for binary downloads, may include an unescaped
41482 @code{0x03} as part of its packet.
41484 @code{BREAK} followed by @code{g} is also known as Magic SysRq g.
41485 When Linux kernel receives this sequence from serial port,
41486 it stops execution and connects to gdb.
41488 In non-stop mode, because packet resumptions are asynchronous
41489 (@pxref{vCont packet}), @value{GDBN} is always free to send a remote
41490 command to the remote stub, even when the target is running. For that
41491 reason, @value{GDBN} instead sends a regular packet (@pxref{vCtrlC
41492 packet}) with the usual packet framing instead of the single byte
41495 Stubs are not required to recognize these interrupt mechanisms and the
41496 precise meaning associated with receipt of the interrupt is
41497 implementation defined. If the target supports debugging of multiple
41498 threads and/or processes, it should attempt to interrupt all
41499 currently-executing threads and processes.
41500 If the stub is successful at interrupting the
41501 running program, it should send one of the stop
41502 reply packets (@pxref{Stop Reply Packets}) to @value{GDBN} as a result
41503 of successfully stopping the program in all-stop mode, and a stop reply
41504 for each stopped thread in non-stop mode.
41505 Interrupts received while the
41506 program is stopped are queued and the program will be interrupted when
41507 it is resumed next time.
41509 @node Notification Packets
41510 @section Notification Packets
41511 @cindex notification packets
41512 @cindex packets, notification
41514 The @value{GDBN} remote serial protocol includes @dfn{notifications},
41515 packets that require no acknowledgment. Both the GDB and the stub
41516 may send notifications (although the only notifications defined at
41517 present are sent by the stub). Notifications carry information
41518 without incurring the round-trip latency of an acknowledgment, and so
41519 are useful for low-impact communications where occasional packet loss
41522 A notification packet has the form @samp{% @var{data} #
41523 @var{checksum}}, where @var{data} is the content of the notification,
41524 and @var{checksum} is a checksum of @var{data}, computed and formatted
41525 as for ordinary @value{GDBN} packets. A notification's @var{data}
41526 never contains @samp{$}, @samp{%} or @samp{#} characters. Upon
41527 receiving a notification, the recipient sends no @samp{+} or @samp{-}
41528 to acknowledge the notification's receipt or to report its corruption.
41530 Every notification's @var{data} begins with a name, which contains no
41531 colon characters, followed by a colon character.
41533 Recipients should silently ignore corrupted notifications and
41534 notifications they do not understand. Recipients should restart
41535 timeout periods on receipt of a well-formed notification, whether or
41536 not they understand it.
41538 Senders should only send the notifications described here when this
41539 protocol description specifies that they are permitted. In the
41540 future, we may extend the protocol to permit existing notifications in
41541 new contexts; this rule helps older senders avoid confusing newer
41544 (Older versions of @value{GDBN} ignore bytes received until they see
41545 the @samp{$} byte that begins an ordinary packet, so new stubs may
41546 transmit notifications without fear of confusing older clients. There
41547 are no notifications defined for @value{GDBN} to send at the moment, but we
41548 assume that most older stubs would ignore them, as well.)
41550 Each notification is comprised of three parts:
41552 @item @var{name}:@var{event}
41553 The notification packet is sent by the side that initiates the
41554 exchange (currently, only the stub does that), with @var{event}
41555 carrying the specific information about the notification, and
41556 @var{name} specifying the name of the notification.
41558 The acknowledge sent by the other side, usually @value{GDBN}, to
41559 acknowledge the exchange and request the event.
41562 The purpose of an asynchronous notification mechanism is to report to
41563 @value{GDBN} that something interesting happened in the remote stub.
41565 The remote stub may send notification @var{name}:@var{event}
41566 at any time, but @value{GDBN} acknowledges the notification when
41567 appropriate. The notification event is pending before @value{GDBN}
41568 acknowledges. Only one notification at a time may be pending; if
41569 additional events occur before @value{GDBN} has acknowledged the
41570 previous notification, they must be queued by the stub for later
41571 synchronous transmission in response to @var{ack} packets from
41572 @value{GDBN}. Because the notification mechanism is unreliable,
41573 the stub is permitted to resend a notification if it believes
41574 @value{GDBN} may not have received it.
41576 Specifically, notifications may appear when @value{GDBN} is not
41577 otherwise reading input from the stub, or when @value{GDBN} is
41578 expecting to read a normal synchronous response or a
41579 @samp{+}/@samp{-} acknowledgment to a packet it has sent.
41580 Notification packets are distinct from any other communication from
41581 the stub so there is no ambiguity.
41583 After receiving a notification, @value{GDBN} shall acknowledge it by
41584 sending a @var{ack} packet as a regular, synchronous request to the
41585 stub. Such acknowledgment is not required to happen immediately, as
41586 @value{GDBN} is permitted to send other, unrelated packets to the
41587 stub first, which the stub should process normally.
41589 Upon receiving a @var{ack} packet, if the stub has other queued
41590 events to report to @value{GDBN}, it shall respond by sending a
41591 normal @var{event}. @value{GDBN} shall then send another @var{ack}
41592 packet to solicit further responses; again, it is permitted to send
41593 other, unrelated packets as well which the stub should process
41596 If the stub receives a @var{ack} packet and there are no additional
41597 @var{event} to report, the stub shall return an @samp{OK} response.
41598 At this point, @value{GDBN} has finished processing a notification
41599 and the stub has completed sending any queued events. @value{GDBN}
41600 won't accept any new notifications until the final @samp{OK} is
41601 received . If further notification events occur, the stub shall send
41602 a new notification, @value{GDBN} shall accept the notification, and
41603 the process shall be repeated.
41605 The process of asynchronous notification can be illustrated by the
41608 <- @code{%Stop:T0505:98e7ffbf;04:4ce6ffbf;08:b1b6e54c;thread:p7526.7526;core:0;}
41611 <- @code{T0505:68f37db7;04:40f37db7;08:63850408;thread:p7526.7528;core:0;}
41613 <- @code{T0505:68e3fdb6;04:40e3fdb6;08:63850408;thread:p7526.7529;core:0;}
41618 The following notifications are defined:
41619 @multitable @columnfractions 0.12 0.12 0.38 0.38
41628 @tab @var{reply}. The @var{reply} has the form of a stop reply, as
41629 described in @ref{Stop Reply Packets}. Refer to @ref{Remote Non-Stop},
41630 for information on how these notifications are acknowledged by
41632 @tab Report an asynchronous stop event in non-stop mode.
41636 @node Remote Non-Stop
41637 @section Remote Protocol Support for Non-Stop Mode
41639 @value{GDBN}'s remote protocol supports non-stop debugging of
41640 multi-threaded programs, as described in @ref{Non-Stop Mode}. If the stub
41641 supports non-stop mode, it should report that to @value{GDBN} by including
41642 @samp{QNonStop+} in its @samp{qSupported} response (@pxref{qSupported}).
41644 @value{GDBN} typically sends a @samp{QNonStop} packet only when
41645 establishing a new connection with the stub. Entering non-stop mode
41646 does not alter the state of any currently-running threads, but targets
41647 must stop all threads in any already-attached processes when entering
41648 all-stop mode. @value{GDBN} uses the @samp{?} packet as necessary to
41649 probe the target state after a mode change.
41651 In non-stop mode, when an attached process encounters an event that
41652 would otherwise be reported with a stop reply, it uses the
41653 asynchronous notification mechanism (@pxref{Notification Packets}) to
41654 inform @value{GDBN}. In contrast to all-stop mode, where all threads
41655 in all processes are stopped when a stop reply is sent, in non-stop
41656 mode only the thread reporting the stop event is stopped. That is,
41657 when reporting a @samp{S} or @samp{T} response to indicate completion
41658 of a step operation, hitting a breakpoint, or a fault, only the
41659 affected thread is stopped; any other still-running threads continue
41660 to run. When reporting a @samp{W} or @samp{X} response, all running
41661 threads belonging to other attached processes continue to run.
41663 In non-stop mode, the target shall respond to the @samp{?} packet as
41664 follows. First, any incomplete stop reply notification/@samp{vStopped}
41665 sequence in progress is abandoned. The target must begin a new
41666 sequence reporting stop events for all stopped threads, whether or not
41667 it has previously reported those events to @value{GDBN}. The first
41668 stop reply is sent as a synchronous reply to the @samp{?} packet, and
41669 subsequent stop replies are sent as responses to @samp{vStopped} packets
41670 using the mechanism described above. The target must not send
41671 asynchronous stop reply notifications until the sequence is complete.
41672 If all threads are running when the target receives the @samp{?} packet,
41673 or if the target is not attached to any process, it shall respond
41676 If the stub supports non-stop mode, it should also support the
41677 @samp{swbreak} stop reason if software breakpoints are supported, and
41678 the @samp{hwbreak} stop reason if hardware breakpoints are supported
41679 (@pxref{swbreak stop reason}). This is because given the asynchronous
41680 nature of non-stop mode, between the time a thread hits a breakpoint
41681 and the time the event is finally processed by @value{GDBN}, the
41682 breakpoint may have already been removed from the target. Due to
41683 this, @value{GDBN} needs to be able to tell whether a trap stop was
41684 caused by a delayed breakpoint event, which should be ignored, as
41685 opposed to a random trap signal, which should be reported to the user.
41686 Note the @samp{swbreak} feature implies that the target is responsible
41687 for adjusting the PC when a software breakpoint triggers, if
41688 necessary, such as on the x86 architecture.
41690 @node Packet Acknowledgment
41691 @section Packet Acknowledgment
41693 @cindex acknowledgment, for @value{GDBN} remote
41694 @cindex packet acknowledgment, for @value{GDBN} remote
41695 By default, when either the host or the target machine receives a packet,
41696 the first response expected is an acknowledgment: either @samp{+} (to indicate
41697 the package was received correctly) or @samp{-} (to request retransmission).
41698 This mechanism allows the @value{GDBN} remote protocol to operate over
41699 unreliable transport mechanisms, such as a serial line.
41701 In cases where the transport mechanism is itself reliable (such as a pipe or
41702 TCP connection), the @samp{+}/@samp{-} acknowledgments are redundant.
41703 It may be desirable to disable them in that case to reduce communication
41704 overhead, or for other reasons. This can be accomplished by means of the
41705 @samp{QStartNoAckMode} packet; @pxref{QStartNoAckMode}.
41707 When in no-acknowledgment mode, neither the stub nor @value{GDBN} shall send or
41708 expect @samp{+}/@samp{-} protocol acknowledgments. The packet
41709 and response format still includes the normal checksum, as described in
41710 @ref{Overview}, but the checksum may be ignored by the receiver.
41712 If the stub supports @samp{QStartNoAckMode} and prefers to operate in
41713 no-acknowledgment mode, it should report that to @value{GDBN}
41714 by including @samp{QStartNoAckMode+} in its response to @samp{qSupported};
41715 @pxref{qSupported}.
41716 If @value{GDBN} also supports @samp{QStartNoAckMode} and it has not been
41717 disabled via the @code{set remote noack-packet off} command
41718 (@pxref{Remote Configuration}),
41719 @value{GDBN} may then send a @samp{QStartNoAckMode} packet to the stub.
41720 Only then may the stub actually turn off packet acknowledgments.
41721 @value{GDBN} sends a final @samp{+} acknowledgment of the stub's @samp{OK}
41722 response, which can be safely ignored by the stub.
41724 Note that @code{set remote noack-packet} command only affects negotiation
41725 between @value{GDBN} and the stub when subsequent connections are made;
41726 it does not affect the protocol acknowledgment state for any current
41728 Since @samp{+}/@samp{-} acknowledgments are enabled by default when a
41729 new connection is established,
41730 there is also no protocol request to re-enable the acknowledgments
41731 for the current connection, once disabled.
41736 Example sequence of a target being re-started. Notice how the restart
41737 does not get any direct output:
41742 @emph{target restarts}
41745 <- @code{T001:1234123412341234}
41749 Example sequence of a target being stepped by a single instruction:
41752 -> @code{G1445@dots{}}
41757 <- @code{T001:1234123412341234}
41761 <- @code{1455@dots{}}
41765 @node File-I/O Remote Protocol Extension
41766 @section File-I/O Remote Protocol Extension
41767 @cindex File-I/O remote protocol extension
41770 * File-I/O Overview::
41771 * Protocol Basics::
41772 * The F Request Packet::
41773 * The F Reply Packet::
41774 * The Ctrl-C Message::
41776 * List of Supported Calls::
41777 * Protocol-specific Representation of Datatypes::
41779 * File-I/O Examples::
41782 @node File-I/O Overview
41783 @subsection File-I/O Overview
41784 @cindex file-i/o overview
41786 The @dfn{File I/O remote protocol extension} (short: File-I/O) allows the
41787 target to use the host's file system and console I/O to perform various
41788 system calls. System calls on the target system are translated into a
41789 remote protocol packet to the host system, which then performs the needed
41790 actions and returns a response packet to the target system.
41791 This simulates file system operations even on targets that lack file systems.
41793 The protocol is defined to be independent of both the host and target systems.
41794 It uses its own internal representation of datatypes and values. Both
41795 @value{GDBN} and the target's @value{GDBN} stub are responsible for
41796 translating the system-dependent value representations into the internal
41797 protocol representations when data is transmitted.
41799 The communication is synchronous. A system call is possible only when
41800 @value{GDBN} is waiting for a response from the @samp{C}, @samp{c}, @samp{S}
41801 or @samp{s} packets. While @value{GDBN} handles the request for a system call,
41802 the target is stopped to allow deterministic access to the target's
41803 memory. Therefore File-I/O is not interruptible by target signals. On
41804 the other hand, it is possible to interrupt File-I/O by a user interrupt
41805 (@samp{Ctrl-C}) within @value{GDBN}.
41807 The target's request to perform a host system call does not finish
41808 the latest @samp{C}, @samp{c}, @samp{S} or @samp{s} action. That means,
41809 after finishing the system call, the target returns to continuing the
41810 previous activity (continue, step). No additional continue or step
41811 request from @value{GDBN} is required.
41814 (@value{GDBP}) continue
41815 <- target requests 'system call X'
41816 target is stopped, @value{GDBN} executes system call
41817 -> @value{GDBN} returns result
41818 ... target continues, @value{GDBN} returns to wait for the target
41819 <- target hits breakpoint and sends a Txx packet
41822 The protocol only supports I/O on the console and to regular files on
41823 the host file system. Character or block special devices, pipes,
41824 named pipes, sockets or any other communication method on the host
41825 system are not supported by this protocol.
41827 File I/O is not supported in non-stop mode.
41829 @node Protocol Basics
41830 @subsection Protocol Basics
41831 @cindex protocol basics, file-i/o
41833 The File-I/O protocol uses the @code{F} packet as the request as well
41834 as reply packet. Since a File-I/O system call can only occur when
41835 @value{GDBN} is waiting for a response from the continuing or stepping target,
41836 the File-I/O request is a reply that @value{GDBN} has to expect as a result
41837 of a previous @samp{C}, @samp{c}, @samp{S} or @samp{s} packet.
41838 This @code{F} packet contains all information needed to allow @value{GDBN}
41839 to call the appropriate host system call:
41843 A unique identifier for the requested system call.
41846 All parameters to the system call. Pointers are given as addresses
41847 in the target memory address space. Pointers to strings are given as
41848 pointer/length pair. Numerical values are given as they are.
41849 Numerical control flags are given in a protocol-specific representation.
41853 At this point, @value{GDBN} has to perform the following actions.
41857 If the parameters include pointer values to data needed as input to a
41858 system call, @value{GDBN} requests this data from the target with a
41859 standard @code{m} packet request. This additional communication has to be
41860 expected by the target implementation and is handled as any other @code{m}
41864 @value{GDBN} translates all value from protocol representation to host
41865 representation as needed. Datatypes are coerced into the host types.
41868 @value{GDBN} calls the system call.
41871 It then coerces datatypes back to protocol representation.
41874 If the system call is expected to return data in buffer space specified
41875 by pointer parameters to the call, the data is transmitted to the
41876 target using a @code{M} or @code{X} packet. This packet has to be expected
41877 by the target implementation and is handled as any other @code{M} or @code{X}
41882 Eventually @value{GDBN} replies with another @code{F} packet which contains all
41883 necessary information for the target to continue. This at least contains
41890 @code{errno}, if has been changed by the system call.
41897 After having done the needed type and value coercion, the target continues
41898 the latest continue or step action.
41900 @node The F Request Packet
41901 @subsection The @code{F} Request Packet
41902 @cindex file-i/o request packet
41903 @cindex @code{F} request packet
41905 The @code{F} request packet has the following format:
41908 @item F@var{call-id},@var{parameter@dots{}}
41910 @var{call-id} is the identifier to indicate the host system call to be called.
41911 This is just the name of the function.
41913 @var{parameter@dots{}} are the parameters to the system call.
41914 Parameters are hexadecimal integer values, either the actual values in case
41915 of scalar datatypes, pointers to target buffer space in case of compound
41916 datatypes and unspecified memory areas, or pointer/length pairs in case
41917 of string parameters. These are appended to the @var{call-id} as a
41918 comma-delimited list. All values are transmitted in ASCII
41919 string representation, pointer/length pairs separated by a slash.
41925 @node The F Reply Packet
41926 @subsection The @code{F} Reply Packet
41927 @cindex file-i/o reply packet
41928 @cindex @code{F} reply packet
41930 The @code{F} reply packet has the following format:
41934 @item F@var{retcode},@var{errno},@var{Ctrl-C flag};@var{call-specific attachment}
41936 @var{retcode} is the return code of the system call as hexadecimal value.
41938 @var{errno} is the @code{errno} set by the call, in protocol-specific
41940 This parameter can be omitted if the call was successful.
41942 @var{Ctrl-C flag} is only sent if the user requested a break. In this
41943 case, @var{errno} must be sent as well, even if the call was successful.
41944 The @var{Ctrl-C flag} itself consists of the character @samp{C}:
41951 or, if the call was interrupted before the host call has been performed:
41958 assuming 4 is the protocol-specific representation of @code{EINTR}.
41963 @node The Ctrl-C Message
41964 @subsection The @samp{Ctrl-C} Message
41965 @cindex ctrl-c message, in file-i/o protocol
41967 If the @samp{Ctrl-C} flag is set in the @value{GDBN}
41968 reply packet (@pxref{The F Reply Packet}),
41969 the target should behave as if it had
41970 gotten a break message. The meaning for the target is ``system call
41971 interrupted by @code{SIGINT}''. Consequentially, the target should actually stop
41972 (as with a break message) and return to @value{GDBN} with a @code{T02}
41975 It's important for the target to know in which
41976 state the system call was interrupted. There are two possible cases:
41980 The system call hasn't been performed on the host yet.
41983 The system call on the host has been finished.
41987 These two states can be distinguished by the target by the value of the
41988 returned @code{errno}. If it's the protocol representation of @code{EINTR}, the system
41989 call hasn't been performed. This is equivalent to the @code{EINTR} handling
41990 on POSIX systems. In any other case, the target may presume that the
41991 system call has been finished --- successfully or not --- and should behave
41992 as if the break message arrived right after the system call.
41994 @value{GDBN} must behave reliably. If the system call has not been called
41995 yet, @value{GDBN} may send the @code{F} reply immediately, setting @code{EINTR} as
41996 @code{errno} in the packet. If the system call on the host has been finished
41997 before the user requests a break, the full action must be finished by
41998 @value{GDBN}. This requires sending @code{M} or @code{X} packets as necessary.
41999 The @code{F} packet may only be sent when either nothing has happened
42000 or the full action has been completed.
42003 @subsection Console I/O
42004 @cindex console i/o as part of file-i/o
42006 By default and if not explicitly closed by the target system, the file
42007 descriptors 0, 1 and 2 are connected to the @value{GDBN} console. Output
42008 on the @value{GDBN} console is handled as any other file output operation
42009 (@code{write(1, @dots{})} or @code{write(2, @dots{})}). Console input is handled
42010 by @value{GDBN} so that after the target read request from file descriptor
42011 0 all following typing is buffered until either one of the following
42016 The user types @kbd{Ctrl-c}. The behaviour is as explained above, and the
42018 system call is treated as finished.
42021 The user presses @key{RET}. This is treated as end of input with a trailing
42025 The user types @kbd{Ctrl-d}. This is treated as end of input. No trailing
42026 character (neither newline nor @samp{Ctrl-D}) is appended to the input.
42030 If the user has typed more characters than fit in the buffer given to
42031 the @code{read} call, the trailing characters are buffered in @value{GDBN} until
42032 either another @code{read(0, @dots{})} is requested by the target, or debugging
42033 is stopped at the user's request.
42036 @node List of Supported Calls
42037 @subsection List of Supported Calls
42038 @cindex list of supported file-i/o calls
42055 @unnumberedsubsubsec open
42056 @cindex open, file-i/o system call
42061 int open(const char *pathname, int flags);
42062 int open(const char *pathname, int flags, mode_t mode);
42066 @samp{Fopen,@var{pathptr}/@var{len},@var{flags},@var{mode}}
42069 @var{flags} is the bitwise @code{OR} of the following values:
42073 If the file does not exist it will be created. The host
42074 rules apply as far as file ownership and time stamps
42078 When used with @code{O_CREAT}, if the file already exists it is
42079 an error and open() fails.
42082 If the file already exists and the open mode allows
42083 writing (@code{O_RDWR} or @code{O_WRONLY} is given) it will be
42084 truncated to zero length.
42087 The file is opened in append mode.
42090 The file is opened for reading only.
42093 The file is opened for writing only.
42096 The file is opened for reading and writing.
42100 Other bits are silently ignored.
42104 @var{mode} is the bitwise @code{OR} of the following values:
42108 User has read permission.
42111 User has write permission.
42114 Group has read permission.
42117 Group has write permission.
42120 Others have read permission.
42123 Others have write permission.
42127 Other bits are silently ignored.
42130 @item Return value:
42131 @code{open} returns the new file descriptor or -1 if an error
42138 @var{pathname} already exists and @code{O_CREAT} and @code{O_EXCL} were used.
42141 @var{pathname} refers to a directory.
42144 The requested access is not allowed.
42147 @var{pathname} was too long.
42150 A directory component in @var{pathname} does not exist.
42153 @var{pathname} refers to a device, pipe, named pipe or socket.
42156 @var{pathname} refers to a file on a read-only filesystem and
42157 write access was requested.
42160 @var{pathname} is an invalid pointer value.
42163 No space on device to create the file.
42166 The process already has the maximum number of files open.
42169 The limit on the total number of files open on the system
42173 The call was interrupted by the user.
42179 @unnumberedsubsubsec close
42180 @cindex close, file-i/o system call
42189 @samp{Fclose,@var{fd}}
42191 @item Return value:
42192 @code{close} returns zero on success, or -1 if an error occurred.
42198 @var{fd} isn't a valid open file descriptor.
42201 The call was interrupted by the user.
42207 @unnumberedsubsubsec read
42208 @cindex read, file-i/o system call
42213 int read(int fd, void *buf, unsigned int count);
42217 @samp{Fread,@var{fd},@var{bufptr},@var{count}}
42219 @item Return value:
42220 On success, the number of bytes read is returned.
42221 Zero indicates end of file. If count is zero, read
42222 returns zero as well. On error, -1 is returned.
42228 @var{fd} is not a valid file descriptor or is not open for
42232 @var{bufptr} is an invalid pointer value.
42235 The call was interrupted by the user.
42241 @unnumberedsubsubsec write
42242 @cindex write, file-i/o system call
42247 int write(int fd, const void *buf, unsigned int count);
42251 @samp{Fwrite,@var{fd},@var{bufptr},@var{count}}
42253 @item Return value:
42254 On success, the number of bytes written are returned.
42255 Zero indicates nothing was written. On error, -1
42262 @var{fd} is not a valid file descriptor or is not open for
42266 @var{bufptr} is an invalid pointer value.
42269 An attempt was made to write a file that exceeds the
42270 host-specific maximum file size allowed.
42273 No space on device to write the data.
42276 The call was interrupted by the user.
42282 @unnumberedsubsubsec lseek
42283 @cindex lseek, file-i/o system call
42288 long lseek (int fd, long offset, int flag);
42292 @samp{Flseek,@var{fd},@var{offset},@var{flag}}
42294 @var{flag} is one of:
42298 The offset is set to @var{offset} bytes.
42301 The offset is set to its current location plus @var{offset}
42305 The offset is set to the size of the file plus @var{offset}
42309 @item Return value:
42310 On success, the resulting unsigned offset in bytes from
42311 the beginning of the file is returned. Otherwise, a
42312 value of -1 is returned.
42318 @var{fd} is not a valid open file descriptor.
42321 @var{fd} is associated with the @value{GDBN} console.
42324 @var{flag} is not a proper value.
42327 The call was interrupted by the user.
42333 @unnumberedsubsubsec rename
42334 @cindex rename, file-i/o system call
42339 int rename(const char *oldpath, const char *newpath);
42343 @samp{Frename,@var{oldpathptr}/@var{len},@var{newpathptr}/@var{len}}
42345 @item Return value:
42346 On success, zero is returned. On error, -1 is returned.
42352 @var{newpath} is an existing directory, but @var{oldpath} is not a
42356 @var{newpath} is a non-empty directory.
42359 @var{oldpath} or @var{newpath} is a directory that is in use by some
42363 An attempt was made to make a directory a subdirectory
42367 A component used as a directory in @var{oldpath} or new
42368 path is not a directory. Or @var{oldpath} is a directory
42369 and @var{newpath} exists but is not a directory.
42372 @var{oldpathptr} or @var{newpathptr} are invalid pointer values.
42375 No access to the file or the path of the file.
42379 @var{oldpath} or @var{newpath} was too long.
42382 A directory component in @var{oldpath} or @var{newpath} does not exist.
42385 The file is on a read-only filesystem.
42388 The device containing the file has no room for the new
42392 The call was interrupted by the user.
42398 @unnumberedsubsubsec unlink
42399 @cindex unlink, file-i/o system call
42404 int unlink(const char *pathname);
42408 @samp{Funlink,@var{pathnameptr}/@var{len}}
42410 @item Return value:
42411 On success, zero is returned. On error, -1 is returned.
42417 No access to the file or the path of the file.
42420 The system does not allow unlinking of directories.
42423 The file @var{pathname} cannot be unlinked because it's
42424 being used by another process.
42427 @var{pathnameptr} is an invalid pointer value.
42430 @var{pathname} was too long.
42433 A directory component in @var{pathname} does not exist.
42436 A component of the path is not a directory.
42439 The file is on a read-only filesystem.
42442 The call was interrupted by the user.
42448 @unnumberedsubsubsec stat/fstat
42449 @cindex fstat, file-i/o system call
42450 @cindex stat, file-i/o system call
42455 int stat(const char *pathname, struct stat *buf);
42456 int fstat(int fd, struct stat *buf);
42460 @samp{Fstat,@var{pathnameptr}/@var{len},@var{bufptr}}@*
42461 @samp{Ffstat,@var{fd},@var{bufptr}}
42463 @item Return value:
42464 On success, zero is returned. On error, -1 is returned.
42470 @var{fd} is not a valid open file.
42473 A directory component in @var{pathname} does not exist or the
42474 path is an empty string.
42477 A component of the path is not a directory.
42480 @var{pathnameptr} is an invalid pointer value.
42483 No access to the file or the path of the file.
42486 @var{pathname} was too long.
42489 The call was interrupted by the user.
42495 @unnumberedsubsubsec gettimeofday
42496 @cindex gettimeofday, file-i/o system call
42501 int gettimeofday(struct timeval *tv, void *tz);
42505 @samp{Fgettimeofday,@var{tvptr},@var{tzptr}}
42507 @item Return value:
42508 On success, 0 is returned, -1 otherwise.
42514 @var{tz} is a non-NULL pointer.
42517 @var{tvptr} and/or @var{tzptr} is an invalid pointer value.
42523 @unnumberedsubsubsec isatty
42524 @cindex isatty, file-i/o system call
42529 int isatty(int fd);
42533 @samp{Fisatty,@var{fd}}
42535 @item Return value:
42536 Returns 1 if @var{fd} refers to the @value{GDBN} console, 0 otherwise.
42542 The call was interrupted by the user.
42547 Note that the @code{isatty} call is treated as a special case: it returns
42548 1 to the target if the file descriptor is attached
42549 to the @value{GDBN} console, 0 otherwise. Implementing through system calls
42550 would require implementing @code{ioctl} and would be more complex than
42555 @unnumberedsubsubsec system
42556 @cindex system, file-i/o system call
42561 int system(const char *command);
42565 @samp{Fsystem,@var{commandptr}/@var{len}}
42567 @item Return value:
42568 If @var{len} is zero, the return value indicates whether a shell is
42569 available. A zero return value indicates a shell is not available.
42570 For non-zero @var{len}, the value returned is -1 on error and the
42571 return status of the command otherwise. Only the exit status of the
42572 command is returned, which is extracted from the host's @code{system}
42573 return value by calling @code{WEXITSTATUS(retval)}. In case
42574 @file{/bin/sh} could not be executed, 127 is returned.
42580 The call was interrupted by the user.
42585 @value{GDBN} takes over the full task of calling the necessary host calls
42586 to perform the @code{system} call. The return value of @code{system} on
42587 the host is simplified before it's returned
42588 to the target. Any termination signal information from the child process
42589 is discarded, and the return value consists
42590 entirely of the exit status of the called command.
42592 Due to security concerns, the @code{system} call is by default refused
42593 by @value{GDBN}. The user has to allow this call explicitly with the
42594 @code{set remote system-call-allowed 1} command.
42597 @item set remote system-call-allowed
42598 @kindex set remote system-call-allowed
42599 Control whether to allow the @code{system} calls in the File I/O
42600 protocol for the remote target. The default is zero (disabled).
42602 @item show remote system-call-allowed
42603 @kindex show remote system-call-allowed
42604 Show whether the @code{system} calls are allowed in the File I/O
42608 @node Protocol-specific Representation of Datatypes
42609 @subsection Protocol-specific Representation of Datatypes
42610 @cindex protocol-specific representation of datatypes, in file-i/o protocol
42613 * Integral Datatypes::
42615 * Memory Transfer::
42620 @node Integral Datatypes
42621 @unnumberedsubsubsec Integral Datatypes
42622 @cindex integral datatypes, in file-i/o protocol
42624 The integral datatypes used in the system calls are @code{int},
42625 @code{unsigned int}, @code{long}, @code{unsigned long},
42626 @code{mode_t}, and @code{time_t}.
42628 @code{int}, @code{unsigned int}, @code{mode_t} and @code{time_t} are
42629 implemented as 32 bit values in this protocol.
42631 @code{long} and @code{unsigned long} are implemented as 64 bit types.
42633 @xref{Limits}, for corresponding MIN and MAX values (similar to those
42634 in @file{limits.h}) to allow range checking on host and target.
42636 @code{time_t} datatypes are defined as seconds since the Epoch.
42638 All integral datatypes transferred as part of a memory read or write of a
42639 structured datatype e.g.@: a @code{struct stat} have to be given in big endian
42642 @node Pointer Values
42643 @unnumberedsubsubsec Pointer Values
42644 @cindex pointer values, in file-i/o protocol
42646 Pointers to target data are transmitted as they are. An exception
42647 is made for pointers to buffers for which the length isn't
42648 transmitted as part of the function call, namely strings. Strings
42649 are transmitted as a pointer/length pair, both as hex values, e.g.@:
42656 which is a pointer to data of length 18 bytes at position 0x1aaf.
42657 The length is defined as the full string length in bytes, including
42658 the trailing null byte. For example, the string @code{"hello world"}
42659 at address 0x123456 is transmitted as
42665 @node Memory Transfer
42666 @unnumberedsubsubsec Memory Transfer
42667 @cindex memory transfer, in file-i/o protocol
42669 Structured data which is transferred using a memory read or write (for
42670 example, a @code{struct stat}) is expected to be in a protocol-specific format
42671 with all scalar multibyte datatypes being big endian. Translation to
42672 this representation needs to be done both by the target before the @code{F}
42673 packet is sent, and by @value{GDBN} before
42674 it transfers memory to the target. Transferred pointers to structured
42675 data should point to the already-coerced data at any time.
42679 @unnumberedsubsubsec struct stat
42680 @cindex struct stat, in file-i/o protocol
42682 The buffer of type @code{struct stat} used by the target and @value{GDBN}
42683 is defined as follows:
42687 unsigned int st_dev; /* device */
42688 unsigned int st_ino; /* inode */
42689 mode_t st_mode; /* protection */
42690 unsigned int st_nlink; /* number of hard links */
42691 unsigned int st_uid; /* user ID of owner */
42692 unsigned int st_gid; /* group ID of owner */
42693 unsigned int st_rdev; /* device type (if inode device) */
42694 unsigned long st_size; /* total size, in bytes */
42695 unsigned long st_blksize; /* blocksize for filesystem I/O */
42696 unsigned long st_blocks; /* number of blocks allocated */
42697 time_t st_atime; /* time of last access */
42698 time_t st_mtime; /* time of last modification */
42699 time_t st_ctime; /* time of last change */
42703 The integral datatypes conform to the definitions given in the
42704 appropriate section (see @ref{Integral Datatypes}, for details) so this
42705 structure is of size 64 bytes.
42707 The values of several fields have a restricted meaning and/or
42713 A value of 0 represents a file, 1 the console.
42716 No valid meaning for the target. Transmitted unchanged.
42719 Valid mode bits are described in @ref{Constants}. Any other
42720 bits have currently no meaning for the target.
42725 No valid meaning for the target. Transmitted unchanged.
42730 These values have a host and file system dependent
42731 accuracy. Especially on Windows hosts, the file system may not
42732 support exact timing values.
42735 The target gets a @code{struct stat} of the above representation and is
42736 responsible for coercing it to the target representation before
42739 Note that due to size differences between the host, target, and protocol
42740 representations of @code{struct stat} members, these members could eventually
42741 get truncated on the target.
42743 @node struct timeval
42744 @unnumberedsubsubsec struct timeval
42745 @cindex struct timeval, in file-i/o protocol
42747 The buffer of type @code{struct timeval} used by the File-I/O protocol
42748 is defined as follows:
42752 time_t tv_sec; /* second */
42753 long tv_usec; /* microsecond */
42757 The integral datatypes conform to the definitions given in the
42758 appropriate section (see @ref{Integral Datatypes}, for details) so this
42759 structure is of size 8 bytes.
42762 @subsection Constants
42763 @cindex constants, in file-i/o protocol
42765 The following values are used for the constants inside of the
42766 protocol. @value{GDBN} and target are responsible for translating these
42767 values before and after the call as needed.
42778 @unnumberedsubsubsec Open Flags
42779 @cindex open flags, in file-i/o protocol
42781 All values are given in hexadecimal representation.
42793 @node mode_t Values
42794 @unnumberedsubsubsec mode_t Values
42795 @cindex mode_t values, in file-i/o protocol
42797 All values are given in octal representation.
42814 @unnumberedsubsubsec Errno Values
42815 @cindex errno values, in file-i/o protocol
42817 All values are given in decimal representation.
42842 @code{EUNKNOWN} is used as a fallback error value if a host system returns
42843 any error value not in the list of supported error numbers.
42846 @unnumberedsubsubsec Lseek Flags
42847 @cindex lseek flags, in file-i/o protocol
42856 @unnumberedsubsubsec Limits
42857 @cindex limits, in file-i/o protocol
42859 All values are given in decimal representation.
42862 INT_MIN -2147483648
42864 UINT_MAX 4294967295
42865 LONG_MIN -9223372036854775808
42866 LONG_MAX 9223372036854775807
42867 ULONG_MAX 18446744073709551615
42870 @node File-I/O Examples
42871 @subsection File-I/O Examples
42872 @cindex file-i/o examples
42874 Example sequence of a write call, file descriptor 3, buffer is at target
42875 address 0x1234, 6 bytes should be written:
42878 <- @code{Fwrite,3,1234,6}
42879 @emph{request memory read from target}
42882 @emph{return "6 bytes written"}
42886 Example sequence of a read call, file descriptor 3, buffer is at target
42887 address 0x1234, 6 bytes should be read:
42890 <- @code{Fread,3,1234,6}
42891 @emph{request memory write to target}
42892 -> @code{X1234,6:XXXXXX}
42893 @emph{return "6 bytes read"}
42897 Example sequence of a read call, call fails on the host due to invalid
42898 file descriptor (@code{EBADF}):
42901 <- @code{Fread,3,1234,6}
42905 Example sequence of a read call, user presses @kbd{Ctrl-c} before syscall on
42909 <- @code{Fread,3,1234,6}
42914 Example sequence of a read call, user presses @kbd{Ctrl-c} after syscall on
42918 <- @code{Fread,3,1234,6}
42919 -> @code{X1234,6:XXXXXX}
42923 @node Library List Format
42924 @section Library List Format
42925 @cindex library list format, remote protocol
42927 On some platforms, a dynamic loader (e.g.@: @file{ld.so}) runs in the
42928 same process as your application to manage libraries. In this case,
42929 @value{GDBN} can use the loader's symbol table and normal memory
42930 operations to maintain a list of shared libraries. On other
42931 platforms, the operating system manages loaded libraries.
42932 @value{GDBN} can not retrieve the list of currently loaded libraries
42933 through memory operations, so it uses the @samp{qXfer:libraries:read}
42934 packet (@pxref{qXfer library list read}) instead. The remote stub
42935 queries the target's operating system and reports which libraries
42938 The @samp{qXfer:libraries:read} packet returns an XML document which
42939 lists loaded libraries and their offsets. Each library has an
42940 associated name and one or more segment or section base addresses,
42941 which report where the library was loaded in memory.
42943 For the common case of libraries that are fully linked binaries, the
42944 library should have a list of segments. If the target supports
42945 dynamic linking of a relocatable object file, its library XML element
42946 should instead include a list of allocated sections. The segment or
42947 section bases are start addresses, not relocation offsets; they do not
42948 depend on the library's link-time base addresses.
42950 @value{GDBN} must be linked with the Expat library to support XML
42951 library lists. @xref{Expat}.
42953 A simple memory map, with one loaded library relocated by a single
42954 offset, looks like this:
42958 <library name="/lib/libc.so.6">
42959 <segment address="0x10000000"/>
42964 Another simple memory map, with one loaded library with three
42965 allocated sections (.text, .data, .bss), looks like this:
42969 <library name="sharedlib.o">
42970 <section address="0x10000000"/>
42971 <section address="0x20000000"/>
42972 <section address="0x30000000"/>
42977 The format of a library list is described by this DTD:
42980 <!-- library-list: Root element with versioning -->
42981 <!ELEMENT library-list (library)*>
42982 <!ATTLIST library-list version CDATA #FIXED "1.0">
42983 <!ELEMENT library (segment*, section*)>
42984 <!ATTLIST library name CDATA #REQUIRED>
42985 <!ELEMENT segment EMPTY>
42986 <!ATTLIST segment address CDATA #REQUIRED>
42987 <!ELEMENT section EMPTY>
42988 <!ATTLIST section address CDATA #REQUIRED>
42991 In addition, segments and section descriptors cannot be mixed within a
42992 single library element, and you must supply at least one segment or
42993 section for each library.
42995 @node Library List Format for SVR4 Targets
42996 @section Library List Format for SVR4 Targets
42997 @cindex library list format, remote protocol
42999 On SVR4 platforms @value{GDBN} can use the symbol table of a dynamic loader
43000 (e.g.@: @file{ld.so}) and normal memory operations to maintain a list of
43001 shared libraries. Still a special library list provided by this packet is
43002 more efficient for the @value{GDBN} remote protocol.
43004 The @samp{qXfer:libraries-svr4:read} packet returns an XML document which lists
43005 loaded libraries and their SVR4 linker parameters. For each library on SVR4
43006 target, the following parameters are reported:
43010 @code{name}, the absolute file name from the @code{l_name} field of
43011 @code{struct link_map}.
43013 @code{lm} with address of @code{struct link_map} used for TLS
43014 (Thread Local Storage) access.
43016 @code{l_addr}, the displacement as read from the field @code{l_addr} of
43017 @code{struct link_map}. For prelinked libraries this is not an absolute
43018 memory address. It is a displacement of absolute memory address against
43019 address the file was prelinked to during the library load.
43021 @code{l_ld}, which is memory address of the @code{PT_DYNAMIC} segment
43024 Additionally the single @code{main-lm} attribute specifies address of
43025 @code{struct link_map} used for the main executable. This parameter is used
43026 for TLS access and its presence is optional.
43028 @value{GDBN} must be linked with the Expat library to support XML
43029 SVR4 library lists. @xref{Expat}.
43031 A simple memory map, with two loaded libraries (which do not use prelink),
43035 <library-list-svr4 version="1.0" main-lm="0xe4f8f8">
43036 <library name="/lib/ld-linux.so.2" lm="0xe4f51c" l_addr="0xe2d000"
43038 <library name="/lib/libc.so.6" lm="0xe4fbe8" l_addr="0x154000"
43040 </library-list-svr>
43043 The format of an SVR4 library list is described by this DTD:
43046 <!-- library-list-svr4: Root element with versioning -->
43047 <!ELEMENT library-list-svr4 (library)*>
43048 <!ATTLIST library-list-svr4 version CDATA #FIXED "1.0">
43049 <!ATTLIST library-list-svr4 main-lm CDATA #IMPLIED>
43050 <!ELEMENT library EMPTY>
43051 <!ATTLIST library name CDATA #REQUIRED>
43052 <!ATTLIST library lm CDATA #REQUIRED>
43053 <!ATTLIST library l_addr CDATA #REQUIRED>
43054 <!ATTLIST library l_ld CDATA #REQUIRED>
43057 @node Memory Map Format
43058 @section Memory Map Format
43059 @cindex memory map format
43061 To be able to write into flash memory, @value{GDBN} needs to obtain a
43062 memory map from the target. This section describes the format of the
43065 The memory map is obtained using the @samp{qXfer:memory-map:read}
43066 (@pxref{qXfer memory map read}) packet and is an XML document that
43067 lists memory regions.
43069 @value{GDBN} must be linked with the Expat library to support XML
43070 memory maps. @xref{Expat}.
43072 The top-level structure of the document is shown below:
43075 <?xml version="1.0"?>
43076 <!DOCTYPE memory-map
43077 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
43078 "http://sourceware.org/gdb/gdb-memory-map.dtd">
43084 Each region can be either:
43089 A region of RAM starting at @var{addr} and extending for @var{length}
43093 <memory type="ram" start="@var{addr}" length="@var{length}"/>
43098 A region of read-only memory:
43101 <memory type="rom" start="@var{addr}" length="@var{length}"/>
43106 A region of flash memory, with erasure blocks @var{blocksize}
43110 <memory type="flash" start="@var{addr}" length="@var{length}">
43111 <property name="blocksize">@var{blocksize}</property>
43117 Regions must not overlap. @value{GDBN} assumes that areas of memory not covered
43118 by the memory map are RAM, and uses the ordinary @samp{M} and @samp{X}
43119 packets to write to addresses in such ranges.
43121 The formal DTD for memory map format is given below:
43124 <!-- ................................................... -->
43125 <!-- Memory Map XML DTD ................................ -->
43126 <!-- File: memory-map.dtd .............................. -->
43127 <!-- .................................... .............. -->
43128 <!-- memory-map.dtd -->
43129 <!-- memory-map: Root element with versioning -->
43130 <!ELEMENT memory-map (memory)*>
43131 <!ATTLIST memory-map version CDATA #FIXED "1.0.0">
43132 <!ELEMENT memory (property)*>
43133 <!-- memory: Specifies a memory region,
43134 and its type, or device. -->
43135 <!ATTLIST memory type (ram|rom|flash) #REQUIRED
43136 start CDATA #REQUIRED
43137 length CDATA #REQUIRED>
43138 <!-- property: Generic attribute tag -->
43139 <!ELEMENT property (#PCDATA | property)*>
43140 <!ATTLIST property name (blocksize) #REQUIRED>
43143 @node Thread List Format
43144 @section Thread List Format
43145 @cindex thread list format
43147 To efficiently update the list of threads and their attributes,
43148 @value{GDBN} issues the @samp{qXfer:threads:read} packet
43149 (@pxref{qXfer threads read}) and obtains the XML document with
43150 the following structure:
43153 <?xml version="1.0"?>
43155 <thread id="id" core="0" name="name">
43156 ... description ...
43161 Each @samp{thread} element must have the @samp{id} attribute that
43162 identifies the thread (@pxref{thread-id syntax}). The
43163 @samp{core} attribute, if present, specifies which processor core
43164 the thread was last executing on. The @samp{name} attribute, if
43165 present, specifies the human-readable name of the thread. The content
43166 of the of @samp{thread} element is interpreted as human-readable
43167 auxiliary information. The @samp{handle} attribute, if present,
43168 is a hex encoded representation of the thread handle.
43171 @node Traceframe Info Format
43172 @section Traceframe Info Format
43173 @cindex traceframe info format
43175 To be able to know which objects in the inferior can be examined when
43176 inspecting a tracepoint hit, @value{GDBN} needs to obtain the list of
43177 memory ranges, registers and trace state variables that have been
43178 collected in a traceframe.
43180 This list is obtained using the @samp{qXfer:traceframe-info:read}
43181 (@pxref{qXfer traceframe info read}) packet and is an XML document.
43183 @value{GDBN} must be linked with the Expat library to support XML
43184 traceframe info discovery. @xref{Expat}.
43186 The top-level structure of the document is shown below:
43189 <?xml version="1.0"?>
43190 <!DOCTYPE traceframe-info
43191 PUBLIC "+//IDN gnu.org//DTD GDB Memory Map V1.0//EN"
43192 "http://sourceware.org/gdb/gdb-traceframe-info.dtd">
43198 Each traceframe block can be either:
43203 A region of collected memory starting at @var{addr} and extending for
43204 @var{length} bytes from there:
43207 <memory start="@var{addr}" length="@var{length}"/>
43211 A block indicating trace state variable numbered @var{number} has been
43215 <tvar id="@var{number}"/>
43220 The formal DTD for the traceframe info format is given below:
43223 <!ELEMENT traceframe-info (memory | tvar)* >
43224 <!ATTLIST traceframe-info version CDATA #FIXED "1.0">
43226 <!ELEMENT memory EMPTY>
43227 <!ATTLIST memory start CDATA #REQUIRED
43228 length CDATA #REQUIRED>
43230 <!ATTLIST tvar id CDATA #REQUIRED>
43233 @node Branch Trace Format
43234 @section Branch Trace Format
43235 @cindex branch trace format
43237 In order to display the branch trace of an inferior thread,
43238 @value{GDBN} needs to obtain the list of branches. This list is
43239 represented as list of sequential code blocks that are connected via
43240 branches. The code in each block has been executed sequentially.
43242 This list is obtained using the @samp{qXfer:btrace:read}
43243 (@pxref{qXfer btrace read}) packet and is an XML document.
43245 @value{GDBN} must be linked with the Expat library to support XML
43246 traceframe info discovery. @xref{Expat}.
43248 The top-level structure of the document is shown below:
43251 <?xml version="1.0"?>
43253 PUBLIC "+//IDN gnu.org//DTD GDB Branch Trace V1.0//EN"
43254 "http://sourceware.org/gdb/gdb-btrace.dtd">
43263 A block of sequentially executed instructions starting at @var{begin}
43264 and ending at @var{end}:
43267 <block begin="@var{begin}" end="@var{end}"/>
43272 The formal DTD for the branch trace format is given below:
43275 <!ELEMENT btrace (block* | pt) >
43276 <!ATTLIST btrace version CDATA #FIXED "1.0">
43278 <!ELEMENT block EMPTY>
43279 <!ATTLIST block begin CDATA #REQUIRED
43280 end CDATA #REQUIRED>
43282 <!ELEMENT pt (pt-config?, raw?)>
43284 <!ELEMENT pt-config (cpu?)>
43286 <!ELEMENT cpu EMPTY>
43287 <!ATTLIST cpu vendor CDATA #REQUIRED
43288 family CDATA #REQUIRED
43289 model CDATA #REQUIRED
43290 stepping CDATA #REQUIRED>
43292 <!ELEMENT raw (#PCDATA)>
43295 @node Branch Trace Configuration Format
43296 @section Branch Trace Configuration Format
43297 @cindex branch trace configuration format
43299 For each inferior thread, @value{GDBN} can obtain the branch trace
43300 configuration using the @samp{qXfer:btrace-conf:read}
43301 (@pxref{qXfer btrace-conf read}) packet.
43303 The configuration describes the branch trace format and configuration
43304 settings for that format. The following information is described:
43308 This thread uses the @dfn{Branch Trace Store} (@acronym{BTS}) format.
43311 The size of the @acronym{BTS} ring buffer in bytes.
43314 This thread uses the @dfn{Intel Processor Trace} (@acronym{Intel
43318 The size of the @acronym{Intel PT} ring buffer in bytes.
43322 @value{GDBN} must be linked with the Expat library to support XML
43323 branch trace configuration discovery. @xref{Expat}.
43325 The formal DTD for the branch trace configuration format is given below:
43328 <!ELEMENT btrace-conf (bts?, pt?)>
43329 <!ATTLIST btrace-conf version CDATA #FIXED "1.0">
43331 <!ELEMENT bts EMPTY>
43332 <!ATTLIST bts size CDATA #IMPLIED>
43334 <!ELEMENT pt EMPTY>
43335 <!ATTLIST pt size CDATA #IMPLIED>
43338 @include agentexpr.texi
43340 @node Target Descriptions
43341 @appendix Target Descriptions
43342 @cindex target descriptions
43344 One of the challenges of using @value{GDBN} to debug embedded systems
43345 is that there are so many minor variants of each processor
43346 architecture in use. It is common practice for vendors to start with
43347 a standard processor core --- ARM, PowerPC, or @acronym{MIPS}, for example ---
43348 and then make changes to adapt it to a particular market niche. Some
43349 architectures have hundreds of variants, available from dozens of
43350 vendors. This leads to a number of problems:
43354 With so many different customized processors, it is difficult for
43355 the @value{GDBN} maintainers to keep up with the changes.
43357 Since individual variants may have short lifetimes or limited
43358 audiences, it may not be worthwhile to carry information about every
43359 variant in the @value{GDBN} source tree.
43361 When @value{GDBN} does support the architecture of the embedded system
43362 at hand, the task of finding the correct architecture name to give the
43363 @command{set architecture} command can be error-prone.
43366 To address these problems, the @value{GDBN} remote protocol allows a
43367 target system to not only identify itself to @value{GDBN}, but to
43368 actually describe its own features. This lets @value{GDBN} support
43369 processor variants it has never seen before --- to the extent that the
43370 descriptions are accurate, and that @value{GDBN} understands them.
43372 @value{GDBN} must be linked with the Expat library to support XML
43373 target descriptions. @xref{Expat}.
43376 * Retrieving Descriptions:: How descriptions are fetched from a target.
43377 * Target Description Format:: The contents of a target description.
43378 * Predefined Target Types:: Standard types available for target
43380 * Enum Target Types:: How to define enum target types.
43381 * Standard Target Features:: Features @value{GDBN} knows about.
43384 @node Retrieving Descriptions
43385 @section Retrieving Descriptions
43387 Target descriptions can be read from the target automatically, or
43388 specified by the user manually. The default behavior is to read the
43389 description from the target. @value{GDBN} retrieves it via the remote
43390 protocol using @samp{qXfer} requests (@pxref{General Query Packets,
43391 qXfer}). The @var{annex} in the @samp{qXfer} packet will be
43392 @samp{target.xml}. The contents of the @samp{target.xml} annex are an
43393 XML document, of the form described in @ref{Target Description
43396 Alternatively, you can specify a file to read for the target description.
43397 If a file is set, the target will not be queried. The commands to
43398 specify a file are:
43401 @cindex set tdesc filename
43402 @item set tdesc filename @var{path}
43403 Read the target description from @var{path}.
43405 @cindex unset tdesc filename
43406 @item unset tdesc filename
43407 Do not read the XML target description from a file. @value{GDBN}
43408 will use the description supplied by the current target.
43410 @cindex show tdesc filename
43411 @item show tdesc filename
43412 Show the filename to read for a target description, if any.
43416 @node Target Description Format
43417 @section Target Description Format
43418 @cindex target descriptions, XML format
43420 A target description annex is an @uref{http://www.w3.org/XML/, XML}
43421 document which complies with the Document Type Definition provided in
43422 the @value{GDBN} sources in @file{gdb/features/gdb-target.dtd}. This
43423 means you can use generally available tools like @command{xmllint} to
43424 check that your feature descriptions are well-formed and valid.
43425 However, to help people unfamiliar with XML write descriptions for
43426 their targets, we also describe the grammar here.
43428 Target descriptions can identify the architecture of the remote target
43429 and (for some architectures) provide information about custom register
43430 sets. They can also identify the OS ABI of the remote target.
43431 @value{GDBN} can use this information to autoconfigure for your
43432 target, or to warn you if you connect to an unsupported target.
43434 Here is a simple target description:
43437 <target version="1.0">
43438 <architecture>i386:x86-64</architecture>
43443 This minimal description only says that the target uses
43444 the x86-64 architecture.
43446 A target description has the following overall form, with [ ] marking
43447 optional elements and @dots{} marking repeatable elements. The elements
43448 are explained further below.
43451 <?xml version="1.0"?>
43452 <!DOCTYPE target SYSTEM "gdb-target.dtd">
43453 <target version="1.0">
43454 @r{[}@var{architecture}@r{]}
43455 @r{[}@var{osabi}@r{]}
43456 @r{[}@var{compatible}@r{]}
43457 @r{[}@var{feature}@dots{}@r{]}
43462 The description is generally insensitive to whitespace and line
43463 breaks, under the usual common-sense rules. The XML version
43464 declaration and document type declaration can generally be omitted
43465 (@value{GDBN} does not require them), but specifying them may be
43466 useful for XML validation tools. The @samp{version} attribute for
43467 @samp{<target>} may also be omitted, but we recommend
43468 including it; if future versions of @value{GDBN} use an incompatible
43469 revision of @file{gdb-target.dtd}, they will detect and report
43470 the version mismatch.
43472 @subsection Inclusion
43473 @cindex target descriptions, inclusion
43476 @cindex <xi:include>
43479 It can sometimes be valuable to split a target description up into
43480 several different annexes, either for organizational purposes, or to
43481 share files between different possible target descriptions. You can
43482 divide a description into multiple files by replacing any element of
43483 the target description with an inclusion directive of the form:
43486 <xi:include href="@var{document}"/>
43490 When @value{GDBN} encounters an element of this form, it will retrieve
43491 the named XML @var{document}, and replace the inclusion directive with
43492 the contents of that document. If the current description was read
43493 using @samp{qXfer}, then so will be the included document;
43494 @var{document} will be interpreted as the name of an annex. If the
43495 current description was read from a file, @value{GDBN} will look for
43496 @var{document} as a file in the same directory where it found the
43497 original description.
43499 @subsection Architecture
43500 @cindex <architecture>
43502 An @samp{<architecture>} element has this form:
43505 <architecture>@var{arch}</architecture>
43508 @var{arch} is one of the architectures from the set accepted by
43509 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
43512 @cindex @code{<osabi>}
43514 This optional field was introduced in @value{GDBN} version 7.0.
43515 Previous versions of @value{GDBN} ignore it.
43517 An @samp{<osabi>} element has this form:
43520 <osabi>@var{abi-name}</osabi>
43523 @var{abi-name} is an OS ABI name from the same selection accepted by
43524 @w{@code{set osabi}} (@pxref{ABI, ,Configuring the Current ABI}).
43526 @subsection Compatible Architecture
43527 @cindex @code{<compatible>}
43529 This optional field was introduced in @value{GDBN} version 7.0.
43530 Previous versions of @value{GDBN} ignore it.
43532 A @samp{<compatible>} element has this form:
43535 <compatible>@var{arch}</compatible>
43538 @var{arch} is one of the architectures from the set accepted by
43539 @code{set architecture} (@pxref{Targets, ,Specifying a Debugging Target}).
43541 A @samp{<compatible>} element is used to specify that the target
43542 is able to run binaries in some other than the main target architecture
43543 given by the @samp{<architecture>} element. For example, on the
43544 Cell Broadband Engine, the main architecture is @code{powerpc:common}
43545 or @code{powerpc:common64}, but the system is able to run binaries
43546 in the @code{spu} architecture as well. The way to describe this
43547 capability with @samp{<compatible>} is as follows:
43550 <architecture>powerpc:common</architecture>
43551 <compatible>spu</compatible>
43554 @subsection Features
43557 Each @samp{<feature>} describes some logical portion of the target
43558 system. Features are currently used to describe available CPU
43559 registers and the types of their contents. A @samp{<feature>} element
43563 <feature name="@var{name}">
43564 @r{[}@var{type}@dots{}@r{]}
43570 Each feature's name should be unique within the description. The name
43571 of a feature does not matter unless @value{GDBN} has some special
43572 knowledge of the contents of that feature; if it does, the feature
43573 should have its standard name. @xref{Standard Target Features}.
43577 Any register's value is a collection of bits which @value{GDBN} must
43578 interpret. The default interpretation is a two's complement integer,
43579 but other types can be requested by name in the register description.
43580 Some predefined types are provided by @value{GDBN} (@pxref{Predefined
43581 Target Types}), and the description can define additional composite
43584 Each type element must have an @samp{id} attribute, which gives
43585 a unique (within the containing @samp{<feature>}) name to the type.
43586 Types must be defined before they are used.
43589 Some targets offer vector registers, which can be treated as arrays
43590 of scalar elements. These types are written as @samp{<vector>} elements,
43591 specifying the array element type, @var{type}, and the number of elements,
43595 <vector id="@var{id}" type="@var{type}" count="@var{count}"/>
43599 If a register's value is usefully viewed in multiple ways, define it
43600 with a union type containing the useful representations. The
43601 @samp{<union>} element contains one or more @samp{<field>} elements,
43602 each of which has a @var{name} and a @var{type}:
43605 <union id="@var{id}">
43606 <field name="@var{name}" type="@var{type}"/>
43613 If a register's value is composed from several separate values, define
43614 it with either a structure type or a flags type.
43615 A flags type may only contain bitfields.
43616 A structure type may either contain only bitfields or contain no bitfields.
43617 If the value contains only bitfields, its total size in bytes must be
43620 Non-bitfield values have a @var{name} and @var{type}.
43623 <struct id="@var{id}">
43624 <field name="@var{name}" type="@var{type}"/>
43629 Both @var{name} and @var{type} values are required.
43630 No implicit padding is added.
43632 Bitfield values have a @var{name}, @var{start}, @var{end} and @var{type}.
43635 <struct id="@var{id}" size="@var{size}">
43636 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
43642 <flags id="@var{id}" size="@var{size}">
43643 <field name="@var{name}" start="@var{start}" end="@var{end}" type="@var{type}"/>
43648 The @var{name} value is required.
43649 Bitfield values may be named with the empty string, @samp{""},
43650 in which case the field is ``filler'' and its value is not printed.
43651 Not all bits need to be specified, so ``filler'' fields are optional.
43653 The @var{start} and @var{end} values are required, and @var{type}
43655 The field's @var{start} must be less than or equal to its @var{end},
43656 and zero represents the least significant bit.
43658 The default value of @var{type} is @code{bool} for single bit fields,
43659 and an unsigned integer otherwise.
43661 Which to choose? Structures or flags?
43663 Registers defined with @samp{flags} have these advantages over
43664 defining them with @samp{struct}:
43668 Arithmetic may be performed on them as if they were integers.
43670 They are printed in a more readable fashion.
43673 Registers defined with @samp{struct} have one advantage over
43674 defining them with @samp{flags}:
43678 One can fetch individual fields like in @samp{C}.
43681 (gdb) print $my_struct_reg.field3
43687 @subsection Registers
43690 Each register is represented as an element with this form:
43693 <reg name="@var{name}"
43694 bitsize="@var{size}"
43695 @r{[}regnum="@var{num}"@r{]}
43696 @r{[}save-restore="@var{save-restore}"@r{]}
43697 @r{[}type="@var{type}"@r{]}
43698 @r{[}group="@var{group}"@r{]}/>
43702 The components are as follows:
43707 The register's name; it must be unique within the target description.
43710 The register's size, in bits.
43713 The register's number. If omitted, a register's number is one greater
43714 than that of the previous register (either in the current feature or in
43715 a preceding feature); the first register in the target description
43716 defaults to zero. This register number is used to read or write
43717 the register; e.g.@: it is used in the remote @code{p} and @code{P}
43718 packets, and registers appear in the @code{g} and @code{G} packets
43719 in order of increasing register number.
43722 Whether the register should be preserved across inferior function
43723 calls; this must be either @code{yes} or @code{no}. The default is
43724 @code{yes}, which is appropriate for most registers except for
43725 some system control registers; this is not related to the target's
43729 The type of the register. It may be a predefined type, a type
43730 defined in the current feature, or one of the special types @code{int}
43731 and @code{float}. @code{int} is an integer type of the correct size
43732 for @var{bitsize}, and @code{float} is a floating point type (in the
43733 architecture's normal floating point format) of the correct size for
43734 @var{bitsize}. The default is @code{int}.
43737 The register group to which this register belongs. It can be one of the
43738 standard register groups @code{general}, @code{float}, @code{vector} or an
43739 arbitrary string. Group names should be limited to alphanumeric characters.
43740 If a group name is made up of multiple words the words may be separated by
43741 hyphens; e.g.@: @code{special-group} or @code{ultra-special-group}. If no
43742 @var{group} is specified, @value{GDBN} will not display the register in
43743 @code{info registers}.
43747 @node Predefined Target Types
43748 @section Predefined Target Types
43749 @cindex target descriptions, predefined types
43751 Type definitions in the self-description can build up composite types
43752 from basic building blocks, but can not define fundamental types. Instead,
43753 standard identifiers are provided by @value{GDBN} for the fundamental
43754 types. The currently supported types are:
43759 Boolean type, occupying a single bit.
43767 Signed integer types holding the specified number of bits.
43775 Unsigned integer types holding the specified number of bits.
43779 Pointers to unspecified code and data. The program counter and
43780 any dedicated return address register may be marked as code
43781 pointers; printing a code pointer converts it into a symbolic
43782 address. The stack pointer and any dedicated address registers
43783 may be marked as data pointers.
43786 Single precision IEEE floating point.
43789 Double precision IEEE floating point.
43792 The 12-byte extended precision format used by ARM FPA registers.
43795 The 10-byte extended precision format used by x87 registers.
43798 32bit @sc{eflags} register used by x86.
43801 32bit @sc{mxcsr} register used by x86.
43805 @node Enum Target Types
43806 @section Enum Target Types
43807 @cindex target descriptions, enum types
43809 Enum target types are useful in @samp{struct} and @samp{flags}
43810 register descriptions. @xref{Target Description Format}.
43812 Enum types have a name, size and a list of name/value pairs.
43815 <enum id="@var{id}" size="@var{size}">
43816 <evalue name="@var{name}" value="@var{value}"/>
43821 Enums must be defined before they are used.
43824 <enum id="levels_type" size="4">
43825 <evalue name="low" value="0"/>
43826 <evalue name="high" value="1"/>
43828 <flags id="flags_type" size="4">
43829 <field name="X" start="0"/>
43830 <field name="LEVEL" start="1" end="1" type="levels_type"/>
43832 <reg name="flags" bitsize="32" type="flags_type"/>
43835 Given that description, a value of 3 for the @samp{flags} register
43836 would be printed as:
43839 (gdb) info register flags
43840 flags 0x3 [ X LEVEL=high ]
43843 @node Standard Target Features
43844 @section Standard Target Features
43845 @cindex target descriptions, standard features
43847 A target description must contain either no registers or all the
43848 target's registers. If the description contains no registers, then
43849 @value{GDBN} will assume a default register layout, selected based on
43850 the architecture. If the description contains any registers, the
43851 default layout will not be used; the standard registers must be
43852 described in the target description, in such a way that @value{GDBN}
43853 can recognize them.
43855 This is accomplished by giving specific names to feature elements
43856 which contain standard registers. @value{GDBN} will look for features
43857 with those names and verify that they contain the expected registers;
43858 if any known feature is missing required registers, or if any required
43859 feature is missing, @value{GDBN} will reject the target
43860 description. You can add additional registers to any of the
43861 standard features --- @value{GDBN} will display them just as if
43862 they were added to an unrecognized feature.
43864 This section lists the known features and their expected contents.
43865 Sample XML documents for these features are included in the
43866 @value{GDBN} source tree, in the directory @file{gdb/features}.
43868 Names recognized by @value{GDBN} should include the name of the
43869 company or organization which selected the name, and the overall
43870 architecture to which the feature applies; so e.g.@: the feature
43871 containing ARM core registers is named @samp{org.gnu.gdb.arm.core}.
43873 The names of registers are not case sensitive for the purpose
43874 of recognizing standard features, but @value{GDBN} will only display
43875 registers using the capitalization used in the description.
43878 * AArch64 Features::
43882 * MicroBlaze Features::
43886 * Nios II Features::
43887 * OpenRISC 1000 Features::
43888 * PowerPC Features::
43889 * RISC-V Features::
43890 * S/390 and System z Features::
43896 @node AArch64 Features
43897 @subsection AArch64 Features
43898 @cindex target descriptions, AArch64 features
43900 The @samp{org.gnu.gdb.aarch64.core} feature is required for AArch64
43901 targets. It should contain registers @samp{x0} through @samp{x30},
43902 @samp{sp}, @samp{pc}, and @samp{cpsr}.
43904 The @samp{org.gnu.gdb.aarch64.fpu} feature is optional. If present,
43905 it should contain registers @samp{v0} through @samp{v31}, @samp{fpsr},
43908 The @samp{org.gnu.gdb.aarch64.sve} feature is optional. If present,
43909 it should contain registers @samp{z0} through @samp{z31}, @samp{p0}
43910 through @samp{p15}, @samp{ffr} and @samp{vg}.
43912 The @samp{org.gnu.gdb.aarch64.pauth} feature is optional. If present,
43913 it should contain registers @samp{pauth_dmask} and @samp{pauth_cmask}.
43916 @subsection ARC Features
43917 @cindex target descriptions, ARC Features
43919 ARC processors are highly configurable, so even core registers and their number
43920 are not completely predetermined. In addition flags and PC registers which are
43921 important to @value{GDBN} are not ``core'' registers in ARC. It is required
43922 that one of the core registers features is present.
43923 @samp{org.gnu.gdb.arc.aux-minimal} feature is mandatory.
43925 The @samp{org.gnu.gdb.arc.core.v2} feature is required for ARC EM and ARC HS
43926 targets with a normal register file. It should contain registers @samp{r0}
43927 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
43928 @samp{lp_count} and @samp{pcl}. This feature may contain register @samp{ilink}
43929 and any of extension core registers @samp{r32} through @samp{r59/acch}.
43930 @samp{ilink} and extension core registers are not available to read/write, when
43931 debugging GNU/Linux applications, thus @samp{ilink} is made optional.
43933 The @samp{org.gnu.gdb.arc.core-reduced.v2} feature is required for ARC EM and
43934 ARC HS targets with a reduced register file. It should contain registers
43935 @samp{r0} through @samp{r3}, @samp{r10} through @samp{r15}, @samp{gp},
43936 @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink}, @samp{lp_count} and @samp{pcl}.
43937 This feature may contain register @samp{ilink} and any of extension core
43938 registers @samp{r32} through @samp{r59/acch}.
43940 The @samp{org.gnu.gdb.arc.core.arcompact} feature is required for ARCompact
43941 targets with a normal register file. It should contain registers @samp{r0}
43942 through @samp{r25}, @samp{gp}, @samp{fp}, @samp{sp}, @samp{r30}, @samp{blink},
43943 @samp{lp_count} and @samp{pcl}. This feature may contain registers
43944 @samp{ilink1}, @samp{ilink2} and any of extension core registers @samp{r32}
43945 through @samp{r59/acch}. @samp{ilink1} and @samp{ilink2} and extension core
43946 registers are not available when debugging GNU/Linux applications. The only
43947 difference with @samp{org.gnu.gdb.arc.core.v2} feature is in the names of
43948 @samp{ilink1} and @samp{ilink2} registers and that @samp{r30} is mandatory in
43949 ARC v2, but @samp{ilink2} is optional on ARCompact.
43951 The @samp{org.gnu.gdb.arc.aux-minimal} feature is required for all ARC
43952 targets. It should contain registers @samp{pc} and @samp{status32}.
43955 @subsection ARM Features
43956 @cindex target descriptions, ARM features
43958 The @samp{org.gnu.gdb.arm.core} feature is required for non-M-profile
43960 It should contain registers @samp{r0} through @samp{r13}, @samp{sp},
43961 @samp{lr}, @samp{pc}, and @samp{cpsr}.
43963 For M-profile targets (e.g. Cortex-M3), the @samp{org.gnu.gdb.arm.core}
43964 feature is replaced by @samp{org.gnu.gdb.arm.m-profile}. It should contain
43965 registers @samp{r0} through @samp{r13}, @samp{sp}, @samp{lr}, @samp{pc},
43968 The @samp{org.gnu.gdb.arm.fpa} feature is optional. If present, it
43969 should contain registers @samp{f0} through @samp{f7} and @samp{fps}.
43971 The @samp{org.gnu.gdb.xscale.iwmmxt} feature is optional. If present,
43972 it should contain at least registers @samp{wR0} through @samp{wR15} and
43973 @samp{wCGR0} through @samp{wCGR3}. The @samp{wCID}, @samp{wCon},
43974 @samp{wCSSF}, and @samp{wCASF} registers are optional.
43976 The @samp{org.gnu.gdb.arm.vfp} feature is optional. If present, it
43977 should contain at least registers @samp{d0} through @samp{d15}. If
43978 they are present, @samp{d16} through @samp{d31} should also be included.
43979 @value{GDBN} will synthesize the single-precision registers from
43980 halves of the double-precision registers.
43982 The @samp{org.gnu.gdb.arm.neon} feature is optional. It does not
43983 need to contain registers; it instructs @value{GDBN} to display the
43984 VFP double-precision registers as vectors and to synthesize the
43985 quad-precision registers from pairs of double-precision registers.
43986 If this feature is present, @samp{org.gnu.gdb.arm.vfp} must also
43987 be present and include 32 double-precision registers.
43989 @node i386 Features
43990 @subsection i386 Features
43991 @cindex target descriptions, i386 features
43993 The @samp{org.gnu.gdb.i386.core} feature is required for i386/amd64
43994 targets. It should describe the following registers:
43998 @samp{eax} through @samp{edi} plus @samp{eip} for i386
44000 @samp{rax} through @samp{r15} plus @samp{rip} for amd64
44002 @samp{eflags}, @samp{cs}, @samp{ss}, @samp{ds}, @samp{es},
44003 @samp{fs}, @samp{gs}
44005 @samp{st0} through @samp{st7}
44007 @samp{fctrl}, @samp{fstat}, @samp{ftag}, @samp{fiseg}, @samp{fioff},
44008 @samp{foseg}, @samp{fooff} and @samp{fop}
44011 The register sets may be different, depending on the target.
44013 The @samp{org.gnu.gdb.i386.sse} feature is optional. It should
44014 describe registers:
44018 @samp{xmm0} through @samp{xmm7} for i386
44020 @samp{xmm0} through @samp{xmm15} for amd64
44025 The @samp{org.gnu.gdb.i386.avx} feature is optional and requires the
44026 @samp{org.gnu.gdb.i386.sse} feature. It should
44027 describe the upper 128 bits of @sc{ymm} registers:
44031 @samp{ymm0h} through @samp{ymm7h} for i386
44033 @samp{ymm0h} through @samp{ymm15h} for amd64
44036 The @samp{org.gnu.gdb.i386.mpx} is an optional feature representing Intel
44037 Memory Protection Extension (MPX). It should describe the following registers:
44041 @samp{bnd0raw} through @samp{bnd3raw} for i386 and amd64.
44043 @samp{bndcfgu} and @samp{bndstatus} for i386 and amd64.
44046 The @samp{org.gnu.gdb.i386.linux} feature is optional. It should
44047 describe a single register, @samp{orig_eax}.
44049 The @samp{org.gnu.gdb.i386.segments} feature is optional. It should
44050 describe two system registers: @samp{fs_base} and @samp{gs_base}.
44052 The @samp{org.gnu.gdb.i386.avx512} feature is optional and requires the
44053 @samp{org.gnu.gdb.i386.avx} feature. It should
44054 describe additional @sc{xmm} registers:
44058 @samp{xmm16h} through @samp{xmm31h}, only valid for amd64.
44061 It should describe the upper 128 bits of additional @sc{ymm} registers:
44065 @samp{ymm16h} through @samp{ymm31h}, only valid for amd64.
44069 describe the upper 256 bits of @sc{zmm} registers:
44073 @samp{zmm0h} through @samp{zmm7h} for i386.
44075 @samp{zmm0h} through @samp{zmm15h} for amd64.
44079 describe the additional @sc{zmm} registers:
44083 @samp{zmm16h} through @samp{zmm31h}, only valid for amd64.
44086 The @samp{org.gnu.gdb.i386.pkeys} feature is optional. It should
44087 describe a single register, @samp{pkru}. It is a 32-bit register
44088 valid for i386 and amd64.
44090 @node MicroBlaze Features
44091 @subsection MicroBlaze Features
44092 @cindex target descriptions, MicroBlaze features
44094 The @samp{org.gnu.gdb.microblaze.core} feature is required for MicroBlaze
44095 targets. It should contain registers @samp{r0} through @samp{r31},
44096 @samp{rpc}, @samp{rmsr}, @samp{rear}, @samp{resr}, @samp{rfsr}, @samp{rbtr},
44097 @samp{rpvr}, @samp{rpvr1} through @samp{rpvr11}, @samp{redr}, @samp{rpid},
44098 @samp{rzpr}, @samp{rtlbx}, @samp{rtlbsx}, @samp{rtlblo}, and @samp{rtlbhi}.
44100 The @samp{org.gnu.gdb.microblaze.stack-protect} feature is optional.
44101 If present, it should contain registers @samp{rshr} and @samp{rslr}
44103 @node MIPS Features
44104 @subsection @acronym{MIPS} Features
44105 @cindex target descriptions, @acronym{MIPS} features
44107 The @samp{org.gnu.gdb.mips.cpu} feature is required for @acronym{MIPS} targets.
44108 It should contain registers @samp{r0} through @samp{r31}, @samp{lo},
44109 @samp{hi}, and @samp{pc}. They may be 32-bit or 64-bit depending
44112 The @samp{org.gnu.gdb.mips.cp0} feature is also required. It should
44113 contain at least the @samp{status}, @samp{badvaddr}, and @samp{cause}
44114 registers. They may be 32-bit or 64-bit depending on the target.
44116 The @samp{org.gnu.gdb.mips.fpu} feature is currently required, though
44117 it may be optional in a future version of @value{GDBN}. It should
44118 contain registers @samp{f0} through @samp{f31}, @samp{fcsr}, and
44119 @samp{fir}. They may be 32-bit or 64-bit depending on the target.
44121 The @samp{org.gnu.gdb.mips.dsp} feature is optional. It should
44122 contain registers @samp{hi1} through @samp{hi3}, @samp{lo1} through
44123 @samp{lo3}, and @samp{dspctl}. The @samp{dspctl} register should
44124 be 32-bit and the rest may be 32-bit or 64-bit depending on the target.
44126 The @samp{org.gnu.gdb.mips.linux} feature is optional. It should
44127 contain a single register, @samp{restart}, which is used by the
44128 Linux kernel to control restartable syscalls.
44130 @node M68K Features
44131 @subsection M68K Features
44132 @cindex target descriptions, M68K features
44135 @item @samp{org.gnu.gdb.m68k.core}
44136 @itemx @samp{org.gnu.gdb.coldfire.core}
44137 @itemx @samp{org.gnu.gdb.fido.core}
44138 One of those features must be always present.
44139 The feature that is present determines which flavor of m68k is
44140 used. The feature that is present should contain registers
44141 @samp{d0} through @samp{d7}, @samp{a0} through @samp{a5}, @samp{fp},
44142 @samp{sp}, @samp{ps} and @samp{pc}.
44144 @item @samp{org.gnu.gdb.coldfire.fp}
44145 This feature is optional. If present, it should contain registers
44146 @samp{fp0} through @samp{fp7}, @samp{fpcontrol}, @samp{fpstatus} and
44150 @node NDS32 Features
44151 @subsection NDS32 Features
44152 @cindex target descriptions, NDS32 features
44154 The @samp{org.gnu.gdb.nds32.core} feature is required for NDS32
44155 targets. It should contain at least registers @samp{r0} through
44156 @samp{r10}, @samp{r15}, @samp{fp}, @samp{gp}, @samp{lp}, @samp{sp},
44159 The @samp{org.gnu.gdb.nds32.fpu} feature is optional. If present,
44160 it should contain 64-bit double-precision floating-point registers
44161 @samp{fd0} through @emph{fdN}, which should be @samp{fd3}, @samp{fd7},
44162 @samp{fd15}, or @samp{fd31} based on the FPU configuration implemented.
44164 @emph{Note:} The first sixteen 64-bit double-precision floating-point
44165 registers are overlapped with the thirty-two 32-bit single-precision
44166 floating-point registers. The 32-bit single-precision registers, if
44167 not being listed explicitly, will be synthesized from halves of the
44168 overlapping 64-bit double-precision registers. Listing 32-bit
44169 single-precision registers explicitly is deprecated, and the
44170 support to it could be totally removed some day.
44172 @node Nios II Features
44173 @subsection Nios II Features
44174 @cindex target descriptions, Nios II features
44176 The @samp{org.gnu.gdb.nios2.cpu} feature is required for Nios II
44177 targets. It should contain the 32 core registers (@samp{zero},
44178 @samp{at}, @samp{r2} through @samp{r23}, @samp{et} through @samp{ra}),
44179 @samp{pc}, and the 16 control registers (@samp{status} through
44182 @node OpenRISC 1000 Features
44183 @subsection Openrisc 1000 Features
44184 @cindex target descriptions, OpenRISC 1000 features
44186 The @samp{org.gnu.gdb.or1k.group0} feature is required for OpenRISC 1000
44187 targets. It should contain the 32 general purpose registers (@samp{r0}
44188 through @samp{r31}), @samp{ppc}, @samp{npc} and @samp{sr}.
44190 @node PowerPC Features
44191 @subsection PowerPC Features
44192 @cindex target descriptions, PowerPC features
44194 The @samp{org.gnu.gdb.power.core} feature is required for PowerPC
44195 targets. It should contain registers @samp{r0} through @samp{r31},
44196 @samp{pc}, @samp{msr}, @samp{cr}, @samp{lr}, @samp{ctr}, and
44197 @samp{xer}. They may be 32-bit or 64-bit depending on the target.
44199 The @samp{org.gnu.gdb.power.fpu} feature is optional. It should
44200 contain registers @samp{f0} through @samp{f31} and @samp{fpscr}.
44202 The @samp{org.gnu.gdb.power.altivec} feature is optional. It should
44203 contain registers @samp{vr0} through @samp{vr31}, @samp{vscr}, and
44204 @samp{vrsave}. @value{GDBN} will define pseudo-registers @samp{v0}
44205 through @samp{v31} as aliases for the corresponding @samp{vrX}
44208 The @samp{org.gnu.gdb.power.vsx} feature is optional. It should
44209 contain registers @samp{vs0h} through @samp{vs31h}. @value{GDBN} will
44210 combine these registers with the floating point registers (@samp{f0}
44211 through @samp{f31}) and the altivec registers (@samp{vr0} through
44212 @samp{vr31}) to present the 128-bit wide registers @samp{vs0} through
44213 @samp{vs63}, the set of vector-scalar registers for POWER7.
44214 Therefore, this feature requires both @samp{org.gnu.gdb.power.fpu} and
44215 @samp{org.gnu.gdb.power.altivec}.
44217 The @samp{org.gnu.gdb.power.spe} feature is optional. It should
44218 contain registers @samp{ev0h} through @samp{ev31h}, @samp{acc}, and
44219 @samp{spefscr}. SPE targets should provide 32-bit registers in
44220 @samp{org.gnu.gdb.power.core} and provide the upper halves in
44221 @samp{ev0h} through @samp{ev31h}. @value{GDBN} will combine
44222 these to present registers @samp{ev0} through @samp{ev31} to the
44225 The @samp{org.gnu.gdb.power.ppr} feature is optional. It should
44226 contain the 64-bit register @samp{ppr}.
44228 The @samp{org.gnu.gdb.power.dscr} feature is optional. It should
44229 contain the 64-bit register @samp{dscr}.
44231 The @samp{org.gnu.gdb.power.tar} feature is optional. It should
44232 contain the 64-bit register @samp{tar}.
44234 The @samp{org.gnu.gdb.power.ebb} feature is optional. It should
44235 contain registers @samp{bescr}, @samp{ebbhr} and @samp{ebbrr}, all
44238 The @samp{org.gnu.gdb.power.linux.pmu} feature is optional. It should
44239 contain registers @samp{mmcr0}, @samp{mmcr2}, @samp{siar}, @samp{sdar}
44240 and @samp{sier}, all 64-bit wide. This is the subset of the isa 2.07
44241 server PMU registers provided by @sc{gnu}/Linux.
44243 The @samp{org.gnu.gdb.power.htm.spr} feature is optional. It should
44244 contain registers @samp{tfhar}, @samp{texasr} and @samp{tfiar}, all
44247 The @samp{org.gnu.gdb.power.htm.core} feature is optional. It should
44248 contain the checkpointed general-purpose registers @samp{cr0} through
44249 @samp{cr31}, as well as the checkpointed registers @samp{clr} and
44250 @samp{cctr}. These registers may all be either 32-bit or 64-bit
44251 depending on the target. It should also contain the checkpointed
44252 registers @samp{ccr} and @samp{cxer}, which should both be 32-bit
44255 The @samp{org.gnu.gdb.power.htm.fpu} feature is optional. It should
44256 contain the checkpointed 64-bit floating-point registers @samp{cf0}
44257 through @samp{cf31}, as well as the checkpointed 64-bit register
44260 The @samp{org.gnu.gdb.power.htm.altivec} feature is optional. It
44261 should contain the checkpointed altivec registers @samp{cvr0} through
44262 @samp{cvr31}, all 128-bit wide. It should also contain the
44263 checkpointed registers @samp{cvscr} and @samp{cvrsave}, both 32-bit
44266 The @samp{org.gnu.gdb.power.htm.vsx} feature is optional. It should
44267 contain registers @samp{cvs0h} through @samp{cvs31h}. @value{GDBN}
44268 will combine these registers with the checkpointed floating point
44269 registers (@samp{cf0} through @samp{cf31}) and the checkpointed
44270 altivec registers (@samp{cvr0} through @samp{cvr31}) to present the
44271 128-bit wide checkpointed vector-scalar registers @samp{cvs0} through
44272 @samp{cvs63}. Therefore, this feature requires both
44273 @samp{org.gnu.gdb.power.htm.altivec} and
44274 @samp{org.gnu.gdb.power.htm.fpu}.
44276 The @samp{org.gnu.gdb.power.htm.ppr} feature is optional. It should
44277 contain the 64-bit checkpointed register @samp{cppr}.
44279 The @samp{org.gnu.gdb.power.htm.dscr} feature is optional. It should
44280 contain the 64-bit checkpointed register @samp{cdscr}.
44282 The @samp{org.gnu.gdb.power.htm.tar} feature is optional. It should
44283 contain the 64-bit checkpointed register @samp{ctar}.
44286 @node RISC-V Features
44287 @subsection RISC-V Features
44288 @cindex target descriptions, RISC-V Features
44290 The @samp{org.gnu.gdb.riscv.cpu} feature is required for RISC-V
44291 targets. It should contain the registers @samp{x0} through
44292 @samp{x31}, and @samp{pc}. Either the architectural names (@samp{x0},
44293 @samp{x1}, etc) can be used, or the ABI names (@samp{zero}, @samp{ra},
44296 The @samp{org.gnu.gdb.riscv.fpu} feature is optional. If present, it
44297 should contain registers @samp{f0} through @samp{f31}, @samp{fflags},
44298 @samp{frm}, and @samp{fcsr}. As with the cpu feature, either the
44299 architectural register names, or the ABI names can be used.
44301 The @samp{org.gnu.gdb.riscv.virtual} feature is optional. If present,
44302 it should contain registers that are not backed by real registers on
44303 the target, but are instead virtual, where the register value is
44304 derived from other target state. In many ways these are like
44305 @value{GDBN}s pseudo-registers, except implemented by the target.
44306 Currently the only register expected in this set is the one byte
44307 @samp{priv} register that contains the target's privilege level in the
44308 least significant two bits.
44310 The @samp{org.gnu.gdb.riscv.csr} feature is optional. If present, it
44311 should contain all of the target's standard CSRs. Standard CSRs are
44312 those defined in the RISC-V specification documents. There is some
44313 overlap between this feature and the fpu feature; the @samp{fflags},
44314 @samp{frm}, and @samp{fcsr} registers could be in either feature. The
44315 expectation is that these registers will be in the fpu feature if the
44316 target has floating point hardware, but can be moved into the csr
44317 feature if the target has the floating point control registers, but no
44318 other floating point hardware.
44320 @node S/390 and System z Features
44321 @subsection S/390 and System z Features
44322 @cindex target descriptions, S/390 features
44323 @cindex target descriptions, System z features
44325 The @samp{org.gnu.gdb.s390.core} feature is required for S/390 and
44326 System z targets. It should contain the PSW and the 16 general
44327 registers. In particular, System z targets should provide the 64-bit
44328 registers @samp{pswm}, @samp{pswa}, and @samp{r0} through @samp{r15}.
44329 S/390 targets should provide the 32-bit versions of these registers.
44330 A System z target that runs in 31-bit addressing mode should provide
44331 32-bit versions of @samp{pswm} and @samp{pswa}, as well as the general
44332 register's upper halves @samp{r0h} through @samp{r15h}, and their
44333 lower halves @samp{r0l} through @samp{r15l}.
44335 The @samp{org.gnu.gdb.s390.fpr} feature is required. It should
44336 contain the 64-bit registers @samp{f0} through @samp{f15}, and
44339 The @samp{org.gnu.gdb.s390.acr} feature is required. It should
44340 contain the 32-bit registers @samp{acr0} through @samp{acr15}.
44342 The @samp{org.gnu.gdb.s390.linux} feature is optional. It should
44343 contain the register @samp{orig_r2}, which is 64-bit wide on System z
44344 targets and 32-bit otherwise. In addition, the feature may contain
44345 the @samp{last_break} register, whose width depends on the addressing
44346 mode, as well as the @samp{system_call} register, which is always
44349 The @samp{org.gnu.gdb.s390.tdb} feature is optional. It should
44350 contain the 64-bit registers @samp{tdb0}, @samp{tac}, @samp{tct},
44351 @samp{atia}, and @samp{tr0} through @samp{tr15}.
44353 The @samp{org.gnu.gdb.s390.vx} feature is optional. It should contain
44354 64-bit wide registers @samp{v0l} through @samp{v15l}, which will be
44355 combined by @value{GDBN} with the floating point registers @samp{f0}
44356 through @samp{f15} to present the 128-bit wide vector registers
44357 @samp{v0} through @samp{v15}. In addition, this feature should
44358 contain the 128-bit wide vector registers @samp{v16} through
44361 The @samp{org.gnu.gdb.s390.gs} feature is optional. It should contain
44362 the 64-bit wide guarded-storage-control registers @samp{gsd},
44363 @samp{gssm}, and @samp{gsepla}.
44365 The @samp{org.gnu.gdb.s390.gsbc} feature is optional. It should contain
44366 the 64-bit wide guarded-storage broadcast control registers
44367 @samp{bc_gsd}, @samp{bc_gssm}, and @samp{bc_gsepla}.
44369 @node Sparc Features
44370 @subsection Sparc Features
44371 @cindex target descriptions, sparc32 features
44372 @cindex target descriptions, sparc64 features
44373 The @samp{org.gnu.gdb.sparc.cpu} feature is required for sparc32/sparc64
44374 targets. It should describe the following registers:
44378 @samp{g0} through @samp{g7}
44380 @samp{o0} through @samp{o7}
44382 @samp{l0} through @samp{l7}
44384 @samp{i0} through @samp{i7}
44387 They may be 32-bit or 64-bit depending on the target.
44389 Also the @samp{org.gnu.gdb.sparc.fpu} feature is required for sparc32/sparc64
44390 targets. It should describe the following registers:
44394 @samp{f0} through @samp{f31}
44396 @samp{f32} through @samp{f62} for sparc64
44399 The @samp{org.gnu.gdb.sparc.cp0} feature is required for sparc32/sparc64
44400 targets. It should describe the following registers:
44404 @samp{y}, @samp{psr}, @samp{wim}, @samp{tbr}, @samp{pc}, @samp{npc},
44405 @samp{fsr}, and @samp{csr} for sparc32
44407 @samp{pc}, @samp{npc}, @samp{state}, @samp{fsr}, @samp{fprs}, and @samp{y}
44411 @node TIC6x Features
44412 @subsection TMS320C6x Features
44413 @cindex target descriptions, TIC6x features
44414 @cindex target descriptions, TMS320C6x features
44415 The @samp{org.gnu.gdb.tic6x.core} feature is required for TMS320C6x
44416 targets. It should contain registers @samp{A0} through @samp{A15},
44417 registers @samp{B0} through @samp{B15}, @samp{CSR} and @samp{PC}.
44419 The @samp{org.gnu.gdb.tic6x.gp} feature is optional. It should
44420 contain registers @samp{A16} through @samp{A31} and @samp{B16}
44421 through @samp{B31}.
44423 The @samp{org.gnu.gdb.tic6x.c6xp} feature is optional. It should
44424 contain registers @samp{TSR}, @samp{ILC} and @samp{RILC}.
44426 @node Operating System Information
44427 @appendix Operating System Information
44428 @cindex operating system information
44434 Users of @value{GDBN} often wish to obtain information about the state of
44435 the operating system running on the target---for example the list of
44436 processes, or the list of open files. This section describes the
44437 mechanism that makes it possible. This mechanism is similar to the
44438 target features mechanism (@pxref{Target Descriptions}), but focuses
44439 on a different aspect of target.
44441 Operating system information is retrived from the target via the
44442 remote protocol, using @samp{qXfer} requests (@pxref{qXfer osdata
44443 read}). The object name in the request should be @samp{osdata}, and
44444 the @var{annex} identifies the data to be fetched.
44447 @appendixsection Process list
44448 @cindex operating system information, process list
44450 When requesting the process list, the @var{annex} field in the
44451 @samp{qXfer} request should be @samp{processes}. The returned data is
44452 an XML document. The formal syntax of this document is defined in
44453 @file{gdb/features/osdata.dtd}.
44455 An example document is:
44458 <?xml version="1.0"?>
44459 <!DOCTYPE target SYSTEM "osdata.dtd">
44460 <osdata type="processes">
44462 <column name="pid">1</column>
44463 <column name="user">root</column>
44464 <column name="command">/sbin/init</column>
44465 <column name="cores">1,2,3</column>
44470 Each item should include a column whose name is @samp{pid}. The value
44471 of that column should identify the process on the target. The
44472 @samp{user} and @samp{command} columns are optional, and will be
44473 displayed by @value{GDBN}. The @samp{cores} column, if present,
44474 should contain a comma-separated list of cores that this process
44475 is running on. Target may provide additional columns,
44476 which @value{GDBN} currently ignores.
44478 @node Trace File Format
44479 @appendix Trace File Format
44480 @cindex trace file format
44482 The trace file comes in three parts: a header, a textual description
44483 section, and a trace frame section with binary data.
44485 The header has the form @code{\x7fTRACE0\n}. The first byte is
44486 @code{0x7f} so as to indicate that the file contains binary data,
44487 while the @code{0} is a version number that may have different values
44490 The description section consists of multiple lines of @sc{ascii} text
44491 separated by newline characters (@code{0xa}). The lines may include a
44492 variety of optional descriptive or context-setting information, such
44493 as tracepoint definitions or register set size. @value{GDBN} will
44494 ignore any line that it does not recognize. An empty line marks the end
44499 Specifies the size of a register block in bytes. This is equal to the
44500 size of a @code{g} packet payload in the remote protocol. @var{size}
44501 is an ascii decimal number. There should be only one such line in
44502 a single trace file.
44504 @item status @var{status}
44505 Trace status. @var{status} has the same format as a @code{qTStatus}
44506 remote packet reply. There should be only one such line in a single trace
44509 @item tp @var{payload}
44510 Tracepoint definition. The @var{payload} has the same format as
44511 @code{qTfP}/@code{qTsP} remote packet reply payload. A single tracepoint
44512 may take multiple lines of definition, corresponding to the multiple
44515 @item tsv @var{payload}
44516 Trace state variable definition. The @var{payload} has the same format as
44517 @code{qTfV}/@code{qTsV} remote packet reply payload. A single variable
44518 may take multiple lines of definition, corresponding to the multiple
44521 @item tdesc @var{payload}
44522 Target description in XML format. The @var{payload} is a single line of
44523 the XML file. All such lines should be concatenated together to get
44524 the original XML file. This file is in the same format as @code{qXfer}
44525 @code{features} payload, and corresponds to the main @code{target.xml}
44526 file. Includes are not allowed.
44530 The trace frame section consists of a number of consecutive frames.
44531 Each frame begins with a two-byte tracepoint number, followed by a
44532 four-byte size giving the amount of data in the frame. The data in
44533 the frame consists of a number of blocks, each introduced by a
44534 character indicating its type (at least register, memory, and trace
44535 state variable). The data in this section is raw binary, not a
44536 hexadecimal or other encoding; its endianness matches the target's
44539 @c FIXME bi-arch may require endianness/arch info in description section
44542 @item R @var{bytes}
44543 Register block. The number and ordering of bytes matches that of a
44544 @code{g} packet in the remote protocol. Note that these are the
44545 actual bytes, in target order, not a hexadecimal encoding.
44547 @item M @var{address} @var{length} @var{bytes}...
44548 Memory block. This is a contiguous block of memory, at the 8-byte
44549 address @var{address}, with a 2-byte length @var{length}, followed by
44550 @var{length} bytes.
44552 @item V @var{number} @var{value}
44553 Trace state variable block. This records the 8-byte signed value
44554 @var{value} of trace state variable numbered @var{number}.
44558 Future enhancements of the trace file format may include additional types
44561 @node Index Section Format
44562 @appendix @code{.gdb_index} section format
44563 @cindex .gdb_index section format
44564 @cindex index section format
44566 This section documents the index section that is created by @code{save
44567 gdb-index} (@pxref{Index Files}). The index section is
44568 DWARF-specific; some knowledge of DWARF is assumed in this
44571 The mapped index file format is designed to be directly
44572 @code{mmap}able on any architecture. In most cases, a datum is
44573 represented using a little-endian 32-bit integer value, called an
44574 @code{offset_type}. Big endian machines must byte-swap the values
44575 before using them. Exceptions to this rule are noted. The data is
44576 laid out such that alignment is always respected.
44578 A mapped index consists of several areas, laid out in order.
44582 The file header. This is a sequence of values, of @code{offset_type}
44583 unless otherwise noted:
44587 The version number, currently 8. Versions 1, 2 and 3 are obsolete.
44588 Version 4 uses a different hashing function from versions 5 and 6.
44589 Version 6 includes symbols for inlined functions, whereas versions 4
44590 and 5 do not. Version 7 adds attributes to the CU indices in the
44591 symbol table. Version 8 specifies that symbols from DWARF type units
44592 (@samp{DW_TAG_type_unit}) refer to the type unit's symbol table and not the
44593 compilation unit (@samp{DW_TAG_comp_unit}) using the type.
44595 @value{GDBN} will only read version 4, 5, or 6 indices
44596 by specifying @code{set use-deprecated-index-sections on}.
44597 GDB has a workaround for potentially broken version 7 indices so it is
44598 currently not flagged as deprecated.
44601 The offset, from the start of the file, of the CU list.
44604 The offset, from the start of the file, of the types CU list. Note
44605 that this area can be empty, in which case this offset will be equal
44606 to the next offset.
44609 The offset, from the start of the file, of the address area.
44612 The offset, from the start of the file, of the symbol table.
44615 The offset, from the start of the file, of the constant pool.
44619 The CU list. This is a sequence of pairs of 64-bit little-endian
44620 values, sorted by the CU offset. The first element in each pair is
44621 the offset of a CU in the @code{.debug_info} section. The second
44622 element in each pair is the length of that CU. References to a CU
44623 elsewhere in the map are done using a CU index, which is just the
44624 0-based index into this table. Note that if there are type CUs, then
44625 conceptually CUs and type CUs form a single list for the purposes of
44629 The types CU list. This is a sequence of triplets of 64-bit
44630 little-endian values. In a triplet, the first value is the CU offset,
44631 the second value is the type offset in the CU, and the third value is
44632 the type signature. The types CU list is not sorted.
44635 The address area. The address area consists of a sequence of address
44636 entries. Each address entry has three elements:
44640 The low address. This is a 64-bit little-endian value.
44643 The high address. This is a 64-bit little-endian value. Like
44644 @code{DW_AT_high_pc}, the value is one byte beyond the end.
44647 The CU index. This is an @code{offset_type} value.
44651 The symbol table. This is an open-addressed hash table. The size of
44652 the hash table is always a power of 2.
44654 Each slot in the hash table consists of a pair of @code{offset_type}
44655 values. The first value is the offset of the symbol's name in the
44656 constant pool. The second value is the offset of the CU vector in the
44659 If both values are 0, then this slot in the hash table is empty. This
44660 is ok because while 0 is a valid constant pool index, it cannot be a
44661 valid index for both a string and a CU vector.
44663 The hash value for a table entry is computed by applying an
44664 iterative hash function to the symbol's name. Starting with an
44665 initial value of @code{r = 0}, each (unsigned) character @samp{c} in
44666 the string is incorporated into the hash using the formula depending on the
44671 The formula is @code{r = r * 67 + c - 113}.
44673 @item Versions 5 to 7
44674 The formula is @code{r = r * 67 + tolower (c) - 113}.
44677 The terminating @samp{\0} is not incorporated into the hash.
44679 The step size used in the hash table is computed via
44680 @code{((hash * 17) & (size - 1)) | 1}, where @samp{hash} is the hash
44681 value, and @samp{size} is the size of the hash table. The step size
44682 is used to find the next candidate slot when handling a hash
44685 The names of C@t{++} symbols in the hash table are canonicalized. We
44686 don't currently have a simple description of the canonicalization
44687 algorithm; if you intend to create new index sections, you must read
44691 The constant pool. This is simply a bunch of bytes. It is organized
44692 so that alignment is correct: CU vectors are stored first, followed by
44695 A CU vector in the constant pool is a sequence of @code{offset_type}
44696 values. The first value is the number of CU indices in the vector.
44697 Each subsequent value is the index and symbol attributes of a CU in
44698 the CU list. This element in the hash table is used to indicate which
44699 CUs define the symbol and how the symbol is used.
44700 See below for the format of each CU index+attributes entry.
44702 A string in the constant pool is zero-terminated.
44705 Attributes were added to CU index values in @code{.gdb_index} version 7.
44706 If a symbol has multiple uses within a CU then there is one
44707 CU index+attributes value for each use.
44709 The format of each CU index+attributes entry is as follows
44715 This is the index of the CU in the CU list.
44717 These bits are reserved for future purposes and must be zero.
44719 The kind of the symbol in the CU.
44723 This value is reserved and should not be used.
44724 By reserving zero the full @code{offset_type} value is backwards compatible
44725 with previous versions of the index.
44727 The symbol is a type.
44729 The symbol is a variable or an enum value.
44731 The symbol is a function.
44733 Any other kind of symbol.
44735 These values are reserved.
44739 This bit is zero if the value is global and one if it is static.
44741 The determination of whether a symbol is global or static is complicated.
44742 The authorative reference is the file @file{dwarf2read.c} in
44743 @value{GDBN} sources.
44747 This pseudo-code describes the computation of a symbol's kind and
44748 global/static attributes in the index.
44751 is_external = get_attribute (die, DW_AT_external);
44752 language = get_attribute (cu_die, DW_AT_language);
44755 case DW_TAG_typedef:
44756 case DW_TAG_base_type:
44757 case DW_TAG_subrange_type:
44761 case DW_TAG_enumerator:
44763 is_static = language != CPLUS;
44765 case DW_TAG_subprogram:
44767 is_static = ! (is_external || language == ADA);
44769 case DW_TAG_constant:
44771 is_static = ! is_external;
44773 case DW_TAG_variable:
44775 is_static = ! is_external;
44777 case DW_TAG_namespace:
44781 case DW_TAG_class_type:
44782 case DW_TAG_interface_type:
44783 case DW_TAG_structure_type:
44784 case DW_TAG_union_type:
44785 case DW_TAG_enumeration_type:
44787 is_static = language != CPLUS;
44795 @appendix Manual pages
44799 * gdb man:: The GNU Debugger man page
44800 * gdbserver man:: Remote Server for the GNU Debugger man page
44801 * gcore man:: Generate a core file of a running program
44802 * gdbinit man:: gdbinit scripts
44803 * gdb-add-index man:: Add index files to speed up GDB
44809 @c man title gdb The GNU Debugger
44811 @c man begin SYNOPSIS gdb
44812 gdb [@option{-help}] [@option{-nh}] [@option{-nx}] [@option{-q}]
44813 [@option{-batch}] [@option{-cd=}@var{dir}] [@option{-f}]
44814 [@option{-b}@w{ }@var{bps}]
44815 [@option{-tty=}@var{dev}] [@option{-s} @var{symfile}]
44816 [@option{-e}@w{ }@var{prog}] [@option{-se}@w{ }@var{prog}]
44817 [@option{-c}@w{ }@var{core}] [@option{-p}@w{ }@var{procID}]
44818 [@option{-x}@w{ }@var{cmds}] [@option{-d}@w{ }@var{dir}]
44819 [@var{prog}|@var{prog} @var{procID}|@var{prog} @var{core}]
44822 @c man begin DESCRIPTION gdb
44823 The purpose of a debugger such as @value{GDBN} is to allow you to see what is
44824 going on ``inside'' another program while it executes -- or what another
44825 program was doing at the moment it crashed.
44827 @value{GDBN} can do four main kinds of things (plus other things in support of
44828 these) to help you catch bugs in the act:
44832 Start your program, specifying anything that might affect its behavior.
44835 Make your program stop on specified conditions.
44838 Examine what has happened, when your program has stopped.
44841 Change things in your program, so you can experiment with correcting the
44842 effects of one bug and go on to learn about another.
44845 You can use @value{GDBN} to debug programs written in C, C@t{++}, Fortran and
44848 @value{GDBN} is invoked with the shell command @code{gdb}. Once started, it reads
44849 commands from the terminal until you tell it to exit with the @value{GDBN}
44850 command @code{quit}. You can get online help from @value{GDBN} itself
44851 by using the command @code{help}.
44853 You can run @code{gdb} with no arguments or options; but the most
44854 usual way to start @value{GDBN} is with one argument or two, specifying an
44855 executable program as the argument:
44861 You can also start with both an executable program and a core file specified:
44867 You can, instead, specify a process ID as a second argument or use option
44868 @code{-p}, if you want to debug a running process:
44876 would attach @value{GDBN} to process @code{1234}. With option @option{-p} you
44877 can omit the @var{program} filename.
44879 Here are some of the most frequently needed @value{GDBN} commands:
44881 @c pod2man highlights the right hand side of the @item lines.
44883 @item break [@var{file}:]@var{function}
44884 Set a breakpoint at @var{function} (in @var{file}).
44886 @item run [@var{arglist}]
44887 Start your program (with @var{arglist}, if specified).
44890 Backtrace: display the program stack.
44892 @item print @var{expr}
44893 Display the value of an expression.
44896 Continue running your program (after stopping, e.g. at a breakpoint).
44899 Execute next program line (after stopping); step @emph{over} any
44900 function calls in the line.
44902 @item edit [@var{file}:]@var{function}
44903 look at the program line where it is presently stopped.
44905 @item list [@var{file}:]@var{function}
44906 type the text of the program in the vicinity of where it is presently stopped.
44909 Execute next program line (after stopping); step @emph{into} any
44910 function calls in the line.
44912 @item help [@var{name}]
44913 Show information about @value{GDBN} command @var{name}, or general information
44914 about using @value{GDBN}.
44917 Exit from @value{GDBN}.
44921 For full details on @value{GDBN},
44922 see @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
44923 by Richard M. Stallman and Roland H. Pesch. The same text is available online
44924 as the @code{gdb} entry in the @code{info} program.
44928 @c man begin OPTIONS gdb
44929 Any arguments other than options specify an executable
44930 file and core file (or process ID); that is, the first argument
44931 encountered with no
44932 associated option flag is equivalent to a @option{-se} option, and the second,
44933 if any, is equivalent to a @option{-c} option if it's the name of a file.
44935 both long and short forms; both are shown here. The long forms are also
44936 recognized if you truncate them, so long as enough of the option is
44937 present to be unambiguous. (If you prefer, you can flag option
44938 arguments with @option{+} rather than @option{-}, though we illustrate the
44939 more usual convention.)
44941 All the options and command line arguments you give are processed
44942 in sequential order. The order makes a difference when the @option{-x}
44948 List all options, with brief explanations.
44950 @item -symbols=@var{file}
44951 @itemx -s @var{file}
44952 Read symbol table from file @var{file}.
44955 Enable writing into executable and core files.
44957 @item -exec=@var{file}
44958 @itemx -e @var{file}
44959 Use file @var{file} as the executable file to execute when
44960 appropriate, and for examining pure data in conjunction with a core
44963 @item -se=@var{file}
44964 Read symbol table from file @var{file} and use it as the executable
44967 @item -core=@var{file}
44968 @itemx -c @var{file}
44969 Use file @var{file} as a core dump to examine.
44971 @item -command=@var{file}
44972 @itemx -x @var{file}
44973 Execute @value{GDBN} commands from file @var{file}.
44975 @item -ex @var{command}
44976 Execute given @value{GDBN} @var{command}.
44978 @item -directory=@var{directory}
44979 @itemx -d @var{directory}
44980 Add @var{directory} to the path to search for source files.
44983 Do not execute commands from @file{~/.gdbinit}.
44987 Do not execute commands from any @file{.gdbinit} initialization files.
44991 ``Quiet''. Do not print the introductory and copyright messages. These
44992 messages are also suppressed in batch mode.
44995 Run in batch mode. Exit with status @code{0} after processing all the command
44996 files specified with @option{-x} (and @file{.gdbinit}, if not inhibited).
44997 Exit with nonzero status if an error occurs in executing the @value{GDBN}
44998 commands in the command files.
45000 Batch mode may be useful for running @value{GDBN} as a filter, for example to
45001 download and run a program on another computer; in order to make this
45002 more useful, the message
45005 Program exited normally.
45009 (which is ordinarily issued whenever a program running under @value{GDBN} control
45010 terminates) is not issued when running in batch mode.
45012 @item -cd=@var{directory}
45013 Run @value{GDBN} using @var{directory} as its working directory,
45014 instead of the current directory.
45018 Emacs sets this option when it runs @value{GDBN} as a subprocess. It tells
45019 @value{GDBN} to output the full file name and line number in a standard,
45020 recognizable fashion each time a stack frame is displayed (which
45021 includes each time the program stops). This recognizable format looks
45022 like two @samp{\032} characters, followed by the file name, line number
45023 and character position separated by colons, and a newline. The
45024 Emacs-to-@value{GDBN} interface program uses the two @samp{\032}
45025 characters as a signal to display the source code for the frame.
45028 Set the line speed (baud rate or bits per second) of any serial
45029 interface used by @value{GDBN} for remote debugging.
45031 @item -tty=@var{device}
45032 Run using @var{device} for your program's standard input and output.
45036 @c man begin SEEALSO gdb
45038 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45039 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45040 documentation are properly installed at your site, the command
45047 should give you access to the complete manual.
45049 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45050 Richard M. Stallman and Roland H. Pesch, July 1991.
45054 @node gdbserver man
45055 @heading gdbserver man
45057 @c man title gdbserver Remote Server for the GNU Debugger
45059 @c man begin SYNOPSIS gdbserver
45060 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
45062 gdbserver --attach @var{comm} @var{pid}
45064 gdbserver --multi @var{comm}
45068 @c man begin DESCRIPTION gdbserver
45069 @command{gdbserver} is a program that allows you to run @value{GDBN} on a different machine
45070 than the one which is running the program being debugged.
45073 @subheading Usage (server (target) side)
45076 Usage (server (target) side):
45079 First, you need to have a copy of the program you want to debug put onto
45080 the target system. The program can be stripped to save space if needed, as
45081 @command{gdbserver} doesn't care about symbols. All symbol handling is taken care of by
45082 the @value{GDBN} running on the host system.
45084 To use the server, you log on to the target system, and run the @command{gdbserver}
45085 program. You must tell it (a) how to communicate with @value{GDBN}, (b) the name of
45086 your program, and (c) its arguments. The general syntax is:
45089 target> gdbserver @var{comm} @var{program} [@var{args} ...]
45092 For example, using a serial port, you might say:
45096 @c @file would wrap it as F</dev/com1>.
45097 target> gdbserver /dev/com1 emacs foo.txt
45100 target> gdbserver @file{/dev/com1} emacs foo.txt
45104 This tells @command{gdbserver} to debug emacs with an argument of foo.txt, and
45105 to communicate with @value{GDBN} via @file{/dev/com1}. @command{gdbserver} now
45106 waits patiently for the host @value{GDBN} to communicate with it.
45108 To use a TCP connection, you could say:
45111 target> gdbserver host:2345 emacs foo.txt
45114 This says pretty much the same thing as the last example, except that we are
45115 going to communicate with the @code{host} @value{GDBN} via TCP. The @code{host:2345} argument means
45116 that we are expecting to see a TCP connection from @code{host} to local TCP port
45117 2345. (Currently, the @code{host} part is ignored.) You can choose any number you
45118 want for the port number as long as it does not conflict with any existing TCP
45119 ports on the target system. This same port number must be used in the host
45120 @value{GDBN}s @code{target remote} command, which will be described shortly. Note that if
45121 you chose a port number that conflicts with another service, @command{gdbserver} will
45122 print an error message and exit.
45124 @command{gdbserver} can also attach to running programs.
45125 This is accomplished via the @option{--attach} argument. The syntax is:
45128 target> gdbserver --attach @var{comm} @var{pid}
45131 @var{pid} is the process ID of a currently running process. It isn't
45132 necessary to point @command{gdbserver} at a binary for the running process.
45134 To start @code{gdbserver} without supplying an initial command to run
45135 or process ID to attach, use the @option{--multi} command line option.
45136 In such case you should connect using @kbd{target extended-remote} to start
45137 the program you want to debug.
45140 target> gdbserver --multi @var{comm}
45144 @subheading Usage (host side)
45150 You need an unstripped copy of the target program on your host system, since
45151 @value{GDBN} needs to examine its symbol tables and such. Start up @value{GDBN} as you normally
45152 would, with the target program as the first argument. (You may need to use the
45153 @option{--baud} option if the serial line is running at anything except 9600 baud.)
45154 That is @code{gdb TARGET-PROG}, or @code{gdb --baud BAUD TARGET-PROG}. After that, the only
45155 new command you need to know about is @code{target remote}
45156 (or @code{target extended-remote}). Its argument is either
45157 a device name (usually a serial device, like @file{/dev/ttyb}), or a @code{HOST:PORT}
45158 descriptor. For example:
45162 @c @file would wrap it as F</dev/ttyb>.
45163 (gdb) target remote /dev/ttyb
45166 (gdb) target remote @file{/dev/ttyb}
45171 communicates with the server via serial line @file{/dev/ttyb}, and:
45174 (gdb) target remote the-target:2345
45178 communicates via a TCP connection to port 2345 on host `the-target', where
45179 you previously started up @command{gdbserver} with the same port number. Note that for
45180 TCP connections, you must start up @command{gdbserver} prior to using the `target remote'
45181 command, otherwise you may get an error that looks something like
45182 `Connection refused'.
45184 @command{gdbserver} can also debug multiple inferiors at once,
45187 the @value{GDBN} manual in node @code{Inferiors and Programs}
45188 -- shell command @code{info -f gdb -n 'Inferiors and Programs'}.
45191 @ref{Inferiors and Programs}.
45193 In such case use the @code{extended-remote} @value{GDBN} command variant:
45196 (gdb) target extended-remote the-target:2345
45199 The @command{gdbserver} option @option{--multi} may or may not be used in such
45203 @c man begin OPTIONS gdbserver
45204 There are three different modes for invoking @command{gdbserver}:
45209 Debug a specific program specified by its program name:
45212 gdbserver @var{comm} @var{prog} [@var{args}@dots{}]
45215 The @var{comm} parameter specifies how should the server communicate
45216 with @value{GDBN}; it is either a device name (to use a serial line),
45217 a TCP port number (@code{:1234}), or @code{-} or @code{stdio} to use
45218 stdin/stdout of @code{gdbserver}. Specify the name of the program to
45219 debug in @var{prog}. Any remaining arguments will be passed to the
45220 program verbatim. When the program exits, @value{GDBN} will close the
45221 connection, and @code{gdbserver} will exit.
45224 Debug a specific program by specifying the process ID of a running
45228 gdbserver --attach @var{comm} @var{pid}
45231 The @var{comm} parameter is as described above. Supply the process ID
45232 of a running program in @var{pid}; @value{GDBN} will do everything
45233 else. Like with the previous mode, when the process @var{pid} exits,
45234 @value{GDBN} will close the connection, and @code{gdbserver} will exit.
45237 Multi-process mode -- debug more than one program/process:
45240 gdbserver --multi @var{comm}
45243 In this mode, @value{GDBN} can instruct @command{gdbserver} which
45244 command(s) to run. Unlike the other 2 modes, @value{GDBN} will not
45245 close the connection when a process being debugged exits, so you can
45246 debug several processes in the same session.
45249 In each of the modes you may specify these options:
45254 List all options, with brief explanations.
45257 This option causes @command{gdbserver} to print its version number and exit.
45260 @command{gdbserver} will attach to a running program. The syntax is:
45263 target> gdbserver --attach @var{comm} @var{pid}
45266 @var{pid} is the process ID of a currently running process. It isn't
45267 necessary to point @command{gdbserver} at a binary for the running process.
45270 To start @code{gdbserver} without supplying an initial command to run
45271 or process ID to attach, use this command line option.
45272 Then you can connect using @kbd{target extended-remote} and start
45273 the program you want to debug. The syntax is:
45276 target> gdbserver --multi @var{comm}
45280 Instruct @code{gdbserver} to display extra status information about the debugging
45282 This option is intended for @code{gdbserver} development and for bug reports to
45285 @item --remote-debug
45286 Instruct @code{gdbserver} to display remote protocol debug output.
45287 This option is intended for @code{gdbserver} development and for bug reports to
45290 @item --debug-file=@var{filename}
45291 Instruct @code{gdbserver} to send any debug output to the given @var{filename}.
45292 This option is intended for @code{gdbserver} development and for bug reports to
45295 @item --debug-format=option1@r{[},option2,...@r{]}
45296 Instruct @code{gdbserver} to include extra information in each line
45297 of debugging output.
45298 @xref{Other Command-Line Arguments for gdbserver}.
45301 Specify a wrapper to launch programs
45302 for debugging. The option should be followed by the name of the
45303 wrapper, then any command-line arguments to pass to the wrapper, then
45304 @kbd{--} indicating the end of the wrapper arguments.
45307 By default, @command{gdbserver} keeps the listening TCP port open, so that
45308 additional connections are possible. However, if you start @code{gdbserver}
45309 with the @option{--once} option, it will stop listening for any further
45310 connection attempts after connecting to the first @value{GDBN} session.
45312 @c --disable-packet is not documented for users.
45314 @c --disable-randomization and --no-disable-randomization are superseded by
45315 @c QDisableRandomization.
45320 @c man begin SEEALSO gdbserver
45322 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45323 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45324 documentation are properly installed at your site, the command
45330 should give you access to the complete manual.
45332 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45333 Richard M. Stallman and Roland H. Pesch, July 1991.
45340 @c man title gcore Generate a core file of a running program
45343 @c man begin SYNOPSIS gcore
45344 gcore [-a] [-o @var{prefix}] @var{pid1} [@var{pid2}...@var{pidN}]
45348 @c man begin DESCRIPTION gcore
45349 Generate core dumps of one or more running programs with process IDs
45350 @var{pid1}, @var{pid2}, etc. A core file produced by @command{gcore}
45351 is equivalent to one produced by the kernel when the process crashes
45352 (and when @kbd{ulimit -c} was used to set up an appropriate core dump
45353 limit). However, unlike after a crash, after @command{gcore} finishes
45354 its job the program remains running without any change.
45357 @c man begin OPTIONS gcore
45360 Dump all memory mappings. The actual effect of this option depends on
45361 the Operating System. On @sc{gnu}/Linux, it will disable
45362 @code{use-coredump-filter} (@pxref{set use-coredump-filter}) and
45363 enable @code{dump-excluded-mappings} (@pxref{set
45364 dump-excluded-mappings}).
45366 @item -o @var{prefix}
45367 The optional argument @var{prefix} specifies the prefix to be used
45368 when composing the file names of the core dumps. The file name is
45369 composed as @file{@var{prefix}.@var{pid}}, where @var{pid} is the
45370 process ID of the running program being analyzed by @command{gcore}.
45371 If not specified, @var{prefix} defaults to @var{gcore}.
45375 @c man begin SEEALSO gcore
45377 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45378 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45379 documentation are properly installed at your site, the command
45386 should give you access to the complete manual.
45388 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45389 Richard M. Stallman and Roland H. Pesch, July 1991.
45396 @c man title gdbinit GDB initialization scripts
45399 @c man begin SYNOPSIS gdbinit
45400 @ifset SYSTEM_GDBINIT
45401 @value{SYSTEM_GDBINIT}
45410 @c man begin DESCRIPTION gdbinit
45411 These files contain @value{GDBN} commands to automatically execute during
45412 @value{GDBN} startup. The lines of contents are canned sequences of commands,
45415 the @value{GDBN} manual in node @code{Sequences}
45416 -- shell command @code{info -f gdb -n Sequences}.
45422 Please read more in
45424 the @value{GDBN} manual in node @code{Startup}
45425 -- shell command @code{info -f gdb -n Startup}.
45432 @ifset SYSTEM_GDBINIT
45433 @item @value{SYSTEM_GDBINIT}
45435 @ifclear SYSTEM_GDBINIT
45436 @item (not enabled with @code{--with-system-gdbinit} during compilation)
45438 System-wide initialization file. It is executed unless user specified
45439 @value{GDBN} option @code{-nx} or @code{-n}.
45442 the @value{GDBN} manual in node @code{System-wide configuration}
45443 -- shell command @code{info -f gdb -n 'System-wide configuration'}.
45446 @ref{System-wide configuration}.
45450 User initialization file. It is executed unless user specified
45451 @value{GDBN} options @code{-nx}, @code{-n} or @code{-nh}.
45454 Initialization file for current directory. It may need to be enabled with
45455 @value{GDBN} security command @code{set auto-load local-gdbinit}.
45458 the @value{GDBN} manual in node @code{Init File in the Current Directory}
45459 -- shell command @code{info -f gdb -n 'Init File in the Current Directory'}.
45462 @ref{Init File in the Current Directory}.
45467 @c man begin SEEALSO gdbinit
45469 gdb(1), @code{info -f gdb -n Startup}
45471 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45472 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45473 documentation are properly installed at your site, the command
45479 should give you access to the complete manual.
45481 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45482 Richard M. Stallman and Roland H. Pesch, July 1991.
45486 @node gdb-add-index man
45487 @heading gdb-add-index
45488 @pindex gdb-add-index
45489 @anchor{gdb-add-index}
45491 @c man title gdb-add-index Add index files to speed up GDB
45493 @c man begin SYNOPSIS gdb-add-index
45494 gdb-add-index @var{filename}
45497 @c man begin DESCRIPTION gdb-add-index
45498 When @value{GDBN} finds a symbol file, it scans the symbols in the
45499 file in order to construct an internal symbol table. This lets most
45500 @value{GDBN} operations work quickly--at the cost of a delay early on.
45501 For large programs, this delay can be quite lengthy, so @value{GDBN}
45502 provides a way to build an index, which speeds up startup.
45504 To determine whether a file contains such an index, use the command
45505 @kbd{readelf -S filename}: the index is stored in a section named
45506 @code{.gdb_index}. The index file can only be produced on systems
45507 which use ELF binaries and DWARF debug information (i.e., sections
45508 named @code{.debug_*}).
45510 @command{gdb-add-index} uses @value{GDBN} and @command{objdump} found
45511 in the @env{PATH} environment variable. If you want to use different
45512 versions of these programs, you can specify them through the
45513 @env{GDB} and @env{OBJDUMP} environment variables.
45517 the @value{GDBN} manual in node @code{Index Files}
45518 -- shell command @kbd{info -f gdb -n "Index Files"}.
45525 @c man begin SEEALSO gdb-add-index
45527 The full documentation for @value{GDBN} is maintained as a Texinfo manual.
45528 If the @code{info} and @code{gdb} programs and @value{GDBN}'s Texinfo
45529 documentation are properly installed at your site, the command
45535 should give you access to the complete manual.
45537 @cite{Using GDB: A Guide to the GNU Source-Level Debugger},
45538 Richard M. Stallman and Roland H. Pesch, July 1991.
45544 @node GNU Free Documentation License
45545 @appendix GNU Free Documentation License
45548 @node Concept Index
45549 @unnumbered Concept Index
45553 @node Command and Variable Index
45554 @unnumbered Command, Variable, and Function Index
45559 % I think something like @@colophon should be in texinfo. In the
45561 \long\def\colophon{\hbox to0pt{}\vfill
45562 \centerline{The body of this manual is set in}
45563 \centerline{\fontname\tenrm,}
45564 \centerline{with headings in {\bf\fontname\tenbf}}
45565 \centerline{and examples in {\tt\fontname\tentt}.}
45566 \centerline{{\it\fontname\tenit\/},}
45567 \centerline{{\bf\fontname\tenbf}, and}
45568 \centerline{{\sl\fontname\tensl\/}}
45569 \centerline{are used for emphasis.}\vfill}
45571 % Blame: doc@@cygnus.com, 1991.